It was tedious, frustrating work. The days sometimes began as early as 6 am and could stretch until 6 pm. Agents' desks were cluttered with stacks and stacks of case files, some of which had been opened as early as 2003 but remained unresolved more than two years later. Much of the agents' time was spent trying to locate victims, perpetrators and eyewitnesses.

Eventually T-shirts were made for the agents. The front displayed the unit's name and DATF motto: Do What Has To Be Done. The back read Detainee Abuse Task Force 2005 and listed the agents' names along with a dark inside joke about the daunting task before them: An Unknown Subject Assaulted an Unknown Victim, at an Unknown Time and Location. Investigation Continues.

A year earlier, in April 2004, searing photos of detainee abuse at Abu Ghraib were broadcast around the world, shocking the public. Political and military leaders condemned the abuse and promised swift action and accountability. President Bush pledged that the United States would "investigate and prosecute all acts of torture and undertake to prevent other cruel and unusual punishment in all territory under our jurisdiction." Within two years, the Defense Department announced it had opened 842 criminal inquiries or investigations into allegations of detainee abuse.

As part of this response, the military produced thirteen comprehensive reports, the FBI produced two reports, and the CIA produced at least one. Also in 2004, according to a senior Army official with knowledge of the DATF, the Army's top CID officer, Maj. Gen. Donald Ryder, initiated a task force to manage what the official called the "volatility" and "sensitivity" of the detainee abuse issue. The task force was run by CID, and it periodically interfaced with the FBI and other federal agencies, as well as Army public affairs, operations and the Congressional liaison office. As part of this effort, in July 2004, the Army created an on-the-ground investigative team in Iraq, the CID's Detainee Abuse Task Force, which in 2004 consisted of agents from the 78th Military Police Detachment and in 2005 agents from the 48th.

During stretches of its existence, according to a joint investigation by The Nation, The Investigative Fund at The Nation Institute and PBS's Need to Know, the DATF had six full-time agents charged with investigating abuse cases that occurred in and around Victory Base Complex—a huge area of responsibility that included the heaviest concentration of detainees. (Individual agents at thirteen military facilities across Iraq dealt with abuse cases in their localities.) DATF agents were also charged with reviewing older cases that had been reopened. Reports generated by the DATF, according to the senior Army official, were reviewed by a team of senior CID agents at CID headquarters in Fort Belvoir, Virginia.

The five CID agents who were interviewed for this article, four of whom worked on the DATF during 2005, said there was no consensus over what constituted abuse, especially when it came to interrogation techniques. They said the case files they received were often missing key pieces of evidence. They said they faced noncooperation from some military units they were investigating. They said they didn't have competent Arabic translators and were rarely able to track down victims once they'd been released from detention. They said they were overwhelmed by hundreds of abuse cases they'd been ordered to reopen, which one agent speculated was done to avoid responding to Freedom of Information Act (FOIA) requests from the ACLU.

Jon Renaud, a retired Army Warrant Officer who headed the task force as the Special Agent in Charge for the first half of 2005, now says of the DATF, "It didn't accomplish anything—it was a whitewash." Neither he nor his fellow agents could recall a single case they investigated that actually advanced to a court-martial hearing, known as an Article 32.

"These investigations needed to take place," said Renaud, a Bronze Star recipient who retired in 2009 after twenty years in the military. "But they needed to be staffed and resourced with the same level of resources that they gave the Abu Ghraib case." He noted that the Army assigned a general and staff to conduct a comprehensive investigation of Abu Ghraib. "That was a single case," he said, "and we had hundreds of others for six people."

Requests to the Army for information about the origins, mission and track record of the DATF were refused, and a FOIA request to CID was denied with this claim: "No documents of the kind you described could be located. No official 'Detainee Abuse Task Force' was ever established by the USACIDC." After a lengthy appeals process, during which we provided several samples of DATF communications on DATF letterhead, this finding was reaffirmed: CID "never created an official 'Detainee Abuse Task Force,'" the denial letter read. "Individual criminal investigation units may have set up informal, ad hoc task forces while deployed to emphasize detainee abuse investigations. In turn, they may have labeled certain investigations as being subject to a 'Detainee Abuse Task Force.'" But "there was no official organization for such a task force."

