Sunday, February 08, 2004

A Day at the Bazaar

A small crowd of people has gathered around an illusionist at a bazaar. In the circle of dust before him are three walnut shells, whose positions he is exchanging with wonderful dexterity. Nearly all the onlookers are agreed that one of them -- look how swiftly he moves it! -- contains a pea. As time passes some of the watchers develop ingenious methods for estimating the correct location of the elusive pea. One fellow has set up a laser tracking system to measure the speed at which the walnut shells move and has deduced from the fact of its slightly slower acceleration that the pea lies under shell number 2. Another has contrived to plant a listening device in the illusionist's home and has recorded conversation from wife to child hinting the pea is under shell number 1. Still another has spread microscopic particles on the dust and by subtle variations in their disturbance concluded the pea was under shell number 3. Each of the three watchers confidently indicates the 'certain' location of the pea and the illusionist upturns the particular shell in turn to reveal nothing. But opening the third shell without reclosing the rest proved the illusionist's undoing and the irate crowd is on the point of thrashing him when a voice from the back says: "Stop. What we have is an intelligence failure. Let's hold an investigation to find out how we could have been fooled."

The Guardian thinks problems like this can be avoided if we ask the right questions.

... unwittingly intelligence analysts can be fooled into assembling the case that appears 'most likely', rather than challenging the evidence to find out what is actually true. And according to recently retired intelligence professionals on both sides of the Atlantic, on the issue of Iraq's WMD that is precisely what occurred. ... 'People have also forgotten how to ask the right sort of questions about what they are seeing - going deeper into the material. You cannot depend on what seems obvious.'

Jane Harman, the leading Democrat on the House Permanent Select Committee on Intelligence, essentially says the same thing. In an article entitled "Four Steps to Better Intelligence" in the Washington Post, Harman says the CIA should try

  1. "to relearn forgotten lessons. In 1992 Robert Gates, then director of central intelligence, wrote that analysts should highlight what they don't know and not try to 'make the tough calls.'"
  2. "quickly correct any other WMD estimates that may contain or be tainted by the same deficiencies found in the Iraq analysis. If estimates of Iraq's WMD programs were so far off the mark, we must be concerned that there are systemic deficiencies in intelligence analysis on other WMD programs and activities, such as those in Iran, Libya, North Korea, Syria and Pakistan."
  3. "improve collection of WMD information, including recruiting and deploying more human sources."
  4. "call for a reinvigorated U.N. inspector force and commit additional U.S. resources to this effort. It turns out that U.N. inspectors had better information than anyone else. Presidential leadership on this would go a long way to restore sadly tarnished U.S. credibility."

During the investigation of the shell game fiasco, one street urchin, who had been called as an expert witness observed, "the key moment came when you were able to look under all the shells simultaneously". Asked to elaborate on the point, the urchin said. "The total quantity of information obtained from calculating the angular momentum of the shells, interpreting signals intelligence or engaging in statistical analysis of the microscopic tracers was wholly dwarfed by the act of looking under them, and subsequently interrogating the illusionist. Now everyone knows that the pea was passed to a confederate in the crowd before the game actually started. And that the whole point of the game was to provide a distraction for you worthy gentlemen while I picked your pockets."

In related developments, "the founder of Pakistan's nuclear weapons program, Abdul Qadeer Khan, has signed a detailed confession admitting that, over the past 15 years, he provided Iran, North Korea and Libya with designs and technology to produce the fuel for nuclear weapons." WMD designs found in Libya, whose dictator opened his country for inspection after the invasion of Iraq showed facilities constructed to a Pakistani design, itself derived from European sources, and of components partially manufactured in Malaysia. The feared WMDs were not being hatched in secret by Saddam Hussein so much as developed as an entire export industry by a staunch American ally, Pakistan and a 'moderate' Muslim country, Malaysia.

As he handed back their wallets, the urchin was asked by the professional shell watchers how they could improve their pea-watching methods. His advice to them was to "keep looking under shells. Whatever you do, don't gaze at your navels. Knowledge comes from flipping over shells." After long deliberation, the watchers are now busy learning how to ask the right questions, avoiding group-think and reinvigorating a dilapidated inspectorate. They have also ordered false moustaches and still undisclosed new technology. Asked why they had disregarded the urchin's advice they said, "that is why we are members of the Select Committee of Shell Watchers while he is still a street Arab".

Talleyrand is supposed to have said of the Bourbon kings, “They have learned nothing, and they have forgotten nothing.”