Eric Holder: Prosecutorial Discretion and Extrajudicial Deaths Commentary
Eric Holder: Prosecutorial Discretion and Extrajudicial Deaths
Edited by:

JURIST Guest Columnist Tung Yin of the Lewis & Clark Law School argues that Attorney General Eric Holder should have given some explanation for his decision to refrain from prosecuting any suspects after spending years investigating the Central Intelligence Agency’s abuse of interrogation methods that led to the deaths of at least two detainees…

Gul Rahman and Manadel al-Jamadi were among hundreds of detainees captured during armed conflict with US forces in Afghanistan or Iraq, but unlike most of the other captives, they died in captivity. Moreover, the circumstantial evidence suggests that their interrogators caused their deaths. On November 20, 2002, Rahman was stripped naked below the waist and left in a secret Central Intelligence Agency (CIA) prison overnight. The ambient temperature in the cell dipped near freezing. In the morning, Rahman was dead from hypothermia. Al-Jamadi died on November 4, 2003, in the Abu Ghraib prison in Iraq from wounds inflicting during a severe beating.

In 2007, Bush administration US Attorney General Michael Mukasey appointed career prosecutor John Durham to investigate the CIA’s admitted destruction of videotapes of detainee interrogations. Mukasey’s successor, Attorney General Eric Holder, directed Durham to investigate the actual interrogation practices, not just the videotape destruction. However, Holder instructed Durham that interrogations conducted within the guidance of legal memoranda provided by the Office of Legal Counsel (OLC), even if later disavowed, would not be subject to prosecution.

In June 2011, Durham recommended a full criminal review of the circumstances of the Rahman and al-Jamadi deaths. Just over a year later, he recommended closing the cases. Holder recently announced that the US Department of Justice (DOJ) indeed would not be filing federal criminal charges against any CIA agents or other Americans involved in the Rahman or al-Jamadi interrogations. Not surprisingly, critics of the Bush administration’s detention and interrogation policies have expressed disappointment and outrage.

Virtually unfettered prosecutorial discretion has long been a feature of the American justice system. Although there are some legal doctrines (such as selective prosecution, expounded in Oyler v. Boles) that theoretically restrain improper prosecutions, these do not apply to decisions not to prosecute. For example, in September 1971, prisoners at the Attica Correctional Facility in New York rioted, beating several guards and killing one. During the effort to take back control, prison guards, state troopers, and other government agents killed 32 inmates and wounded 400 more. Most of the prisoners were African-American, and most of the guards were white, adding a racial complexion to the perception of excessive violence in the re-capture of the prison. Although New York Governor Nelson Rockefeller appointed a special prosecutor to investigate the actions of the prisoners as well as the guards and troopers, he did not bring any charges against any state officers. The inmates sued for a court order to force local and federal prosecutors to seek indictments.

In Inmates of Attica Correctional Facility v. Rockefeller, the US Court of Appeals for the Second Circuit affirmed the district court’s dismissal of the lawsuit, even though the court acknowledged that the complaint alleged “serious questions … as to the protection of the civil rights and physical security of a definable class of victims of crime and as to the fair administration of the criminal justice system[.]” In addition to the familiar separation of powers concerns raised by the specter of having a court oversee an executive branch agency’s performance of discretionary duties, the Second Circuit pointed out that judges lack the institutional competence to carry out such oversight.

Given the strong similarities between the Rahman and al-Jamadi deaths on the one hand and the Attica prisoner deaths on the other hand, Inmates of Attica suggests that Holder’s decision not to pursue charges is effectively final. But what might explain Holder’s decision, given that the alleged wrongdoing seems indefensible and occurred during the previous administration?

Holder has explained that “[b]ased on the fully developed factual record concerning the two deaths, the Department has declined prosecution because the admissible evidence would not be sufficient to obtain and sustain a conviction beyond a reasonable doubt.”

According to Section 9-27.220 of the US Attorneys’ Manual:

The attorney for the government should commence or recommend Federal prosecution if he/she believes that the person’s conduct constitutes a Federal offense and that the admissible evidence will probably be sufficient to obtain and sustain a conviction, unless, in his/her judgment, prosecution should be declined because:

  1. No substantial Federal interest would be served by prosecution;
  2. The person is subject to effective prosecution in another jurisdiction; or
  3. There exists an adequate non-criminal alternative to prosecution.