The agents' accounts of failed detainee abuse investigations are bolstered by a 2006 study by Human Rights Watch and two other human rights groups, which found that of 330 cases in which US military and civilian personnel allegedly abused or killed detainees after 2001, "only about half appear to have been properly investigated." In many cases, investigators "failed to follow basic investigative techniques, including interviewing victims and witnesses and gathering physical evidence."

Angela Birt, the Operations Officer who oversaw CID's felony investigations across Iraq during 2005, including the DATF, expressed disbelief at the military's response. According to Birt, the DATF did not receive an official unit designator; "there was no heraldry behind it," she said. "But to say it didn't exist in the terms that they said in the letter? Wow, that's really embarrassing for them," said Birt.

"To say, 'You never existed,'" Renaud said, "It's insulting. It's insulting to the agents that worked on it.

"I have to assume they just don't want to release the cases," he went on, "because if anybody actually got ahold of all the cases [and] read over them, they would obviously see huge holes."

* * *

Attorney Susan Burke spent her childhood bouncing around military bases and regards the military as an extended family. She'd long been proud of the Army's tradition of military law. But when the Abu Ghraib photos went public, she knew something had gone terribly wrong. She was determined to find out if such grave abuses were widespread. It was in her quest for accountability that Burke first stumbled on the existence of the DATF.

In the months after Abu Ghraib, Burke, a litigation attorney now based in Washington, DC, hired an Iraqi researcher and an American private investigator, Keith Rohman, to try to locate and interview other victims of detainee abuse, as well as witnesses who could corroborate their accounts. Burke's staff combed through evidence provided by detainees, including prison bracelets containing detainee ID numbers. Burke's team had their clients examined by doctors to verify their injuries and interviewed American soldiers, contractors and linguists who worked at facilities where the detainees had been held, as well as detainees in nearby cells who may have witnessed the abuse.

What they found troubled them greatly. In contrast to the emerging narrative that Abu Ghraib was the act of a few "bad apples," Burke's team discovered evidence of abuse that extended well beyond the walls of that infamous prison.

"I think Americans don't want to realize how widespread it was, how awful it was," said Burke. "They like to think, 'Oh, it was just a few night guards at Abu Ghraib.' What we have learned is some of the other sites were actually worse. Camp Cropper—the physical torture there was just horrific. The mobile interrogation units—again, the torture there, terrible."

Burke's team uncovered detailed information about what she called "very, very serious cases of physical abuse. People who've been electrocuted, people who have been mock executed, people who have been hung by their arms for hours on end."

Eventually, Burke joined forces with Shereef Akeel, a Michigan-based civil rights attorney, and the Center for Constitutional Rights in New York City to file a lawsuit against private military contractors who had abused detainees in Iraq. The case, Saleh v. CACI and L-3, is pending. Separately, Burke sought criminal accountability for the allegations of abuse by military personnel. Reading through the case summaries she compiled, Abu Ghraib begins to feel less and less like an aberration:

A male was arrested [on] January 5, 2004, and released in June of 2004.... He was taken to [Mustansiriya] where he was hung by his hands and feet and beaten with a stick, kept nude for hours, and made to sleep outside.... He was taken to Abu Ghraib and threatened with the rape of his family, denied the ability to pray, forced to watch military personnel having sex, forced to break fast, forced to view pornography, and forced to masturbate. A female sodomized him with an artificial penis.

A female was arrested with her husband on September 4, 2003, in front of her two daughters.... While incarcerated, she was dragged on the ground, and asked to pick up feces with her hands. When she stopped to vomit, she was threatened by American personnel that if she continued, she would be raped. She was held at gunpoint, hooded, denied prayers, touched improperly. Her hands and legs were tied and she was put in the sun for hours. For 15 days, she did not have food, little water, and was forbidden from going to the bathroom.... She was photographed nude.

A man was detained by American forces at the [Samarra] Police Station on January 13, 2004 and was released on July 22, 2004.... He was transported to a facility near Baghdad Airport. During an interrogation there, an African-American interrogator demanded that the detainee confess to supporting the resistance. He put a gun to his head, threatened him with a knife, and tore off his clothes, stripping him naked. The detainee was beaten and later when a doctor came to see him he learned that his breastbone was broken.

