A U.S. Army general dispatched by senior Pentagon officials to bolster the collection of intelligence from prisoners in Iraq last fall inspired and promoted the use of guard dogs there to frighten the Iraqis, according to sworn testimony by the top U.S. intelligence officer at the Abu Ghraib prison.
According to the officer, Col. Thomas Pappas, the idea came from Maj. Gen. Geoffrey D. Miller, who at the time commanded the U.S. militarydetention center at Guantanamo Bay, Cuba, and was implemented under a policy approved by Lt. Gen. Ricardo S. Sanchez, the top U.S. military official in Iraq.
"It was a technique I had personally discussed with General Miller, when he was here" visiting the prison, testified Pappas, head of the 205th Military Intelligence Brigade and the officer placed in charge of the cellblocks at Abu Ghraib prison where abuses occurred in the wake of Miller's visit to Baghdad between Aug. 30 and Sept. 9, 2003.
"He said that they used military working dogs at Gitmo [the nickname for Guantanamo Bay], and that they were effective in setting the atmosphere for which, you know, you could get information" from the prisoners, Pappas told the Army investigator, Maj. Gen. Antonio M. Taguba, according to a transcript provided to The Washington Post.
Pappas, who was under pressure from Taguba to justify the legality and appropriateness of using guard dogs to frighten detainees, said at two separate points in the Feb. 9 interview that Miller gave him the idea. He also said Miller had indicated the use of the dogs "with or without a muzzle" was "okay" in booths where prisoners were taken for interrogation.
But Miller, whom the Bush administration appointed as the new head of Abu Ghraib this month, denied through a spokesman that the conversation took place.
"Miller never had a conversation with Colonel Pappas regarding the use of military dogs for interrogation purposes in Iraq. Further, military dogs were never used in interrogations at Guantanamo," said Brig. Gen. Mark Kimmitt, spokesman for U.S. forces in Iraq.
Pappas's statements nonetheless provide the fullest public account to date of how he viewed the interrogation mission at Abu Ghraib and Miller's impact on operations there. Pappas said, among other things, that interrogation plans involving the use of dogs, shackling, "making detainees strip down," or similar aggressive measures followed Sanchez's policy, but were often approved by Sanchez's deputy, Maj. Gen. Walter Wojdakowski, or by Pappas himself.
The claims and counterclaims between Pappas and Miller concern one of the most notorious aspects of U.S. actions at Abu Ghraib, as revealed by Taguba's March 9 report and by pictures taken by military personnel that became public late last month. The pictures show unmuzzled dogs being used to intimidate Abu Ghraib detainees, sometimes while the prisoners are cowering, naked, against a wall.
Taguba, in a rare classified passage within his generally unclassified report, listed "using military working dogs (without muzzles) to intimidate and frighten detainees" as one of 13 examples of "sadistic, blatant, and wanton criminal abuses" inflicted by U.S. military personnel at Abu Ghraib.
Experts on the laws of war have charged that using dogs to coerce prisoners into providing information, as was done at Abu Ghraib, constitutes a violation of the Geneva Conventions that protect civilians under the control of an occupying power, such as the Iraqi detainees.
"Threatening a prisoner with a ferocious guard dog is no different as a matter of law from pointing a gun at a prisoner's head and ordering him to talk," said James Ross, senior legal adviser at Human Rights Watch. "That's a violation of the Geneva Conventions."
Article 31 of the Fourth Geneva Convention bars use of coercion against protected persons, and Common Article Three bars any "humiliating and degrading treatment," Ross said. Experts do not consider the presence in a prison of threatening dogs, by itself, to constitute torture, but a 1999 United Nations-approved manual lists the "arranging of conditions for attacks by animals such as dogs" as a "torture method."
But Pappas, who was charged with overseeing interrogations at Abu Ghraib involving those suspected of posing or knowing about threats to U.S. forces in Iraq, told Taguba that "I did not personally look at that [use of dogs] with regard to the Geneva Convention," according to the transcript.
Pappas also said he did not have "a program" to inform his civilian employees, including a translator and an interrogator, of what the Geneva Conventions stated, and said he was unaware if anyone else did. He said he did not believe using force to coerce, intimidate or cause fear violated the conventions.
Brig. Gen. Janis L. Karpinski, who commanded the prison guards at Abu Ghraib's cellblocks 1A and 1B until Nov. 19, when Pappas assumed control, said in an interview that Navy, Army and Air Force dog teams were used there for security purposes. But she said military intelligence officers "were responsible for assigning those dogs and where they would go."
