Suhail Najim Abdullah AL SHIMARI; Taha Yaseen Arraq Rashid; Sa‘ad Hamza Hantoosh Al–Zuba‘e; Salah Hasan Nusaif Jasim Al–Ejaili, Plaintiffs–Appellees, v. CACI INTERNATIONAL, INCORPORATED; CACI Premier Technology, Incorporated, Defendants–Appellants. Kellogg Brown & Root Services, Incorporated, Amicus Supporting Appellants.
Four Iraqi citizens, who were seized by the U.S. military in the Iraq war zone and detained by the military in Abu Ghraib prison, near Baghdad, commenced this tort action against a civilian contractor, retained by the military to assist it at the prison in conducting interrogations for the purpose of obtaining intelligence. The plaintiffs allege that while they were detained, the contractor's employees and military personnel conspired among themselves and with others to torture and abuse them and to cover up that conduct.
The contractor filed a motion to dismiss on numerous grounds, including the political question doctrine; federal pre-emption under Boyle v. United Technologies Corp., 487 U.S. 500 (1988), and Saleh v. Titan Corp., 580 F.3d 1 (D.C.Cir.2009); and derivative sovereign immunity. The district court denied the contractor's motion, concluding that the “[p]laintiffs' claims are justiciable because civil tort claims against private actors for damages do not interfere with the separation of powers”; that defendant's claim of immunity must be developed through discovery, and dismissal now would be premature; and that plaintiffs' claims “are not preempted by the combatant activities exception at this stage because discovery is required to determine whether the interrogations here constitute ‘combatant activities' within the meaning of the exception.” Al Shimari v. CACI Premier Technology, Inc., 657 F.Supp.2d 700, 731 (E.D.Va.2009).
On the contractor's appeal, we reverse and remand with instructions to dismiss this case. We conclude that the plaintiffs' state law claims are preempted by federal law and displaced by it, as articulated in Saleh v. Titan Corp., 580 F.3d 1, 8–12 (D.C.Cir.2009).
In response to the unprovoked attacks on the United States on September 11, 2001, during which some 3,000 people were killed, a multi-national force, led by the United States and Great Britain, invaded Iraq in March 2003 to depose Sad-dam Hussein and rid Iraq of weapons of mass destruction. While Hussein was quickly deposed and no weapons of mass destruction were found, the war in Iraq continued at least for the period relevant to the claims asserted in this action. Indeed, according to various published data, a substantial number of deaths and casualties of both Iraqi civilians and members of the U.S. military continued even up to the time of oral argument, although at a reduced level from the peak in 2006 and 2007. See, e.g., Hannah Fischer, Cong. Research Serv., R40824, Iraq Casualties (Oct. 7, 2010), available at www.fpc.state.gov/documents/organization/150201.pdf; U.S. Casualties in Iraq, www.gl obalsecurity.org/military/ops/iraq_casualties.htm (last visited Jan. 10, 2011).
During the course of the war, the U.S. military seized and detained Iraqi citizens suspected of being enemy combatants or thought to have value in possessing useful intelligence. Some of these detainees were imprisoned at Abu Ghraib prison, near Baghdad. Although the prison was operated in the war zone by the United States Army, “a severe shortage” of military intelligence personnel “prompt[ed] the U.S. government to contract with private corporations to provide civilian interrogators and interpreters.” J .A. 408. These contractors included CACI Premier Technology, Inc., a subsidiary of CACI International, Inc. (collectively herein, “CACI”). The contractors were required to comply with Department of Defense interrogation policies and procedures when conducting “[i]ntelligence interrogations, detainee debriefings, and tactical questioning” of persons in the custody of the U.S. military. J.A. 270–71.
In the Executive Summary of the Senate Armed Services Committee Inquiry into the Treatment of Detainees in U.S. Custody, the Committee detailed the history of the standards and practices applied in interrogations at Guantanamo Bay, Iraq, and Afghanistan. J.A. 360–65. The Executive Summary noted that the President signed an order on February 7, 2002, stating that the Third Geneva Convention did not apply to the conflict with al-Qaeda and the Taliban and that detainees were not entitled to the protections afforded prisoners of war by the Third Geneva Convention. But the order stated that, as “a matter of policy, the United States Armed Forces shall continue to treat detainees humanely and, to the extent appropriate and consistent with military necessity, in a manner consistent with the principles of the Geneva Conventions.” J.A. 354 (emphasis added). Later, in December 2002, following requests from the field to employ aggressive interrogation techniques to obtain intelligence, the Secretary of Defense approved a list of techniques for interrogation, such as stress positions, removal of clothing, use of phobias (such as fear of dogs), and deprivation of light and auditory stimuli. J.A. 360. While the approval was directed at interrogations being conducted at Guantanamo Bay, it was also circulated to military personnel in Iraq and Afghanistan. J.A. 363. But even as aggressive techniques were being employed for interrogation conducted in those theatres, the Secretary rescinded his memorandum approving the specific techniques. J.A. 363. It was unclear, however, what techniques thereafter remained authorized by the Secretary. J.A. 363–64. During the following year, high-level military personnel directed that interrogators in Iraq be more aggressive—telling field personnel that “the gloves are coming off” and “we want these detainees broken.” J.A. 365.
While the record reflects an ongoing policy not to engage in torture, the definition of torture was the subject of continuing debate in the Executive Branch and the military. See J.A. 356–60. Nonetheless, the military believed it to be in the national interest to pursue intelligence through aggressive interrogation techniques, inasmuch as intelligence, especially in the context of the wars in Iraq and Afghanistan, was an especially significant tool of war. Even so, the Senate Armed Services Committee concluded that the approval and use of aggressive techniques were a direct cause of detainee abuse inasmuch as they conveyed a message that it was acceptable to mistreat and degrade detainees in U.S. custody.
While some of the abuses that the plaintiffs detailed in the allegations of their complaint appear to have been approved by the military at one point or another, others were clearly not.
The four Iraqi citizens who commenced this action—Suhail Najim Abdullah Al Shimari, Taha Yaseen Arraq Rashid, Sa‘ad Hamza Hantoosh Al–Zuba‘e, and Salah Hasan Nusaif Jasim Al–Ejaili—were detained by the U.S. military in Abu Ghraib prison during various periods between 2003 and 2008. They alleged that during their detention, they were interrogated in dangerous and unauthorized stress positions; that they were subjected to sexual assault, repeated beatings, deprivations of food, water and sleep, forced witnessing of the rape of another prisoner, and imprisonment under conditions of sensory deprivation; and that the facts of abuse were covered up. They allege that the abuse and cover-up were carried out by CACI employees in conspiracy with U.S. military personnel.