In August 2004, Burke met with Captain Christopher Graveline, one of the Abu Ghraib prosecutors, to share her findings. She contacted senators on the Armed Services Committee. And she wrote to military investigators with the Army's CID to offer them access to her clients for their investigations. In October 2004, Burke told military personnel via e-mail, "We are most interested in working with you and others within the military who are acting in good faith to investigate fully and remedy the wrongs that occurred.... As the daughter of an Army Colonel (now retired), I am personally confident that many on active duty are as appalled as we are by the terrible abuses that occurred."

Burke first traded e-mails with Renaud, the head of the DATF, in March 2005. She felt encouraged by his warm response. "We had every reason to believe that her information was extremely credible and valuable," Renaud recalled. The result was a series of much awaited meetings with DATF agents scheduled for June 2005 in Kuwait. But just days before Burke and Rohman left to meet with Renaud, he was detailed to another mission. So on June 3, 2005, Burke and Rohman arrived at the Kuwait City Marriott for two days of meetings with two other DATF agents, Kenneth Dean and Julie Tyler (now Julie Kuykendall).

* * *

What Burke didn't know at the time was that the DATF was in disarray. Even the most basic aspect of its job was ill defined: what exactly constituted abuse? "What is abuse? was the million-dollar question over there because it was never clearly established," said Renaud. "Everybody liked to use the term 'detainee abuse,' but nobody really knows what detainee abuse was. It was never clarified. Were we using the US definition of assault?"

In particular, there was widespread confusion over which "enhanced interrogation" techniques were allowed and which were regarded as abusive. During his fourteen years with CID, Renaud had worked narcotics, fraud and homicide cases as well as protection details for Donald Rumsfeld, Paul Wolfowitz and Gen. George Casey. In 2003 he interrogated suspects connected with Saddam Hussein's regime as part of an Iraqi war crimes investigation by the 3rd Military Police Group. But even he didn't have a clear sense of which interrogation techniques were allowed.

"I was an interrogator for a year in 2003. I ran the Detainee Abuse Task Force, and I can't give you an absolute definition of what detainee abuse is—and none of my bosses can," said Renaud. "Everyone thinks that every interrogator in theater had read a list of enhanced interrogation techniques. Nobody knew what those were."

Several former agents speculated that this confusion over what was a permissible "enhanced interrogation" technique may explain why so very few cases relating to abuse during interrogations ever reached them. Renaud said it is plausible that interrogations complaints never reached them because when military intelligence commanders received a complaint against an interrogator they'd say, "No, I believe this falls within what is an acceptable interrogation technique" and therefore wouldn't file a report. "If you believe your folks didn't do anything wrong...why would you report it?"

The agents said the vast majority of their cases instead involved physical abuse that occurred during detainees' capture and transport. And for injuries sustained at capture, Renaud recalled, "we gave them some leeway."

Former Army interrogator Tony Lagouranis documented the detainee abuse he witnessed in his 2007 book Fear Up Harsh. He agreed that interrogators were unlikely to report on abuse they regarded as permissible. "If people didn't consider what we were doing abuse they had no inclination to report [it]," Lagouranis said. "When people think about torture they're thinking of some kind of horrific cutting off of fingers or burning alive or something like that. But it can be much, much less."

Lagouranis recounted one example. While working in a mobile interrogation unit in Mosul in 2004, he learned that a special operations unit used a combination of flashing lights and loud music, along with ice water and cold temperature to induce hypothermia, on detainees—a regimen nicknamed "the disco" or "the discotheque." Lagouranis's unit eventually copied the technique for its interrogations in Mosul.

Renaud said he heard that the disco was also used at a special ops site in Anbar province in 2003, but he said DATF agents had difficulty getting enough information to investigate it. "It came up in a lot of cases," he said. "I knew it existed. I know folks that had worked there. I knew where it was located. When I'd do a formal request and investigation saying, 'Can you confirm the existence and the location of the disco?' [I was told,] 'We don't have any place like that.' In my role as detainee abuse investigator, I would get zero information."

Renaud was aware that certain units employed harsh interrogation techniques during his first tour of Iraq in 2003, but few such complaints ever reached him. "Not the forced standing, or forced to stay awake, or listen to the loud music," said Renaud. "I don't remember seeing any forced standing. I don't remember any complaints about sleep deprivation. I never saw [complaints related to] dogs used in interrogations."

