At no point in the last two-and-a-half years has the spectre of Iraq been more sharply evident than in the events surrounding Secretary of State John Kerry's statement of the United States' intelligence and national-security case for limited military intervention in Syria following the apparent use of chemical weapons in attacks on Damascus on 21 August. Indeed, Secretary Kerry himself called it to centre stage with his promise that 'We will not repeat that moment' - a reference to what is now considered to have been a significant intelligence failure as well as the alleged politicisation of intelligence by the George W Bush administration by fitting the facts to the argument.
Yet fears that history is repeating itself were raised by the reported comments of an anonymous senior US official that the intelligence case is 'not a slam dunk' - evoking the phrase used by then-CIA chief George Tenet to describe the intelligence case in the run-up to the Iraq War - and the acknowledgement in a report by the Office of the Director for National Intelligence that the evidence against Syria is 'thick with caveats'.
Such alarm at the lack of absolute certainty around the intelligence case ascribing blame for the attack to Assad's regime shows a fundamental misunderstanding of the nature of intelligence - and the inherent difficulty in using it to make a public case for any policy, especially one that is a matter of life and death.
Intelligence cannot be taken as 'fact' and it should never be described as a 'slam dunk'. It is simply an estimate or an assessment of often fragmentary information by analysts using deduction, past experience and the comparison of sources and information to judge its likely veracity. Rarely is intelligence as clear-cut as the images of Soviet missiles on Cuba presented to the UN by Adlai Stevenson in 1962. More usually, it sits on a scale of (un)certainty and its substance is rooted in the caveats used to qualify it. Together these pieces create a picture that is more or less convincing, depending on the assessed quality and quantity of the information collected.
And that is precisely why it can be a difficult line to tread when using intelligence to publicly advocate a particular policy.
All arguments structure the information available to create as 'black and white' a picture as possible in order to persuade. Language must be assertive and, where possible, devoid of ambiguity. And examples must be forceful and dynamic. But intelligence is not black and white and, stripped of its caveats - even if only to protect the sources and methods used in its collection - it can appear more concrete than it actually is.
For example, in his (relatively brief) statement, Secretary Kerry used the words 'know' or 'knows' no less than thirty-five times. Indeed, 'The United States Government now knows that at least 1,429 Syrians were killed in this attack, including at least 426 children' (emphasis added) - startlingly precise figures which, in themselves, are extremely compelling. But how certain is the intelligence community (IC) of the veracity of this information and what were its sources for this? Without such information, it is difficult for the public properly to assess the credibility of the government's claims and therefore of its case.
The importance of maintaining the integrity of intelligence in such circumstances can best be seen looking back ten years to the US presentation of evidence at the UN and the UK's earlier publication of an intelligence dossier - subsequently dubbed the 'dodgy dossier' - in the run-up to the 2003 Iraq War.
In July 2004, the US Senate Select Committee on Intelligence reported that the key judgements regarding the extent of Saddam Hussein's WMD programme contained in the October 2002 National Intelligence Estimate (NIE) 'either overstated, or were not supported by, the underlying intelligence reporting'. Yet it was on the NIE that then-Secretary of State Colin Powell relied when presenting the US case for intervention in Iraq to the UN in February 2003, leading him to state categorically that, 'We know that Saddam Hussein is determined to keep his weapons of mass destruction, is determined to make more'. In reality, very few of the individual pieces of intelligence material cited - such as the 'evidence' that Iraq had tried to buy yellowcake from Niger - merited the strength lent by Powell's assertive language in putting the United States' argument.
Meanwhile, the UK's parliamentary Intelligence and Security Committee also subsequently deemed the UK government too assertive in its statement of the intelligence case for intervention; for example, in its inclusion of the now notorious 45-minute claim in the Prime Minister's Foreword and Executive Summary of the September 2002 dossier - without explaining that the intelligence suggested this applied only to battlefield, not strategic, weapons. Furthermore, the Butler Committee found that although the dossier generally reflected the Joint Intelligence Committee's assessments, its failure to repeat the warnings about the limited nature of the intelligence available was 'a serious weakness'.
Of course, the situation regarding Syria today is significantly different to that of Iraq a decade ago. And given the range of open-source information available - from YouTube videos, to journalistic and NGO reports, and accounts by both international and Syrian medical staff on the ground - very few would argue that the attacks on the residents of the Damascus suburbs did not involve some sort of chemical weapon. Indeed, the Obama administration now says that it has physical evidence that sarin was used in the August attacks, based on samples taken from victims - something the UN is expected to confirm within days.
But what is subject to debate is who was behind the attacks and therefore against whom any punitive action should be directed.
In making the case for the culpability of the Assad regime, Secretary Kerry showed that the Obama administration is clearly very mindful of the lessons of the past. He moved to reassure the American people that the IC has not made the same mistakes as a decade ago, stating that it has 'carefully reviewed and re-reviewed information regarding this attack'. Alongside his assertive statements of what 'we know' and the 'facts' and 'evidence', the choice of language in discussing the overall estimate generally hinted at the somewhat subjective nature of intelligence conclusions, with Kerry describing it as a 'high confidence' evaluation (thereby leaving some room for doubt).
Yet despite this more circumspect approach to the presentation of the case, there remains an unavoidable tension between the desire to prove culpability before taking military action and the fact that intelligence can only point towards a conclusion, rather than 'proving' it.
As such, the administration is again asking the American people to place their trust in those making the judgement call on the quality of the intelligence available and the picture it combines to create. Given the burdens borne in the decade since the invasion of Iraq, this will require a huge leap of faith.
This may be one reason why President Obama has made the somewhat surprising decision to seek the approval of the people's representatives in Congress, having made the underlying intelligence available to them in full. In this way, perhaps it will be possible to restore public trust in the executive branch of the US government when it comes to the appropriate use of intelligence - and finally to lay that particular ghost of Iraq to rest.