After the district court granted CACI's motion to stay discovery, CACI filed a motion under Rules 12(b)(1) and 12(b)(6) to dismiss, based on numerous grounds, including the political question doctrine, federal preemption, and derivative sovereign immunity. The district court denied the motion, and CACI filed this interlocutory appeal, challenging the district court's rulings on immunity and on the defenses involving the political question doctrine and federal preemption. See Nixon v. Fitzgerald, 457 U .S. 731 (1982) (recognizing that a ruling on the President's absolute immunity based on the separation of powers was immediately appealable); see also Al–Quraishi v. L–3 Servs., Inc., ––– F.3d ––––, Nos. 10–1891 & 10–1921 (4th Cir. Sept. 21, 2011) (holding that an appeal raising the same issues presented here is immediately appealable).
Considering CACI's preemption challenge, we conclude, based on the uniquely federal interests involved in this case, that the plaintiffs' tort claims are preempted and displaced under the reasoning articulated in Boyle v. United Technologies Corp., 487 U.S. 500 (1988), as applied to circumstances virtually identical to those before us in Saleh v. Titan Corp., 580 F.3d 1 (D.C.Cir.2009), cert. denied, ––– U.S. ––––, No. 09–1313, 2011 WL 2518834 (June 27, 2011). In Saleh, the D.C. Circuit held that where a civilian contractor is integrated into combat activities over which the military retains command authority, a tort claim arising out of the contractor's engagement in such activities is preempted. Saleh, 580 F.3d at 9. In reaching its holding, the court applied the rationale of Boyle to circumstances practically identical to those before us.
In Boyle, a marine pilot's estate filed suit under Virginia tort law against United Technologies Corporation, a civilian contractor of the Department of Defense, alleging the negligent design of a helicopter. Boyle, 487 U.S. at 503. When the helicopter crashed into the water, the pilot was unable to open the escape hatch, which opened outward rather than inward, causing the pilot to drown. Id. United Technologies contended that the door's design was specified by the Department of Defense and that the uniquely federal interests implicated by its procurement from civilian contractors preempted Virginia tort law, and the Supreme Court agreed. The Court determined that the contractor should not be held liable for implementing the government's design and that entertaining the pilot's tort case would undermine the unique federal interests in the procurement of equipment for the national defense. If state tort liability were permitted, the federal interests would be adversely affected because “either the contractor [would] decline to manufacture the design specified by the Government, or it [would] raise its price.” Id. at 507.
The Boyle Court held that protecting these uniquely federal interests conflicted with the purposes and operation of state tort law and therefore the state law was preempted. It looked to the discretionary function exception of the Federal Tort Claims Act to demonstrate that the federal government must have the flexibility to select the appropriate design for military equipment and that allowing state tort liability for a defective design, where the government had participated in that design, would significantly conflict with the policy embodied in and defined by the Federal Tort Claims Act's discretionary function exception. Id. at 511–12. Thus, Boyle recognized a government contractor preemption defense and applied it so that contractors would be protected from state law liability where such protection was necessary to safeguard uniquely federal interests.
While Boyle's preemption holding thus functions to displace state law to protect “uniquely federal interests,” it did not rely on any act of Congress to animate the preemption. Rather, the Boyle preemption, which leaves no federal law addressing the claim, operates more in effect like sovereign immunity that is extended to protect civilian government contractors' functioning on behalf of the sovereign. Thus, the shape of Boyle preemption, rather than being defined by the presence of federal law, is defined by the priority of uniquely federal interests over countervailing state interests as manifested in state law.
As did the courts in Boyle and Saleh, we too conclude that this case implicates important and uniquely federal interests. The potential liability under state law of military contractors for actions taken in connection with U.S. military operations overseas would similarly affect the availability and costs of using contract workers in conjunction with military operations. In this case, that uniquely federal interest was especially important in view of the recognized shortage of military personnel and the need for assistance in interrogating detainees at Abu Ghraib prison. Not only would potential tort liability against such contractors affect military costs and efficiencies and contractors' availability, it would also present the possibility that military commanders could be hauled into civilian courts for the purpose of evaluating and differentiating between military and contractor decisions. That effort could become extensive if contractor employees and the military worked side by side in questioning detainees under military control, as the complaint alleges in this case. Moreover, such interference with uniquely federal interests would be aggravated by the prison's location in the war zone. Finally, potential liability under state tort law would undermine the flexibility that military necessity requires in determining the methods for gathering intelligence.
The dissenting opinion takes the position that CACI should not enjoy any immunity from liability based on its repeated (and wrong) assertions that CACI acted independently, apart from the military, and “contrary to military directives.” Post, at 41; see also post, at 30 (noting that “no federal interest encompasses the torture and abuses that plaintiffs allege”); post, at 32–33 (“Ultimately, the government rather than the contractor must be in charge of decisionmaking in order for the contractor to be shielded from liability”); post, at 33 (“the government's precise control over its contractor, which is so integral to Boyle's reasoning, is absent”); post, at 34 (noting that government authority for alleged conduct can only be determined by looking at CACI's contract with the military); post, at 35 (“there is no evidence to support the majority's supposition of ‘integration’ ․ other than what can be gleaned from the bare allegations of the Complaint”); post, at 36 (arguing that absence of government role precludes application of Boyle). But the dissent's position is belied by the allegations of the complaint, which assert that all the misconduct charged was the product of a conspiracy between CACI personnel and military personnel. See, e.g., Amended Complaint ¶ 1 (alleging a conspiracy between CACI employees and military personnel, who are “now serving time in military prison” for their participation); ¶ 70 (alleging that “CACI employees repeatedly conspired with military personnel to give Plaintiffs the ‘special treatment’ which was code for torture of the type endured by Plaintiffs in this hard site”); ¶ 71 (alleging that “CACI employees repeatedly conspired with military personnel to harm Plaintiffs in various manners and methods referred to above”); ¶ 118 (alleging that CACI employees “agreed with each other and others to participate in a series of unlawful acts”); ¶ 124 (alleging that CACI employees “aided and abetted others who were torturing Plaintiffs”); and ¶ 135 (alleging that CACI's “knowing participation in the conspiracy caused grave and foreseeable damages to Plaintiffs”). In view of these allegations of the complaint, which, at this stage, we accept as true, we can only assume for purposes of our decision that CACI employees were integrated into the military activities at Abu Ghraib prison in Baghdad, over which the military retained command authority.