Former Sgt. Cooper Tieaskie, a member of the DATF during 2005, recalled instances when the DATF would receive records of abuse by military intelligence (MI) units only after they'd been heavily redacted for "operational security reasons."

"Sometimes, it was just like, Here's what we're going to give you—one sentence," Tieaskie said.

"MI became very restrictive about who could participate in or observe their interviews," said Birt, the former head of CID in Iraq, "probably because they were doing something that wouldn't be classically accepted by law enforcement." She explained that charges of abuse would mostly surface if an MI officer or translator who was present, or a member of the medical staff who treated wounds after the fact, chose to report it.

Yet Lagouranis pointed out that medics may have been hesitant to report abuse. "Medics are military people," he said. "If you have a medic who's an E-3 [a private], he's not going to want to make waves and report on anything that his unit is doing."

Kuykendall also explained that any abuse that didn't leave any marks—such as water torture, forced standing, sleep deprivation and other "enhanced interrogation" techniques—was unlikely to be pursued by JAG attorneys. "How are you going to prosecute that?" asked Kuykendall. "That's not to say that they just swept them under a rug. [But] you prosecute what you can prosecute." As a 2006 investigation by the New York Times revealed, the motto of one unit at Camp Nama was "No Blood, No Foul."

Without a medic or unit member documenting abuse, CID would rarely learn of detainee abuse committed by MI forces in the first place. Investigating such units, Birt said, "became increasingly hard because it was hidden from the light."

* * *

Tieaskie and Kuykendall recalled one covert unit based out of Fort Bragg about which there were serious allegations of abuse, but the unit was noncooperative.

"We wouldn't get the interrogators' real full names," said Kuykendall. "We would get their made-up, pseudonym names. Pretty much every case we had with that group of interrogators all went the same way.... We had to close them because there just wasn't enough information to go forward."

According to Birt even noncovert military intelligence would classify the most basic information such as detainee height, weight, age, scars, marks and tattoos. "Something that typically would be in a normal police investigative file," she said, would be labeled classified.

The covert units, said Birt, wouldn't even interface with regular Army CID agents because they had their own dedicated CID agents to "investigate any crimes committed by those soldiers who are black ops." The senior Army official with knowledge of the DATF said these investigators were embedded with special ops units "to keep them on the straight and narrow."

But the former DATF agents said the black ops investigators seldom answered their inquiries about detainee abuse, preferring their own internal review process, which was shrouded in secrecy. The agents associated with the covert unit would "take the initial report...and then it just vanished. I know it exists somewhere, but it does not exist in a manner that we can see or that we can track."

Those agents would never tell the DATF anything about the outcome of their cases—including the cases at Fort Bragg. "We repeatedly asked our command to [look into] it...to go from boss to boss," said Renaud. "Zero support. We could have at least cleared up who was involved in it. But because of the lack of cooperation, and the lack of will, a lot of cases went unresolved."

The former DATF agents said that in some cases ordinary troops also failed to cooperate. Kuykendall vividly recalled one such case, which involved four or five detainees who were severely beaten at Forward Operating Base Iskandariyah, between Baghdad and Karbala, after a soldier was shot in the face during an ambush. According to the complaint, the soldiers took out their anger over the ambush on detainees at the base.

"There was no question about it," she said. "They had black eyes, they had bruises, they had cuts and scrapes—they got beat up. They got the crap kicked out of them." It was the unit's medic who reported the abuse, Kuykendall said, but he wanted to remain anonymous.

"We had to arrange for the commander to get this guy away from his unit so we could interview him and find out what happened," she said. "He talked to us no problem. In that first interview we pretty much find out what happened: who did what, when and how—basically we had just about the whole story."

Then she and her fellow agents started interviewing witnesses for corroborating statements. What should have been "everyday, run-of-the-mill, ordinary witness interviews," she said, "turned into a two- to three-hour interrogation because not a single one of these unit members wanted to talk. Not a one. They were protecting everybody else. They were a real close-knit unit—they were National Guard out of Mississippi...and they weren't gonna rat on their buddy, no matter what."

She pushed on with her investigation, but without corroborating witnesses, it never led to a court-martial.