Holder’s statement did not reference any of the three bases for declining to prosecute where there is sufficient evidence to sustain a conviction, but it is apparent that none of those would apply here. Wrongdoing by federal agents calls specifically for federal prosecution; by contrast, failing to prosecute sends a message that the US does not abide by its own laws. There is also little prospect of effective prosecution in another jurisdiction. Although the alleged crimes took place in foreign countries, it is far from clear that Iraqi or Afghan officials can exercise jurisdiction, even assuming the political will to do so. Finally, it does not appear that there are adequate non-criminal alternatives to prosecution. It is true that many of the military officers involved in al-Jamadi’s death were disciplined and the US Navy leader was court-martialed — but any CIA agents involved do not appear to have been punished.

Thus, the only way to understand Holder’s decision not to prosecute is to analyze the assessment that “the admissible evidence would not be sufficient to obtain and sustain a conviction beyond a reasonable doubt.” One possibility is that Durham had identified the perpetrators but through means that rendered the evidence vulnerable to suppression. This seems unlikely, although threats to fire public employees if they refuse to answer questions that might incriminate themselves can run afoul of the Self-Incrimination Clause. A second possibility is that Durham was unable to identify the specific wrongdoers, perhaps due to some sort of “code of silence.” This is, of course, a situation that prosecutors routinely face in conspiracy cases and deal with through a variety of means, including “flipping” the lower ranked targets.

Perhaps the most likely explanation is that Holder doubted that a jury would convict any defendants for the conduct in question. This is somewhat hard to understand, given that a military pathology had ruled al-Jamadi’s death a homicide. However, as Jane Mayer has reported in The Dark Side, al-Jamadi was suspected of taking part in or carrying out a number of atrocities, including a 2003 car bombing of the International Red Cross (IRC) headquarters in Iraq that killed more than a dozen aid workers. Of course, even if he were in fact complicit in that bombing and other attacks, it still would not justify killing him during interrogation. But it might have made a jury more willing to acquit any government agents charged with murdering him. Should prosecutors take into account their perceptions of jurors’ reluctance to convict defendants in deciding whether to pursue criminal charges? If prosecutors are correct in their assessment, it would seem futile to expect them to proceed to trial. The time and resources spent on a hopeless case could surely have been used on other matters.

Yet, prosecutors may well be incorrect in their assessment of jurors. One can imagine that if there has been a history of jury acquittals for the same kind of crime — such as hate crimes against African-Americans in the Jim Crow South — prosecutors might be justified in concluding that additional prosecutions would likely have the same result. But the in-custody killings of Rahman and al-Jamadi have no immediately apparent precedents from which to predict jury hostility to prosecution. While one could expect a certain lack of sympathy for Rahman and al-Jamadi, it is also possible that a jury would be disturbed at the torture-like interrogations that led to their deaths.

Without access to the government’s files and classified information, it is difficult to draw any definitive conclusions about the DOJ’s decision to decline charges. But nothing that is known publicly readily supports that decision, especially when one remembers that Durham reviewed only those cases where the interrogations strayed from the heavily criticized OLC “torture” memos. Prosecutors do have discretion to charge or not to charge, of course, and in the typical instance, there would be no public explanation or statement. The high-profile nature and national (as well as international) significance of the Rahman and al-Jamadi deaths, however, called for such an explanation. Unfortunately, Holder’s statement falls short of providing a satisfying explanation.

Tung Yin is a professor of law at Lewis & Clark Law School specializing in national security law and terrorism and the law. He has examined such matters as the jurisdiction of the federal courts to entertain habeas petitions by Guantanamo Bay detainees, the theory of unilateral executive branch war powers and the potential constitutional rights available to alien detainees outside the country.

Suggested citation: Tung Yin, Eric Holder: Prosecutorial Discretion and Extrajudicial Deaths, JURIST – Forum, Sept. 26, 2012,

This article was prepared for publication by Caleb Pittman, head of JURIST’s academic commentary service. Please direct any questions or comments to him at

Opinions expressed in JURIST Commentary are the sole responsibility of the author and do not necessarily reflect the views of JURIST's editors, staff, donors or the University of Pittsburgh.