In addition to the specific adverse impacts on the uniquely federal interests of interrogating detainees in foreign battlefields, a broader and perhaps more significant conflict with federal interests would arise from allowing tort law generally to apply to foreign battlefields. “[T]he traditional rationales for tort law—deterrence of risk-taking behavior, compensation of victims, and punishment of tortfeasors—are singularly out of place in combat situations.” Saleh, 580 F.3d at 7 (emphasis omitted). In Boyle, the Supreme Court looked to the Federal Tort Claims Act exceptions for the purpose of determining whether a significant conflict between state tort law and federal interests existed. Although the relevant Federal Tort Claims Act provision in Boyle was the discretionary function exception, when we employ the same approach to determine the nature and extent of any conflict here, the relevant provision is the combatant activities exception. See 28 U.S.C. § 2680(j). This exception retains the United States' sovereign immunity for claims “arising out of the combatant activities of the military or naval forces, or the Coast Guard, during time of war.” Id. As the D.C. Circuit observed in Saleh, Congress intended the exception to “eliminat[e] ․ tort from the battlefield, both to preempt state or foreign regulation of federal wartime conduct and to free military commanders from the doubts and uncertainty inherent in potential subjection to civil suit.” Saleh, 580 F.3d at 7. And we agree with the D.C. Circuit's conclusion that this interest is implicated even when the suit is brought indirectly—against a civilian contractor—rather than directly against the United States itself. The acuteness of a need to preempt state tort law in the context of battlefield activities is well articulated in Saleh:
The nature of the conflict in this case is somewhat different from that in Boyle—a sharp example of discrete conflict in which satisfying both state and federal duties (i.e., by designing a helicopter hatch that opens both inward and outward) was impossible. In the context of the combatant activities exception, the relevant question is not so much whether the substance of the federal duty is inconsistent with a hypothetical duty imposed by the state or foreign sovereign. Rather, it is the imposition per se of the state or foreign tort law that conflicts with the FTCA's policy of eliminating tort concepts from the battlefield. The very purposes of tort law are in conflict with the pursuit of warfare. Thus, the instant case presents us with a more general conflict preemption, to coin a term, “battle-field preemption”: the federal government occupies the field when it comes to warfare, and its interest in combat is always “precisely contrary” to the imposition of a non-federal tort duty.
Saleh, 580 F.3d at 7 (emphasis added) (citing Boyle, 487 U .S. at 500).1
The uniquely federal interest in conducting and controlling the conduct of war, including intelligence-gathering activities within military prisons, thus is simply incompatible with state tort liability in that context.
This case involves allegations of misconduct in connection with the essentially military task of interrogation in a war zone military prison by contractors working in close collaboration with the military. We hold that under these circumstances, where a civilian contractor is integrated into wartime combatant activities over which the military broadly retains command authority, tort claims arising out of the contractors' engagement in such activities are preempted. See Saleh, 580 F.3d at 9.
The nation rightly reacted with moral indignation to the pictures circulated from Abu Ghraib prison. And if these four Iraqi citizens did in fact suffer in a similar manner from the unauthorized conduct of military and civilian guards and interrogators, the nation, including its judges, would react similarly. Nothing we say in this opinion is intended to condone the torture, abuse, and cover-up alleged in the complaint.
Of course, nothing we say should be taken as passing judgment on the substance of these allegations. For our purposes, they remain allegations that we have accepted as true, but only for purposes of deciding this appeal.
What we hold is that conduct carried out during war and the effects of that conduct are, for the most part, not properly the subject of judicial evaluation. The Commander in Chief and the military under him have adopted policies, regulations, and manuals and have issued orders and directives for military conduct, and they have established facilities and procedures for addressing violations and disobedience. On this structural ground alone, and not on any judgment about the conduct itself, we are requiring that the claims of these four Iraqi detainees alleging abuse in a military prison in Iraq be dismissed by the district court.
Therefore, we reverse the district court's order denying CACI's motion to dismiss and remand with instructions to dismiss.
REVERSED AND REMANDED WITH INSTRUCTIONS
I would conclude that in addition to preemption, the political question doctrine under Baker v. Carr, 369 U.S. 186 (1962), and derivative absolute immunity under Mangold v. Analytic Services, Inc., 77 F.3d 1442 (4th Cir.1996), require dismissal of this case. I note that Judge King would apparently agree with application of the political question doctrine were he to have addressed the issue. See Taylor v. Kellogg, ––– F.3d ––––, No. 10–1543 (4th Cir. Sept. 21, 2011) (King, J.). But, in his dissenting opinion, he has chosen to address only federal preemption and not the political question doctrine. Nor has he addressed derivative absolute immunity, even though all three issues were raised by CACI on appeal.
On the political question issue, CACI contends that plaintiffs' claims are nonjusticiable because the conduct of its employees, on which the claims are based, was part of the military effort undertaken in a war zone and resolution of those claims would inextricably be tied to an evaluation of the exercise of war powers, committed under Articles I and II of the Constitution to coordinate political branches. See Baker, 369 U.S. at 208–17. More specifically, CACI argues that the interrogation techniques, which lie at the core of plaintiffs' claims, were an inseparable component of war, and that “many if not most of the alleged forms of abuse here were interrogation techniques approved at the highest levels of the Executive Branch.” CACI adds that it is not relevant whether the “chosen techniques were in fact appropriate—that is precisely the political question that the courts may not ask or answer,” citing Lin v. United States, 561 F.3d 502, 507 (D.C.Cir.2009).
The plaintiffs argue that resolution of their claims is not textually committed by the Constitution to coordinate political branches but, because their claims are tort claims, to the Judicial Branch. The plaintiffs note that their “tort claims do not even arise out of actions by a coordinate political branch.” Rather, the tort claims arise from conduct by CACI, which is not, nor is it like, a coordinate branch of government. They also argue that the torture allegedly committed by CACI employees was never authorized by the military.
The political question doctrine, at its core, recognizes as nonjusticiable any question whose resolution is committed to a coordinate branch of government and whose evaluation by a court would require the application of standards judicially undiscoverable or judicially unmanageable. As the Baker Court summarized, “The nonjusticiability of a political question is primarily a function of the separation of powers.” 369 U.S. at 210. Even so, the “delicate exercise” of determining whether questions are indeed political remains the responsibility of the Judicial Branch as the “ultimate interpreter of the Constitution.” Id. at 211.
In Baker, the Court analyzed prior representative decisions of the Court “to infer from them the analytical threads that make up the political question doctrine.” Id. It observed, for example, that earlier foreign relations cases presented political questions where they turned on “standards that defy judicial application,” thus demanding the “single-voiced statement” of the government's views. Id. In another example, it observed, in connection with the war powers, that “isolable reasons for the presence of political questions” arise in determining “when or whether a war has ended,” id. at 213, and it pointed out that the war power “includes the power to remedy the evils which have arisen from its rise and progress and continues during that emergency,” id. at 213 (internal quotation marks omitted). Distilling the core nature of political questions, the Court explained that a “lack of judicially discoverable standards and the drive for even-handed application” requires referring such questions to the political departments. Id. at 214. The Court summarized the circumstances that present a political question:
It is apparent that several formulations which vary slightly according to the settings in which the questions arise may describe a political question, although each has one or more elements which identify it as essentially a function of the separation of powers. Prominent on the surface of any case held to involve a political question is found a textually demonstrable constitutional commitment of the issue to a coordinate political department; or a lack of judicially discoverable and manageable standards for resolving it; or the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion; or the impossibility of a court's undertaking independent resolution without expressing lack of the respect due coordinate branches of government; or an unusual need for unquestioning adherence to a political decision already made; or the potentiality of embarrassment from multifarious pronouncements by various departments on one question.
Unless one of these formulations is inextricable from the case at bar, there should be no dismissal for non-justiciability on the ground of a political question's presence.
Baker, 369 U.S. at 217. In short, the substantial presence of any one of the articulated formulations would indicate a political question.