* * *

Burke and Rohman arrived in Kuwait in June 2005 with a tremendous sense of optimism. They spent hours laying out their findings to Dean and Kuykendall. "We talked about allegations of anal sodomy, with fingers and other objects, allegations of electrocution, sleep deprivation, doused with cold water, rape of women, rape of children and serious visible beatings," said Rohman. He and Burke said they also walked the agents through the various steps they'd taken to corroborate these claims—an account confirmed by extensive notes both Burke and Rohman took during the meetings.

But Dean, now a police captain in Fort Worth, Texas, was dismissive, they claimed. According to Rohman's detailed notes, Dean said, "Yeah, I've heard it all. Steel rods, wooden sticks, brooms, you name it." He dismissed the allegations as rumors, dismissed the possibility that rapes had occurred at Abu Ghraib, dismissed reports that dogs had been used against detainees and repeatedly circled back to what appeared to be his chief concern—that soldiers' careers not be harmed by being subjected to false accusations. (Dean did not respond to repeated calls, e-mails or faxes over a period of months.)

Kuykendall, who described the meeting as "more of a courtesy," did not recall many details but said she and Dean "probably would have had similar reactions. I don't want to say in doubt, but just taking a lot of what they [Burke and Rohman] were saying with a grain of salt." (She did not recall the specific Dean statements quoted by Rohman.)

According to Rohman and Burke's notes, Dean and Kuykendall spoke to the formidable obstacles they faced in pursuing such cases. The agents told Burke that they lacked resources to manage their case loads effectively, that victims and witnesses were often impossible to locate after they had been released from detention, that they faced noncooperation from certain US forces and that they regularly had problems securing physical evidence and medical records. Getting into the field to interview witnesses required securing a military escort, they said, requests that were not always accommodated. And CID could not compel anyone who'd left active duty to cooperate.

In a recent interview, Kuykendall, now studying to become an art teacher in North Carolina, confirmed what she'd told Burke in 2005: that DATF cases were often closed without charges when investigators lacked physical and medical evidence.

Kuykendall also said that DATF would often close cases when agents weren't able to get witnesses and corroboration—including the testimony of abused detainees. "In all of my detainee abuse cases there was only one, maybe two, where I actually got to interview the subjects," she said. Typically, she said she had to depend on transcripts of old interviews conducted at the time the detainee filed the complaint.

Kuykendall said that she and her team rarely pursued alleged victims once they had been released. "There's no tracking people in that country. It's an impossibility," she said.

Burke said she understood the agents' frustration, but thought they had vastly overstated the difficulty of locating Iraqi victims. "With the Internet, e-mail and cellphones, it's not as if these people are impossible to find," she said. Anyway, she added, the agents didn't have to go looking for her clients—she'd offered to deliver them to the DATF, at a safe location of their choice.

The meetings, Rohman recalled, "became increasingly difficult. They strongly discouraged us from providing to them names of other cases to investigate." According to Rohman's notes, Dean finally said that if Burke insisted on sending the cases over, she should wait until December 15, when his reserve duty was up. Dean's remark "would've been a joke," Kuykendall said. "Dean's not that kind of cop." But she does not recall that she or Dean followed up on any of Burke's leads. "The last thing we wanted was a bunch more cases," she said.

"I thought that was the perfect scenario," Renaud said. "We had a lawyer who wanted to come forward and make all these people available in a safe location. Had I handled it, I would have met with every one of these people she represented."

Hearing details of the meeting recently, Birt had stronger words: "We've seriously screwed the pooch here if we have [hundreds of] serious detainee allegations that never saw the light of day. They still deserve to—I don't care if it's ten weeks, ten months, ten years after the offense."

Throughout 2005, Burke said, she repeatedly offered the military access to her clients, at her expense, in stable Middle Eastern countries like Jordan, Kuwait or Turkey. "We weren't saying, 'Go out in the streets of Baghdad and do it,'" said Burke. "We were going to bring the victims to do the interviews in safety."

Her offer was never accepted. Lawyers from the ACLU and Human Rights First, who are representing detainees in similar cases, say that military investigators have never sought out their clients for testimony either—though they did not actively approach the military as Burke did.

