available intelligence, even if uncorroborated. But we have noted that the report turned out to be wrong on several counts:on the total stockpile of chemical agent, on t he availability of particular types of agent and on the ballistic missile systems available for their delivery.
Estimates of the size of the Iraqi chemical agent stockpile were revised radically downwards in the immediate pre-war period, from the November 1990 estimate described above to an assessed range of 6,000–10,000 tonnes. This was drawn up to provide military commanders with an indication of the possible scale of Iraq’s use of chemical weapons, and of how long such use could be sustained. We questioned the derivation of the figures. We were told that the calculation started from an estimate of Iraq’s chemical agent production capacity, derived from past intelligence about production at individual plants, pieced together to provide a figure for the combined capacity of Iraq’s production plants of 3,000–5,000 tonnes per annum1. Estimates of the possible size of the stockpile were derived by assuming two years’ production at full capacity over the period from the end of the Iran/Iraq war until the start of the first Gulf war. Such estimates assumed that no chemical agent stocks had been left over from the Iran/Iraq war. The sizeable range given is a reflection of the uncertainty inherent in this estimate, and especially in the scale of operation of the production plants. Less agent would have been available had the plants been operating at less than full capacity; more would have been available had some stocks remained after the Iran/Iraq war.
We understand why the JIC chose that method of calculation, given the limited evidence available in the immediate post-war period of residual Iraqi chemical weapons capabilities. We also noted that the assumptions behind the estimate were clearly spelt out in the JIC assessment. But we have also concluded that one consequence was to leave the intelligence community with an estimate for the size of the Iraqi chemical agent stockpile which was over-cautious, and at its upper end worst case. We have also noted that, after May 1991, JIC assessments did not spell out that the figures inside them were calculated on the same worst case basis. There will inevitably have been a risk that that estimate, shorn of its assumptions, may have become the ‘prevailing wisdom’, with subsequent Iraqi declarations being tested against it for truthfulness, especially in circumstances where intelligence was sparse. If so, that process would have tended to lead to deductions by analysts and policy-makers that there were shortfalls in Iraqi declarations. Furthermore, suspicions here will have been exacerbated by Iraqi prevarication, concealment and deception in the early- and mid-1990s, reinforcing any suspicions that Iraq had substantial stocks to hide.
We have also noted, however, that by 1994/95 the JIC was becoming more sanguine about the size of the Iraqi chemical agent stockpile and indeed of the value to Iraq of retaining a stockpile at all. A JIC assessment in September 1994 noted that:
. . . we do not believe the full extent of the CW programme has yet been revealed
but also that:
Iraq later declared to UNSCOM that, during the entire period of its chemical warfare programme, it produced 3,859 tonnes of chemical agent. Of this quantity, it weaponised 3,315 tonnes, of which about 80% was used during the Iran/ Iraq war. UNSCOM was unable to verify this information fully.