The Baker formulations led Justice Powell to distill the inquiry into three questions:
(i) Does the issue involve resolution of questions committed by the text of the Constitution to a coordinate branch of government?
(ii) Would resolution of the question demand that a court move beyond areas of judicial expertise?
(iii) Do prudential considerations counsel against judicial intervention?
Goldwater v. Carter, 444 U.S. 996, 998 (1979) (Powell, J., concurring in the judgment).
To answer the first question, I begin by noting that the claims made in this case arose in the context of the war in Iraq. And seizing, in the war zone, foreigners suspected of hostile activity or of possessing useful intelligence and then interrogating them in the field were integral parts of the war effort. Indeed, the function of detaining and interrogating to obtain intelligence was undoubtedly critical to the success of military strategies and campaigns. In such circumstances, the judgment of whom to interrogate, what to inquire about, and the techniques to use falls comfortably within the powers of the Commander in Chief and his subordinates in the chain of command.
It is not disputed that this power to conduct war and determine its objectives and means is explicitly committed by the Constitution to Congress and the President. See U.S. Const. art. I, § 8, cl. 11–14 (authorizing Congress to declare war, to raise armies and create a navy, and to make rules for the military); id. art. II, § 2 (providing that the President “shall be Commander in Chief of the Army and Navy of the United States, and of the Militia of the several States, when called into the actual Service of the United States”). This assignment to the President was deliberate and considered. As the Federalist papers explain, “Of all the cares or concerns of government, the direction of war most peculiarly demands those qualities which distinguish the exercise of power by a single hand. The direction of war implies the direction of the common strength; and the power of directing and employing the common strength, forms a usual and essential part in the definition of the executive authority.” The Federalist No. 74, at 383 (Alexander Hamilton, March 25, 1788) (George W. Carey & James McClellan eds., 1990).
We must thus ask whether plaintiffs' claims, arising in the context of a war, challenge the exercise of war powers so committed to coordinate political branches. While it would certainly be so if their challenges were directed to military actions and personnel, when directed at a conspiracy of U.S. military personnel and employees of civilian contractors engaged to conduct military functions, the issue is more nuanced. Making the question more complex is the allegation that the members of the conspiracy are alleged to have disobeyed orders and violated limits established by persons higher in the chain of military command.
In a case brought against the military directly, rather than a contractor, the allegation that a soldier disobeyed orders surely would not make the claim justiciable if it otherwise was nonjusticiable. Thus, if interrogation was designed to uncover the location and names of enemy personnel and their plans, the fact that a military interrogator applied techniques more aggressive than those approved by his commander for aggressive interrogation would not remove the activity from the military effort, any more than would a soldier's shooting an enemy soldier even after he had been seized and disarmed. Such conduct, albeit disobedient, is undertaken grossly in the course of prosecuting war and advancing the strategy of the military adopted by upper level commanders for carrying out the war. Just as the President and his designees are given the authority to conduct the war and interrogate battlefield prisoners free from judicial oversight, they are given the authority to address disobedience and impose discipline.
To be sure, this analysis, when applied to conduct engaged in by civilian contractors, becomes more attenuated because civilian contractors do not enjoy every protection from suit that the military might enjoy. As our dissenting colleague recognizes in another opinion today dismissing a claim against a military contractor based on the political question doctrine, “we are obliged to carefully assess the relationship between the military and [the contractor], and to ‘look beyond the complaint, [and] consider[ ] how [the plaintiff] might prove [his] claim[ ] and how [defendant] would defend.’ “ Taylor v. Kellogg, ––– F.3d at ––––, No. 10–1543 (4th Cir. Sept. 21, 2011), slip op. at 13 (quoting Lane v. Halliburton, 529 F.3d 548, 565 (5th Cir.2008) (some alterations in original)). When, as here, this assessment demonstrates that the civilian contractors were working side by side with military personnel to carry out military operations under the ultimate supervision and command of the military in a war zone, evaluation of their conduct raises the same political question that would be raised by a direct challenge to the military.
CACI's function here (interrogating persons seized by the military for interrogation) was ultimately a military function under the control of the military, and therefore the decision to dismiss the plaintiff's claims is not affected by the fact that CACI was a civilian contractor. The U.S. military had picked up the detainees in the war zone and believed that they should be interrogated. The detainees remained in the custody of the military throughout interrogations, and the military both operated and guarded the prison. Because of personnel shortages, however, the interrogation activities were carried out not only by military personnel but also by civilian employees engaged to perform the same function. They were instructed on approved interrogation techniques and ordered not to violate the limitations. In addition, the intelligence being sought through interrogation was defined by the military's goals such that the substance of the questions posed to detainees was of U .S. military origin. Moreover, the actions complained of are alleged to have been committed jointly by CACI employees and military personnel, and all activities are alleged to have fallen within the scope of a conspiracy that included CACI employees and military personnel.
Accordingly, in response to the first question in considering the political question doctrine—whether resolution of the questions in this case is committed by the text of the Constitution to a coordinate branch of government—I conclude that the answer is undoubtedly yes, even though the allegations may involve imperfect or disobedient conduct by contractors.
The answer to the second question is more complicated and requires a determination of whether resolution of the plaintiffs' claims challenging aggressive interrogation techniques would take the courts into areas beyond their judicial expertise or competence. See Taylor, ––– F.3d at ––––, No. 10–1543, slip op. at 14 (“[W]e must, to resolve this appeal, gauge the degree to which national defense interests may be implicated in a judicial assessment of [the tort claim]”).
As a central component of conducting war, the President, the Executive Branch, and the military determined that aggressive interrogation techniques were a military necessity inasmuch as the war in Iraq involved an enemy that was spread out among numerous factions and cells within the population, without a distinguishing organization, uniforms, or bases of operation. Thus, as a matter of policy, the President found it inconsistent with military necessity to afford seized enemy combatants the protections of the Third Geneva Convention. And in carrying out that determination, the Secretary of Defense and high-level military officers directed that aggressive interrogation be employed. There was, to be sure, a debate within the Executive Branch about what were morally appropriate techniques and what could be justified by military necessity. But these questions were not addressed by applying standards that were judicially cognizable; they were difficult judgments that involved a delicate weighing of public policy, the public sense of morality, public decency, the customs of war, international treaties, and military necessity. One could hardly find a question more unsuited for the judiciary.
Indeed, in any given war, the President might choose to impose no limits on specific military actions ordered. For example, we know that in connection with the response to the 9/11 attacks launched against the United States, the President considered, and perhaps even approved, an order to shoot down a U.S. civilian airplane carrying innocent American citizens, determining that the order was in the greater public interest. In that case, the President had information that the airplane was headed for the White House or the Capitol in Washington, D.C. That type of question could hardly have been addressed or reviewed by a court, which would have had few if any standards to apply.