At the time of her meeting with the DATF agents, Burke and her team had interviewed more than 140 detainees. Today, Burke's roster of clients has grown to 337 former detainees who have alleged that they were tortured or abused by private contractors or military personnel while in military custody from 2003 to 2005. According to Burke, more than five years after she began writing to military investigators, neither US military investigators nor the Justice Department has interviewed a single one of her clients. "We remain more than willing to put any government official that's investigating torture in touch with victims," she said.

"I don't think it's too late," Burke added. "We ran prisons in a foreign nation, and we permitted torture to go on in those prisons. We have an obligation to identify and prosecute the perpetrators."

After the Kuwait meetings with CID agents, Burke said she provided the military and the Justice Department with documentation of several of their worst cases, including the man hung nude by his hands and feet in Mustansiriya and the woman ordered to pick up feces at Abu Ghraib. No one contacted her.

"I just could not believe it," said Burke. "This is the Detainee Abuse Task Force."

Several DATF agents expressed frustration that their efforts to focus on serious cases were often stymied by CID headquarters. Staff at Fort Belvoir, who were interfacing with Army leadership, would often reopen seemingly dead end cases, overwhelming their case loads.

"We'd close it and bam!—they'd kick it back to us," said Renaud. "So, now you have an open case that you have to do something with that there's nothing to do with."

Renaud explained that his superiors at Fort Belvoir sent him weekly e-mails containing an itemized list of cases they were ordering reopened. He also separately received a list of cases about which the ACLU had filed FOIA requests. And he began to notice a correlation.

"I challenged folks on this. I said, 'Hey, are we reopening these cases because we're going to work them? Or are we reopening them to play hide the ball because we don't want to release them?'"

"We did discuss the potential that they were just sending these back because as long as they're open, they're not subject to FOIA," said Birt. "The rule with [the] Crimes Records Center is: if a case is open, they will not honor a FOIA request because it might jeopardize open and valid investigative pursuits."

"I thought it was borderline illegal what they were doing," Renaud said. "I expressed that, and they assured me it wasn't and...that one had nothing to do with the other."

The senior Army official said, "I'm not aware of that—that would be against our policy." In his experience, cases were reopened because of small reporting lapses or new leads.

Kuykendall too noticed the correlation between the two lists but had a different take on why cases were reopened.

"Would we maybe have opened cases just to rewrite the final reports to make it more clear as to what happened and what we couldn't do and why we had to close it for the ACLU lawsuits? Yes. I'm not surprised if we did that."

Alexander Abdo, an attorney with the ACLU's National Security Project, said, "If it is true that the government deliberately reopened investigations for no other purpose than to avoid transparency under the Freedom of Information Act, that conduct is nothing short of outrageous, and it violated FOIA and the government's obligations of transparency."

* * *

As overwhelmed as they were by their case load, several agents suspected that a vast number of abuse cases never even reached them at all. According to Birt, after detainees were captured they were typically sent to larger detention facilities where medical staff examined and documented their injuries. Sometimes, however, detainees were released from a base without ever being sent to a facility with medical personnel. Birt said that among such detainees, there could have been "a lot" of abuse that was never reported to military investigators.

Sometimes ordinary troops who witnessed abuse faced serious challenges trying to report it. Several military personnel and human rights researchers interviewed for this article said the tone had been set after Abu Ghraib.

Sgt. Joseph Darby was the soldier who provided the Abu Ghraib photos to military investigators. At the time, Secretary of Defense Rumsfeld praised "his courage and his values," but in so doing made his identity public. During his remaining nights in Iraq, Darby slept with a pistol under his pillow, fearing that he'd be attacked. Back home, Darby's house was vandalized so badly that he had to leave his hometown and seek out the military equivalent of witness protection. His experience had a chilling effect.

"I think it all stemmed from that case," said Renaud. "I don't care what you say, these guys are tight—they're fighting a war together. If they give information on their buddy, there is legitimate fear that there could be some retaliation. Anybody who says there isn't is kidding themselves."

"Most troops don't report abuse because they think it's a waste of time—and they're right," said John Sifton, a former senior researcher on terrorism and counterterrorism at Human Rights Watch. "It's a waste of time to risk angering your fellow troops and your commanding officer."

