Discussions about torture often start with this hypothetical: Imagine that there is a terrorist in the middle of Manhattan who has planted a nuclear bomb set to go off within hours. You capture him and are faced with a moral dilemma. Do you torture him to get the information that will allow you to defuse the bomb, thereby saving the lives of millions of people? Or do you stand on principle and sacrifice multitudes?
The purpose of this piece is to shed some light on the way the intelligence community operates, to describe how legal rules shape some of its most sensitive work, and to offer a perspective on the way the Central Intelligence Agency (CIA or Agency) fits into the debate about interrogation and torture. The debate is not about, and indeed cannot be about, whether our government should conduct torture. The answer to that question is and must be, by law and standards of human decency, no. As recently as March 2005, CIA Director Porter Goss reiterated the Agency’s position that it is bound by the laws banning torture and that the Agency adheres to those laws. But at a level deeper than the denials and the blanket statements, there is a difficulty that cannot be avoided. That difficulty lies not in the abstract form of the question, but in the real, on-the-ground scenarios that develop where interrogations are taking place. What can an interrogator do? When can she use deception, discomfort, fear, fatigue, punishment, physical contact, and similar tactics?
Lawyers often represent clients in criminal cases when the odds are long or a catastrophe likely. The facts might be harmful, the evidence overwhelming, or the law clearly on the side of the prosecution. Still, we do the best we can. But what if the system is rigged? What if the system has the trappings of a fair fight but is, in fact, skewed to one side and, by design, the lawyer cannot fully defend the client? What if the lawyer can only lend legitimacy to a process that at its core is biased, slanted in favor of the other side, or fundamentally unfair? Indeed, what if the system is rigged so as to prevent the lawyer from zealously representing the client, or if it compromises the lawyer’s undivided loyalty to the client? Should lawyers refuse to participate in such systems, or should they – should we – still do the best we can?
In the fall of 2001, the Bush administration was looking for a place to imprison and interrogate alleged al Qaeda members away from the prying eyes of other countries and insulated from the supervision of United States courts. The Defense Department believed that the Naval Base at Guantánamo Bay, Cuba might work, so it asked the Justice Department’s Office of Legal Counsel (OLC) whether federal courts would entertain habeas corpus petitions filed by prisoners at Guantánamo, or whether they would dismiss such petitions as beyond their jurisdiction. On December 28, 2001, OLC responded with a thorough and balanced analysis of how the federal courts were likely to resolve the jurisdictional question. The memorandum prepared by OLC explained the arguments against such jurisdiction, but it also explored possible strengths in the opposing position. The memorandum predicted that federal courts would not exercise jurisdiction but explained the risk of a contrary ruling. Acting in reliance on this memorandum, the government started imprisoning and interrogating alleged al Qaeda members at Guantánamo the following month, cognizant of the risk that a federal court might find habeas jurisdiction.
Two innovative federal crimes, 18 U.S.C. §2339A and §2339B, have been frequently charged in prosecutions since September 11, 2001, becoming key elements in the government’s anti-terrorism efforts. Section 2339A makes it a federal crime knowingly to provide material support or resources in preparation for or in carrying out specified crimes of terrorism. Section 2339B prohibits knowingly providing material support to a foreign terrorist organization. These statutes bear some resemblance to criminal liability for complicity, but they are being used like a new kind of conspiracy charge, the familiar “prosecutor’s darling.” Unlike conspiracy, they are framed with a mens rea of knowledge rather than of purpose, and because they are substantive offenses, they can be combined with traditional conspiracy charges. These provisions can be used to impose punishment for conduct remote from the commission of criminal harms, often conduct involving minimal and outwardly non-criminal acts.