That is not to say that the evaluation of battlefield interrogations calls for the same intensity of response as does the response to an enemy-captured civilian airplane en route to the nation's capital. Nonetheless, interrogation was a military tool for use in prosecuting the war effort. To engage a court in the question of which techniques were militarily necessary but yet morally acceptable and consistent with American policy, at least as defined by the President and Congress, would require a court to exercise the very powers committed to those branches. The military necessity of actions in the war zone, including battlefield interrogations of detainees, cannot be explored by a court without requiring it to evaluate judgments about which the judiciary lacks expertise and competence. For a court to evaluate military policy that interrogation had to be more aggressive, that “the gloves are coming off,” and that “these detainees must be broken,” it would have to evaluate the entire basis for the military decisions or be at a loss as to where to begin. Such questions go to the heart of the political question doctrine.
On this question, as noted above with respect to military personnel who disobey orders, the fact that CACI employees may have disobeyed orders does not remove their activities from the military function and would not change the analysis. A court's attempt to evaluate the disobedient activities of CACI employees would inappropriately enmesh the court into military strategies, decisions, and activities to the same extent as if they were undertaken entirely by military personnel. The political question doctrine recognizes that the Constitution assigns such matters to Congress, the Commander in Chief, and the Executive Branch generally. See Tiffany v. United States, 931 F.2d 271, 277 (4th Cir.1991) (“Of the legion of governmental endeavors, perhaps the most clearly marked for judicial deference are provisions for national security and defense․ The strategy and tactics employed on the battlefield are clearly not subject to judicial review”).
Finally, addressing Justice Powell's third question, I conclude that it would be imprudent for civilian courts to attempt to adjudge military acts under common law tort principles. To entertain the plaintiffs' claims under those principles would introduce, for the first time, tort principles in a field of battle, raising a yet broader array of interferences by the judiciary into the military functions committed to Congress, the President, and the Executive Branch. When deciding whether this claim raises a political question, we must assess “first, the extent to which [the contractor] was under the military's control, and second, whether national defense interests were closely intertwined with the military's decisions governing [the contractor's] conduct.” Taylor, ––– F.3d at ––––, No. 10–1543, slip op. at 17. Here, the CACI was engaged by the military to pursue interrogations under the command and control of military personnel, and decisions about the scope and nature of these interrogations, even more so than decisions about “whether back-up power should have been supplied” to a particular area, id. at 18, were intricately intertwined with national defense interests.
For these reasons, I would defer to the political branches for how best to manage military prisons, to interrogate detainees for military intelligence, and to punish those within the prison who disobey military directives. See Carmichael v. Kellogg, Brown & Root Servs., Inc., 572 F.3d 1271 (11th Cir.2009).
I would also conclude that this suit is barred by the doctrine of derivate absolute immunity, as articulated in Mangold v. Analytic Services, Inc., 77 F.3d 1442 (4th Cir.1996). See also Murray v. Northrop Grumman Info. Tech., Inc., 444 F.3d 169, 175 (2d Cir.2006) (government contractor absolutely immune from tort liability for performing contracted-for governmental function, citing Mangold, 77 F.3d at 1447); Pani v. Empire Blue Cross/Blue Shield, 152 F.3d 67, 71–73 (2d Cir.1998) (same); Midland Psychiatric Assocs., Inc. v. United States, 145 F.3d 1000, 1005 (8th Cir.1998) (common law official immunity barred tort suit against Medicare insurer). Derivative absolute immunity protects contractors from suit where such immunity is necessary to protect a discretionary government function and the benefits of immunity outweigh its costs.
In Mangold, we held that a government contractor was absolutely immune from a state tort suit for defamation based on statements that the contractor made in response to an official government investigation about its dealings with the government. There, the Air Force had conducted an investigation into the activities of an Air Force colonel who allegedly exerted his influence to pressure a government contractor to hire a family friend. Mangold, 77 F.3d at 1444–45. In response to questions posed by the Air Force, the contractor provided information to the Air Force confirming that the colonel did indeed press the contractor to hire the friend, despite the friend's lack of credentials for the position. Id. Following the contractor's response to the Air Force, the colonel sued the contractor for defamation under Virginia law. Id. We concluded that the discretionary governmental action of investigating suspected fraud was protected by absolute immunity and that the immunity extended “to persons in the private sector who are government contractors participating in official investigations of government contracts” “to the extent that the public benefits obtained by granting immunity outweigh[ed] its costs.” Id. at 1447. Such immunity could be extended to a private contractor because the “immunity [was] defined by the nature of the function being performed and not by the office or the position of the particular employee involved.” Id. Thus,
[i]f absolute immunity protect[ed] a particular government function, no matter how many times or to what level that function [was] delegated, it [was] a small step to protect that function when delegated to private contractors, particularly in light of the government's unquestioned need to delegate governmental functions. The government cannot perform all necessary and proper services itself and must therefore contract out some services for performance by the private sector. When the government delegates discretionary governmental functions through contracting with private contractors, therefore, the same public interest identified in Barr [v. Matteo, 360 U.S. 564 (1959) ] and Westfall [v. Erwin, 484 U.S. 292 (1988) ]—the interest in efficient government—demands that the government possess the ability meaningfully to investigate these contracts to ensure that they are performed without fraud, waste, or mismanagement.
Id. at 1447–48 (emphasis added).
As in Mangold, the military made the discretionary determination to interrogate detainees and required the assistance of civilian contractors to perform the interrogations. Here, as in Mangold, extending immunity to the contractors is necessary to protect the underlying discretionary governmental activity, in this case, performing wartime interrogations.
Nonetheless, for derivative absolute immunity to apply, its benefits must outweigh its costs. The costs of immunity obviously arise from denying injured parties access to courts to assert otherwise legitimate claims. Its benefit is that it prevents vexatious litigation from impairing the efficient functioning of government. In Mangold, we concluded that the government had a strong interest in receiving contractor assistance during investigations of contracting improprieties, and that such assistance would be less forthcoming if contractors could be subject to suit for their participation. Mangold, 77 F.3d at 1447. The court held that this interest outweighed that of potentially defamed individuals in seeking compensation. Id. Here, the military had a strong need to receive contractor assistance in its interrogations because of a substantial shortage of personnel. And interrogations were a major component of the war effort designed to gather military intelligence. Like in Mangold, subjecting contractors to tort actions would risk interference with interrogations, as well as the availability of civilian assistance. Because of the important public interest in the effective prosecution of war and the alternative mechanisms already in place to ensure against, and compensate for, the abuse for which the plaintiffs seek compensation in this case, I would conclude, as in Mangold, that the benefits of immunity outweigh its costs.
At bottom, I would rely on these additional grounds—the political question doctrine and derivative absolute immunity—to reverse the district court's order and remand this case to the district court for dismissal.
I write to dissent from my distinguished colleagues in the majority. For the same reasons I discuss at length in my dissenting opinion in our companion case of Al–Quraishi v. L–3 Services, Inc., ––– F.3d ––––, No. 10–1891(L) (4th Cir. Sept. 21, 2011), we lack jurisdiction over this interlocutory appeal to decide, as the majority does, that the plaintiffs' claims are preempted by federal law. Were we authorized to adjudicate the merits of the preemption defense, however, we should rule it unavailing here.