In 2008 the AP investigated the way the military had handled whistleblowers in cases of alleged misconduct or wrongdoing and found that "the Pentagon inspector general, the internal watchdog for the Defense Department, hardly ever sides with service members who complain that they were punished for reporting wrongdoing." The inspector general's office, the AP found, stood by the military and came down against whistleblowers more than 90 percent of the time.

"You know that most likely your commanding officer is not going to listen to you, is not going to take any action to correct what is going on," said Sifton. "It was perfectly reasonable to think that by reporting abuse you were putting yourself at risk. Even if you weren't putting yourself physically at risk, you were putting yourself at professional risk."

Asked if troops reporting abuse felt discouraged, ignored or even threatened, Kuykendall said, "I have no doubt about that happening at all. These commanders are not going to want to have CID all up in their unit asking about detainee abuse."

Sometimes troops who went through the proper channels to report abuse found that their complaints went nowhere. Specialist Stephen Lewis, a friend of Tony Lagouranis, the former interrogator, served in Iraq at the same time and was equally frustrated by the military's follow-through on his reports. Lewis, also a former interrogator, detailed various brutal cases of prisoner abuse he witnessed, including "detainees being beaten, sodomized with a squeegee handle, locked in confined spaces like shipping containers in the heat, mock executions, degrading treatment while being naked, degrading treatment from females—playing on them being Muslim and so forth." He said that he filed at least a dozen reports—but like Lagouranis, who had also filed several reports to CID, he got little reaction.

"CID didn't respond at all, really," he said. "They talked to me twice, I think, and never really followed up."

The former DATF agents also pointed to the problem of internal unit investigations, known as 15-6s, which rarely reached their desks. A 15-6 is a nonjudicial process whereby a superior officer assigns an officer or civilian to investigate misconduct in his or her unit. A 15-6 does not have to be referred to CID unless there is a finding of criminality.

Former DATF agents pointed out that it was up to a commander's discretion whether to initiate a 15-6; if the commander thought the allegations were invalid or didn't warrant follow-up, the inquiry stopped there. Most commanding officers, in their experience, genuinely wanted to root out abuse and misconduct in their units, but for some officers, they said, protecting their troops may have been a higher priority.

Even with a committed commanding officer, the subordinate officers tasked to investigate 15-6 abuse cases typically lacked any experience. They aren't "proper investigators," said Tieaskie. "They don't really know what they're doing."

"When you're talking about detainee abuse cases, highly publicized in the States, possibly resulting in death and/or confinement for somebody...to do a 15-6 and have a noninvestigative person do it is extremely irresponsible," Renaud said. "You're investigating your brother in arms. So, I could be a platoon leader investigating my platoon sergeant, or I could be the executive officer investigating the platoon I was just in charge of six months ago.... There's very little unbiased investigation."

CID agents said that as a result of these dynamics and the lack of trained 15-6 investigators, there were likely many 15-6 cases that were never referred to CID.

"I have no doubt that there are probably hundreds, if not thousands of unit-level investigations that were conducted, founded or unfounded, that nobody is aware of...that have never seen the light of day, that never got to CID," Renaud said.

"I don't think that's without merit," said Birt, when asked about Renaud's estimate of the number of unreported cases of abuse. "I think that's a reasonable conclusion."

* * *

In 2009 Barack Obama, who had campaigned on promises to reverse many of the Bush administration's worst detainee practices, made a decision to block the release of additional photographs requested by the ACLU that may have contained fresh images of detainee abuse. "Individuals who violated standards of behavior in these photos have been investigated and held accountable," Obama said. "Nothing has been concealed to absolve perpetrators of crimes."

Despite such assertions, there has still been little serious accountability for US abuse and torture. Punishment for detainee abuse has been doled out to a handful of soldiers. Few officers have seen the inside of a military courtroom, much less a lengthy sentence. Not one political official has been held responsible.

In 2006 the UN Committee Against Torture found accountability for US military abuses inadequate and issued a statement urging the United States to eradicate torture and to "promptly and thoroughly investigate such acts, prosecute those responsible for such acts and ensure they are appropriately punished." In response, the State Department announced that the military had taken action against 250 service personnel, producing 103 courts-martial and eighty-nine convictions. US officials have often invoked such figures to show how vigilantly they have addressed these cases.

