The prisoner seemed to be a Viet Cong officer of some sort, and he was basically just laughing at his American captors. He didn't even bother to make his lies consistent. His story changed once. Then twice. Three times.
That's the way the Army interrogator remembers it, anyway. His job as a US officer was to try to elicit useful information from POWs. But guidelines for his actions were rigid - the most he was allowed to do was shout.
So the interrogator tried patience. He waited. And waited. Then he threw the man a pack of cigarettes and said that if he didn't give up something the Americans would be forced to turn him over to the South Vietnamese police.
The man talked. It turned out he was a province chief.
"The point is, you get better results from being relatively humane to [prisoners], rather than beating on them," says the interrogator today, who asked to remain anonymous due to continuing work with US intelligence.
That may be so - but it is also likely to be only part of the story. Interrogation is a complicated process in which two people knock against each other under tense conditions. According to past and present US training manuals, loosening a prisoner's tongue requires careful planning, methodical implementation - and, at times, the inspiration of an artist.
How much of the abuse of Iraqi detainees was related to interrogations remains unclear, but the controversy has brought this process under a spotlight - and shown, graphically, what can happen when it goes wrong. An Army field manual on intelligence interrogation quotes Gen. Douglas MacArthur: "In no other profession are the penalties for employing untrained personnel so appalling or so irrevocable as in the military."
US military scientists have studied interrogation methods for decades, according to documents from the library of a retired high-ranking officer made available to the Monitor.
Three generations of CIA and military interrogation manuals show how methods have evolved and been refined. After World War II, personnel pored over the testimony of Hanns Joachim Scharff, a genial German interrogator who questioned every downed US fighter pilot and was famous for his use of props meant to put prisoners at ease. During the Korea and Vietnam conflicts, they weighed the effectiveness of less-savory psychological methods characterized in the press as "brainwashing."
Through much of the cold war, US interrogators trained allies in Latin America and elsewhere in the use of humiliation, nakedness, physical discomfort, and other harsh coercive techniques.
Sensory deprivation in dark, sound-proofed cells can deeply affect an interrogee, said the CIA's notorious Vietnam-era KUBARK interrogation manual. "An environment still more subject to control, such as water-tank or iron lung, is even more effective," it stated.
Officially this approach was abandoned following congressional hearings in the mid-1980s. However, even training manuals that predate this change emphasized that the most important quality for an interrogator was not necessarily intimidation. The chief qualification for a questioner is "a genuine insight into the subject's character and motives," concludes the CIA's 1983 Human Resources Exploitation Training Manual.
The first rule of successful interrogation is to set the mood, according to military and CIA manuals. Ideally, that means using a spartan interrogation room. Mirrors, see-through or otherwise, should be positioned so that detainees can't see themselves. If the interrogator has a comfortable chair, and the subject does not, that subtly conveys an impression of power. There should never be a phone in the room. It is both a subliminal connection to the outside world, and an actual source of distraction. And distractions break the spell of an interrogation.
"The effect of someone wandering in because he forgot his pen or wants to invite the interrogator to lunch can be devastating," notes the CIA's KUBARK manual.
The second step is generally an analysis of what type of person the detainee is. Both the KUBARK manual and its 1983 successor, for instance, list presumed categories of prisoner personalities. These include the "orderly-obstinate subject" and the "guilt-ridden subject."
Then comes questioning itself. Interrogators must be alert to any sign that subjects are withholding information, note manuals. They should match demeanor to prisoner psychology - "orderly-obstinate" prisoners, for example, are thought to clam up in response to table-pounding.
Rapid-fire nonsense questions are one way to disorient a prisoner and wear down resistance, noted the CIA in 1983.
The Army's FM 34-52 Intelligence Interrogation manual, still in use today, lists a number of approaches. "Emotional hate," for instance, tries to build on any grievance the prisoner might have, through such statements as, "You owe them no loyalty because of the way they've treated you." The "fear-up" approach tries to exploit preexisting fears, according to the manual, without actually lying about the consequences of noncooperation. The "file and dossier" method builds a paper case against a detainee, often with padding to make it look thicker, to trick the prisoner into believing it would be explosive if released to family members or the public.
None of the techniques outlined in FM 34-52 should be construed as allowing physical or mental torture, or any other form of mental coercion, says the manual.
Yet the abuse of prisoners in Iraq shows that the harsh practices that were supposed to have ended decades ago have continued. New iterations of interrogation techniques have been issued by the Pentagon - many of them echoing the harsher practices listed in the earlier CIA manuals.
The investigations into what happened at Abu Ghraib, and why, are widening to look at the military intelligence side of the equation, as well as looking at suspected abuses and even deaths at military prisons in Afghanistan and Guantánamo.
Meanwhile, the fallout continues. So far, seven members of the military police who participated in the abuses have been charged. On Monday, Brig. Gen. Janis Karpinski, in charge of 16 Iraq prisons when the abuses at Abu Ghraib occurred, was indefinitely suspended, pending results of further investigations.
The Pentagon has also announced that Lt. Gen. Ricardo Sanchez, commander of the forces in Iraq, will leave Baghdad this summer. Gen. George Casey Jr., the Army's second-ranking general, has been tipped as his replacement.
Gen. George Fay's report, examining the military intelligence role in the abuses, is expected within the next 10 days.