The plaintiffs' claims arise from their maltreatment while detained at the Abu Ghraib prison during our nation's military campaign in Iraq. According to the operative Amended Complaint (the “Complaint”), the allegations of which we are bound to take as true at this stage of the proceedings, civilian employees of CACI International, Inc., and CACI Premier Technology, Inc. (collectively “CACI”), while interrogating the plaintiffs or assisting in their interrogation, conspired with military personnel to “instigate[ ], direct [ ], participate[ ] in, [and] aid[ ] and abet[ ] conduct towards detainees that clearly violated the Geneva Conventions, the Army Field Manual, and the laws of the United States.” Complaint ¶ 67.1 One plaintiff alleges that he was “forcibly subjected to sexual acts by a female as he was cuffed and shackled to cell bars,” was “dragged by a rope where part of it was tied tightly to his penis,” and was “subjected to [a] mock execution.” Id. ¶¶ 32, 37, 39. Other asserted abuses include beatings, food and sleep deprivation, humiliation, and being forced to witness the rape of a female detainee. See generally id. ¶¶ 11–63.
The Complaint relates that CACI has “admitted ․ that it had the ability to control, direct and influence the actions performed by employees,” and it insists that CACI was able “to prevent employees from torturing plaintiffs.” Complaint ¶¶ 76–77. The plaintiffs further maintain that “CACI at all times [was] obliged by the terms of its contract to supervise [its] employees.” Id. ¶ 78. CACI was aware, according to the plaintiffs, “that the United States intended and required that any person acting under the contract [with] the United States would conduct themselves in accordance with the relevant domestic and international laws.” Id. ¶ 98. Nonetheless, by engaging in and directing the torture of the plaintiffs, CACI “directly contradicted the contract terms, domestic law and the United States' express policy against torture.” Id. ¶ 115. CACI, the plaintiffs say, is consequently liable to them under Virginia law for the torts of assault and battery, sexual assault, intentional and negligent infliction of emotional distress, and negligent hiring and supervision.
Pursuant to Rules 12(b)(1) and 12(b)(6) of the Federal Rules of Civil Procedure, CACI moved to dismiss the Complaint, asserting, among other things: (1) that the suit raised a nonjusticiable political question; (2) that CACI was entitled to immunity derived from its association with the sovereign; and (3) that, as a logical extension of the Supreme Court's decision in Boyle v. United Technologies Corp., 487 U.S. 500 (1988), the plaintiffs' state law claims were preempted, having arisen in the context of combatant activities conducted in the federal interest. The district court denied CACI's motion, rejecting its argument that the plaintiffs' claims were nonjusticiable. See Al Shimari v. CACI Premier Tech., Inc., 657 F.Supp.2d 700, 708–14 (E.D.Va.2009). The court declined to decide the immunity issue at the dismissal stage, concluding that it could not “determine the scope of Defendants' government contract, the amount of discretion it afforded Defendants in dealing with detainees, or the costs and benefits of recognizing immunity in this case without examining a complete record after discovery has taken place.” Id. at 714. The limited record, according to the district court, also cast doubt that CACI's interrogation practices amounted to “combatant activities.” Id. at 725. The court ruled that the plaintiffs' claims were in any event “not preempted under Boyle,” because they “do not present a significant conflict with a uniquely federal interest.” Id.2 Five days following the district court's ruling, before discovery could commence, CACI noted this appeal.
I need not reiterate in extravagant detail why jurisdiction over this appeal is lacking, having devoted considerable space to the subject in my dissenting opinion in today's companion case of Al–Quraishi v. L–3 Services, Inc., –––F.3d ––––, No. 10–1891(L) (4th Cir. Sept. 21, 2011). Suffice it to say that the only basis that could arguably support the exercise of collateral order jurisdiction, see Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541 (1949), the denial of dismissal on the ground of derivative sovereign immunity, was not “conclusively determined” by the district court as required by Will v. Hallock, 546 U.S. 345, 349 (2006). The denials of dismissal based on the political question doctrine and on Boyle preemption, as applied by the District of Columbia Circuit in Saleh v. Titan Corp., 580 F.3d 1 (D.C.Cir.2009), though conclusively determined, abridged no immunity. As a result, neither ground meets the additional prerequisite of being “effectively unreviewable” on appeal from a final judgment. See Will, 546 U.S. at 349.
The majority nevertheless accepts appellate jurisdiction, see ante at 6, reversing the district court's interlocutory order and remanding with instructions to dismiss the plaintiffs' remaining claims as preempted on the same theory underlying the D.C. Circuit's decision in Saleh. Putting aside the jurisdictional defect for argument's sake, I take issue with the majority's embrace of Saleh preemption to relieve CACI of its potential liability in this matter.3
The majority purports merely to apply the Supreme Court's decision in Boyle v. United Technologies Corp., 487 U.S. 500 (1988), but by adopting the reasoning of Saleh v. Titan Corp., 580 F.3d 1 (D.C.Cir.2009), a case presenting facts highly similar to this one, it affords Boyle an excessively robust elasticity. The Boyle Court recognized a form of implicit preemption of state law, based on a “significant conflict” between “uniquely federal interests” and state law duties the plaintiff sought to impose on a private contractor. See 487 U.S. at 504, 506, 512.
The contract in Boyle was one for procurement in which the government contractor was to manufacture and deliver military helicopters with an outward-opening escape hatch. This hatch could not be opened underwater, which allegedly rendered the design defective under state law. To determine whether a significant conflict was present, the Court looked to the statutory “discretionary functions” exception to the Federal Tort Claims Act (the “FTCA”), which reserves the sovereign immunity of the United States for, among other things, “[a]ny claim ․ based upon the exercise or performance or the failure to exercise or perform a discretionary function or duty on the part of a federal agency or an employee of the Government, whether or not the discretion involved be abused.” 28 U.S.C. § 2680(a).
Guided by this specific FTCA exception, the Supreme Court reasoned that “the selection of the appropriate design for military equipment to be used by our Armed Forces is assuredly a discretionary function” under the FTCA because “[i]t often involves not merely engineering analysis but judgment as to the balancing of many technical, military, and even social considerations, including specifically the trade-off between greater safety and great combat effectiveness.” Boyle, 487 U.S. at 511. Accordingly, the Court concluded that “state law which holds Government contractors liable for design defects in military equipment does in some circumstances present a ‘significant conflict’ with federal policy and must be displaced.” Id. at 512. The Court acknowledged that the Boyle preemptive principle was distinct from “ordinary” preemption and was not tethered to “legislation specifically immunizing Government contractors from liability.” Id. at 504, 507.
The Supreme Court stated in no uncertain terms, however, that the presence of a federal interest “merely establishes a necessary, not a sufficient, condition for the displacement of state law.” 487 U.S. at 507. Such a “[d]isplacement will occur only where ․ a significant conflict exists between an identifiable federal policy or interest and the operation of state law, or the application of state law would frustrate specific objectives of federal legislation .” Id. (citations, internal quotation marks, and alterations omitted). Although “[t]he conflict with federal policy need not be as sharp as that which must exist for ordinary preemption ․, conflict there must be.” Id. at 507–08 (emphasis added).