"That's the thing that was so galling," said Burke. The Bush administration claimed that the military was doing a good job by citing "the sheer number of cases that they had investigated and closed." But closing a case, she said, may only mean "it has run into a dead end."

"By the Numbers," the 2006 report by the three human rights groups, found that "of the hundreds of personnel implicated in detainee abuse, only ten people have been sentenced to a year or more in prison." In the course of updating another 2006 report, "Command's Responsibility," Human Rights First found that at least 184 detainees have died while in the custody of US forces since the launch of the "war on terror." "No high-level military officials have faced responsibility for the death of a detainee," said Daphne Eviatar, an HRF senior associate. "No CIA personnel have ever been charged with wrongdoing in the death of any detainee, despite clear evidence in the military's own investigations of CIA involvement in several deaths [and] the longest sentence of any US government personnel in any torture-related death continues to stand at five months."

These failings have not been lost on the international community. In November, during the UN's Universal Periodic Review process, several governments criticized the United States for "the persistent impunity" of officials involved in torture policies and personnel who carried out detainee abuse. Brazil, Norway and Russia have called for more thorough investigations and for the perpetrators to be brought to justice.

Also in November, State Department legal adviser Harold Koh, appearing before the UN, affirmed that the United States was abiding by anti-torture treaties, investigating torture allegations and prosecuting perpetrators. Koh assured the assembled nations that "to our knowledge, all credible allegations of detainee abuse by United States forces have been thoroughly investigated and appropriate corrective action has been taken."

None of the DATF agents interviewed for this article could recall a single case that advanced to a court-martial hearing, known as an Article 32. "As far as I remember, not a single one," said Kuykendall, who served a full year on the DATF. "I would've remembered if one of our cases went to an Article 32." The senior Army official could not recall such an instance either.

"There's no way that you could have hundreds of cases opened, investigated and closed and not have some type of adjudication," said Renaud. "There had to be some wrongdoing. Somebody had to have looked at some of these investigations and found fault somewhere. But nobody was ever tried for ours."

Of the few CID cases that were prosecuted, most of those convicted avoided severe punishment. In one of many conversations, Birt referred to the case of Army Col. Allen West, who, in 2003 in Taji, used mock execution to interrogate an Iraqi police officer accused of insurgent attacks on US forces. The military released the police officer, Yehiya Kadoori Hamoodi, forty-five days after his arrest, realizing that he didn't possess any intelligence about insurgent attacks. According to press reports, Hamoodi remains traumatized from the experience and panics when he sees US troops on Iraqi streets. West's commanders charged him with aggravated assault for threatening a prisoner at gunpoint. West accepted an Article 15 nonjudicial punishment; his sentence amounted to a $5,000 fine and a demotion. West submitted his resignation, but seven years later was elected to represent Florida's 22nd Congressional district on the Republican ticket.

Birt, who retired as a chief warrant officer in 2007, was most deeply troubled by the lack of accountability for cases involving detainee deaths. She was the primary investigator on a case that has stayed with her, involving incidents in Afghanistan in 2002 in which US military police working in the Bagram detention facility beat two Afghan detainees until they died. The victims were Mullah Habibullah, the brother of a Taliban commander, and a 22-year-old taxi driver simply named Dilawar, who was later found to be innocent of any insurgent activity.

"I'd never seen just regular old soldiers chain a guy to a ceiling and beat him until his legs look like he was run over by a bus," said Birt, referring to Dilawar. She and other CID agents thoroughly investigated the allegations, producing a case file thousands of pages long, and the MPs and interrogators involved were court-martialed. The final outcome of the case "had a profound effect on my trust of the justice system," Birt recalled.

"We had eighteen people who confessed to complicity in two homicides, and no one served over six months in jail," she said. "People were convicted and given a slap on the wrist." Of the twenty-seven Army personnel charged in the Afghans' deaths and related abuses, only four troops were sentenced to jail time.

Birt had wanted to be a police officer ever since she was 8 and passionately believed in accountability through sober law enforcement. Yet the minimal punishment for the Bagram beating deaths rattled her.

"The outcome of that investigation and the lack of justice was my primary reason for leaving the military," she said. "We tell people all the time that we're going to be exempt from the Geneva and Hague war crimes tribunal because we're going to police our own. But we didn't police our own."