Using dogs to intimidate or attack detainees was very much against regulations, Karpinski said. "You cannot use the dogs in that fashion, to attack or be aggressive with a detainee. . . . Why were there guys so willing to take these orders? And who was giving the orders? The military intelligence people were in charge of them."
Taguba never interviewed Miller or any officer above Karpinski's rank for his report. Nor did he conduct a detailed probe of the actions of military intelligence officials. But he said he suspected that Pappas and several of his colleagues were "either directly or indirectly responsible for the abuses at Abu Ghraib."
In a Feb. 11 written statement accompanying the transcript, Pappas shifted the responsibility elsewhere. He said "policies and procedures established by the [Abu Ghraib] Joint Interrogation and Debriefing Center relative to detainee operations were enacted as a specific result of a visit" by Miller, who in turn has acknowledged being dispatched to Baghdad by Undersecretary of Defense Stephen A. Cambone, after a conversation with Secretary of Defense Donald H. Rumsfeld.
Cambone told lawmakers recently that he wanted Miller to go because he had done a good job organizing the detention center at Guantanamo Bay, and wanted Miller to help improve intelligence-gathering in Iraq.
Some senators, however, have noted that the Bush administration considers Guantanamo detainees exempt from the protections of the Geneva Conventions, and wondered if Miller brought the same aggressive interrogation ideas with him to Iraq, where the conventions apply.
When asked at a May 19 Senate hearing if he and his colleagues had "briefed" military officers in Iraq about specific Guantanamo interrogation techniques that did not comply with the Geneva Conventions, Miller said no.
He said he brought "our SOPs [standard operating procedures] that we had developed for humane detention, interrogation, and intelligence fusion" to Iraq for use as a "starting point." He added that it was up to the officers in Iraq to decide which were applicable and what modifications to make.
But Pappas said the result of Miller's visit was that "the interrogators and analysts developed a set of rules to guide interrogations" and assigned specific military police soldiers to help interrogators -- an approach Miller had honed in Guantanamo.
After calling the use of dogs Miller's idea, Pappas explained that "in the execution of interrogation, and the interrogation business in general, we are trying to get info from these people. We have to act in an environment not to permanently damage them, or psychologically abuse them, but we have to assert control and get detainees into a position where they're willing to talk to us."
Pappas added that it "would never be my intent that the dog be allowed to bite or in any way touch a detainee or anybody else." He said he recalled speaking to one dog handler and telling him "they could be used in interrogations" anytime according to terms spelled out in a Sept. 14, 2003, memo signed by Sanchez.
That memo included the use of dogs among techniques that did not require special approval. The policy was changed on Oct. 12 to require Sanchez's approval on a case-by-case basis for certain techniques, including having "military working dogs" present during interrogations.
That memo also demanded -- in what Taguba referred to during the interview as its "fine print" -- that detainees be treated humanely and in accordance with the Geneva Conventions.
But Pappas told Taguba that "there would be no way for us to actually monitor whether that happened. We had no formal system in place to do that -- no formal procedure" to check how interrogations were conducted. Moreover, he expressed frustration with a rule that the dogs be muzzled. "It's not very intimidating if they are muzzled," Pappas said. He added that he requested an exemption from the rule at one point, and was turned down.
In the interview transcript, Taguba's disdain for using dogs is clear. He asked Pappas if he knew that after a prison riot on Nov. 24, 2003, five dogs were "called in to either intimidate or cause fear or stress" on a detainee. Pappas said no, and acknowledged under questioning that such an action was inappropriate.
Taguba also asked if he believed the use of dogs is consistent with the Army's field manual. Pappas replied that he could not recall, but reiterated that Miller instigated the idea. The Army field manual bars the "exposure to unpleasant and inhumane treatment of any kind."
At least four photographs obtained by The Washington Post -- each apparently taken in late October or November -- show fearful prisoners near unmuzzled dogs.
One MP charged with abuses, Spec. Sabrina D. Harman, recalled for Army investigators an episode "when two dogs were brought into [cellblock] 1A to scare an inmate. He was naked against the wall, when they let the dogs corner him. They pulled them back enough, and the prisoner ran . . . straight across the floor. . . . The prisoner was cornered and the dog bit his leg. A couple seconds later, he started to move again, and the dog bit his other leg."
Staff writer Josh White contributed to this report.