The rather obvious problem with invoking the government's “interest in conducting and controlling the conduct of war,” ante at 11, to preempt the plaintiffs' claims of gratuitous torture by an independent contractor, is that there is no conflict between the two. No federal interest implicates the torture and abuse of detainees. To the contrary, the repeated declarations of our executives, echoed by the Congress, expressly disavow such practices.
For example, shortly after graphic photos depicting detainee abuse at Abu Ghraib became public, President Bush vowed that “the practices that took place in that prison are abhorrent and they don't represent America.” White House, Press Release, President Bush Meets with Al Arabiya Television, 2004 WLNR 2540883 (May 5, 2004). He pledged to “[t]he people of the Middle East ․ that we will investigate fully, that we will find out the truth,” and further assured that “justice will be served.” Id. Similarly, Secretary of Defense Rumsfeld testified before Congress that the Abu Ghraib prisoner abuses were “inconsistent with the values of our nation,” asserting that “[p]art of [our] mission—part of what we believe in—is making sure that when wrongdoing or scandal occur, that they are not covered up, but exposed, investigated, publicly disclosed—and the guilty brought to justice.” Donald H. Rumsfeld, Testimony Before the Senate and House Armed Services Committees 1, 6 (May 7, 2004).
For its part, the Senate “condemn[ed] in the strongest possible terms the despicable acts at Abu Ghraib prison.” S. Res. 356, 108th Cong. (2004). Meanwhile, the House of Representatives declared that the practices at Abu Ghraib “offen[d] ․ the principles and values of the American people and the United States military ․ and contradict the policies, orders, and laws of the United States military and undermine the ability of the United States military to achieve its mission in Iraq.” H.R. Res. 627, 108th Cong. (2004).
The point is not confined to the facile observation that no federal interest encompasses the torture and abuses that the plaintiffs allege. Indeed, it is quite plausible that the government would view private tort actions against the perpetrators of such abuses as advancing the federal interest in effective military activities. The government has not intervened on behalf of the contractors in this dispute, and, in fact, the Department of Defense (the “DOD”) has promulgated a final rule advising contractors that the “[i]nappropriate use of force could subject a contractor or its subcontractors or employees to prosecution or civil liability under the laws of the United States and the host nation.” Contractor Personnel Authorized to Accompany U.S. Armed Forces, 73 Fed.Reg. 16,764, 16,764, 16,767 (Mar. 31, 2008) (the “DOD Rule”).
The DOD Rule “may reflect the government's general view that permitting contractor liability will advance, not impede, U.S. foreign policy by demonstrating that ‘the United States is committed to ensuring that its contractors are subject to proper oversight and held accountable for their actions.’ “ Saleh, 580 F.3d at 28 (Garland, J., dissenting) (quoting U.S. Dep't of State, Press Release, Department of State Legal Adviser Promotes Accountability for Private Military and Security Companies (Sept. 17, 2008)). As the Saleh dissent emphasizes:
the government's failure to defend the contractors may reflect the Executive Branch's view that the country's interests are better served by demonstrating that “people will be held to account according to our laws.” And the Executive may believe that one way to show that “people will be held to account” is to permit this country's legal system to take its ordinary course and provide a remedy for those who were wrongfully injured.
At bottom, Boyle does not countenance the majority's approach because there simply is no conflict—much less, a “significant conflict”—between the asserted state law duties and any uniquely federal interest. Quite the opposite: the plaintiffs allege that CACI violated federal policy. Boyle does not apply, because, as the Saleh dissent explained:
Boyle has never been applied to protect a contractor from liability resulting from the contractor's violation of federal law and policy. And there is no dispute that the conduct alleged, if true, violated both. Hence, these cases are not within the area where the policy of the “discretionary function” would be frustrated, and they present no significant conflict with federal interests. Preemption is therefore not justified under Boyle.
Saleh, 580 F.3d at 23 (Garland, J., dissenting) (internal quotation marks and citations omitted).
Another premise underlying Boyle's reasoning—the rigid control that the government exerts over contractors in procuring military equipment—is absent where, as here, the government contracted for general services only. As the Boyle Court acknowledged, selecting military equipment “often involves not merely engineering analysis but judgment as to the balancing of many technical, military, and even social considerations.” 487 U.S. at 511. Ultimately, the government rather than the contractor must be in charge of decisionmaking in order for the contractor to be shielded from liability. Consistently with that principle, the Boyle test for preemption “assure[s] that the suit is within the area where the policy of the ‘discretionary function’ would be frustrated”—that is, “that the design feature in question was considered by a Government officer, and not merely by the contractor itself.” Id.4
By contrast, the government itself has recognized that such judgments are not present in general services contracts. As the DOD explained in a recent rulemaking, “[t]he public policy rationale behind Boyle does not apply when a performance-based statement of work is used in a services contract, because the Government does not, in fact, exercise specific control over the actions and decisions of the contractor or its employees or subcontractors.” DOD Rule, 73 Fed.Reg. at 16,768. In other words, the government's precise control over its contractor, which was so integral to Boyle's reasoning, see 487 U.S. at 509–12, is absent in a general services contract in which the government simply requires “a contractor to ensure its employees comply with host nation law and other authorities,” DOD Rule, 73 Fed.Reg. at 16,768.
It follows that while military contractors might be able to assert Boyle-type arguments when the government's decisions result in injuries to third parties, the DOD adamantly opposes “send[ing] a signal that would invite courts to shift the risk of loss to innocent third parties” where “contractors ․ seek[ ] to avoid accountability to third parties for their own actions by raising defenses based on the sovereignty of the United States.” DOD Rule, 73 Fed.Reg. at 16,768 (emphasis added). Accordingly, the DOD elected to “retain[ ] the current rule of law, holding contractors accountable for the negligent or willful actions of their employees, officers, and subcontractors.” Id. In obstinate opposition to the government's prescribed path, the majority would protect contractors from civil liability even when there is no indication that the government authorized the conduct underlying the asserted liability.
Contrary to the majority's position, whether the government authorized CACI's conduct in this case can only be ascertained by examining the contract between the parties, which, as the district court lamented, is not in the record at the dismissal stage. The contract would shed light on:
• The contractor's delegated discretionary authority—that is, the services the contractor was to provide under the contract—and whether the contractor acted within the bounds of such authority, see Rodriguez v. Lockheed Martin Corp., 627 F.3d 1259, 1266 (9th Cir.2010) (refusing to deem claim preempted under Boyle where “there is no proof to establish as a matter of law that the equipment [alleged to have injured the victims] conformed to the government's precise specifications”);
• Whether such authority was “validly conferred” to the contractor, see Boyle, 487 U.S. at 506 (quoting Yearsley, 309 U.S. at 20–21); and
• Whether and to what extent the government had a significant interest in the specific services to be provided, see id. at 509 (recognizing that “significant conflict” justifying preemption may not be present even where state duty is “precisely contrary” to contractual duty, since government may lack “significant interest in th[e] particular feature” specified in contract).
The majority's extra-contractual inquiry into whether “a civilian contractor is integrated into wartime combatant activities over which the military broadly retains command authority,” ante at 11–12 (citing Saleh, 580 F.3d at 9), is of scant moment considering the lack of agency possessed by the rank-and-file military to alter or augment the material terms of the contract.5
Of course, there is no evidence to support the majority's supposition of “integration” (whatever that means) in this case, other than what can be gleaned from the bare allegations of the Complaint. But the question is wholly irrelevant absent any allegation that the terms of the written agreement were materially supplemented or changed (or even could be, in the event that the contract contained a valid provision barring parol alterations), either by representatives with authority to act or through the parties' course of conduct or dealing. Here, although the plaintiffs allege a conspiracy with members of the military, they are entitled to the inference that the conspiracy did not define the contract, but instead permitted CACI to act outside its bounds. Cf. ante at 5 (“While some of the abuses that the plaintiffs detailed in the allegations of the complaint appear to have been approved by the military at one point or another, others were clearly not.”).6
By relying on the discretionary function exception to the Federal Tort Claims Act (“FTCA”) to identify the pertinent federal interest, the Supreme Court in Boyle required, at a minimum, that reviewing courts would examine a contractor's allegedly tortious conduct to determine whether it was truly the product of the government's exercise of discretion, or merely an ordinary, unprovoked lapse of care. The majority's approach avoids even that minimal analysis by grounding the asserted federal interest in a different exception to the FTCA—the combatant activities exception—the umbrella of which the majority would deploy over government contractors whenever there are “actions taken in connection with U.S. military operations overseas.” Ante at 8.7
The majority thereby ignores the Supreme Court's warning that the FTCA's exceptions are not equally equipped to define the contours of an implicit preemption. The Boyle Court made the point through its discussion of Feres v. United States, 340 U.S. 135 (1950), in which it was held that the FTCA does not waive sovereign immunity with respect to suits brought against the United States by service members for injuries sustained in the course of their military service.
The Supreme Court declared the Feres doctrine unsuitable to ascertain whether a significant conflict exists between federal interests and an asserted state duty, in that it “logically produces results that are in some respects too broad and in some respects too narrow.” Boyle, 487 U.S. at 510. As an example of the former, the Court observed that “[s]ince Feres prohibits all service-related tort claims against the Government, a contractor defense that rests upon it should prohibit all service-related tort claims against the manufacturer,” id., a result that the Supreme Court deemed inadvisable. See also Al–Quraishi v. Nakhla, 728 F.Supp.2d 702, 740 (D.Md.2010) (declining to adopt rationale of Saleh, based in part on Supreme Court's rejection of Feres as basis for preemption, “because [the Feres defense] does not take into account whether the Government exercised any discretion or played any role in the contractor's alleged tortious acts, as required by the three part test ultimately articulated in Boyle ”).
The majority's invocation of the combatant activities exception suffers from the same defects. While the Supreme Court sought to discern an appropriate “limiting principle” to assist in identifying any significant conflict between state and federal policies under the discretionary function exception, Boyle, 487 U.S. at 509, the majority's version of preemption under the combatant activities exception is “extraordinarily broad, ․ result[ing] not in conflict preemption but in field preemption.” Saleh, 580 F.3d at 23 (Garland, J., dissenting) (internal quotation marks omitted).8
The majority makes no attempt to conceal the sweeping breadth of the preemption doctrine it adopts today, confidently maintaining that its approach properly implements what it characterizes as “the FTCA's policy of eliminating tort concepts from the battlefield.” Ante at 11 (quoting Saleh, 580 F.3d at 7). The majority vastly overstates its case, however, because, much more narrowly,
the FTCA's policy is to eliminate the U.S. government's liability for battlefield torts. That, after all, is what the FTCA says. But it is not plain that the FTCA's policy is to eliminate liability when the alleged tortfeasor is a contractor rather than a soldier. That, after all, is not what the FTCA says.
Saleh, 580 F.3d at 26 (Garland, J., dissenting). Judge Garland's eye is keen: the FTCA waives, with certain specific exceptions, the sovereign immunity constitutionally afforded the United States, which operates through its various federal agencies. See 28 U.S.C. §§ 2674, 2675. Government contractors, however, are expressly excluded from the FTCA's reach. See id. § 2671 (“[T]he term ‘Federal agency’ ․ does not include any contractor with the United States.”). The majority's description of the FTCA's policy as the wholesale elimination of wartime torts, even those committed by private parties, is therefore inaccurate.
Congress has had no difficulty exempting private parties from liability in other contexts. Consider, for example, the statute found at 22 U.S.C. § 2291–4(b), which provides that the interdiction of an aircraft over a foreign country, conducted pursuant to a presidentially approved program, “shall not give rise to any civil action ․ against the United States or its employees or agents.” Id. (emphasis added). Congress has issued no similar exemption here. If anything, its wholesale exclusion of government contractors from the limited protections of the FTCA leads to the opposite conclusion—that CACI should be held liable for its civil misdeeds.
Further, the FTCA addresses only the immunity of the United States; it does not shield members of the armed services or other government employees from tort suits. Instead, the Westfall Act provides that sort of protection, so long as the Attorney General certifies “that the defendant employee was acting within the scope of his office or employment.” 28 U.S.C. § 2679(d)(1). Upon such certification, the employee is dismissed from the lawsuit and the United States is substituted as the party defendant, after which the dispute is governed by the FTCA (as well as its exceptions that retain sovereign immunity). See Osborn v. Haley, 549 U.S. 225, 230 (2007). But because the Westfall Act incorporates the FTCA's definitions, it too excludes government contractors. Yet the majority deems the plaintiffs' claims preempted in the absence of an Attorney General's certification that would have been essential were these defendants soldiers or sailors rather than contractors. The majority thereby grants the defendants unqualified protection that even our citizens in uniform do not enjoy.
The majority also gleans several specific policy conflicts that tort suits against contractors would bring about, but these concerns evaporate upon closer inspection. The majority asserts that “[n]ot only would potential tort liability against ․ contractors affect military costs and efficiencies and contractors' availability,” but “would also present the possibility that military commanders could be hauled into civilian courts for the purpose of evaluating and differentiating between military and contractor decisions.” Ante at 8. But the possibility of cost-passing is already taken into consideration at an earlier stage of the Boyle inquiry, that is, in determining whether a uniquely federal interest “will be directly affected.” 487 U.S. at 507.9
With respect to the majority's concern that military commanders may be called to provide testimony in private tort suits, wholesale preemption remains unwarranted. Ordinary mechanisms of civil procedure and other legal doctrines provide ample safeguards against such interference. Federal Rule of Civil Procedure 45, for example, compels the district courts to quash subpoenas calling for privileged matter or that would cause an undue burden. Moreover, the government remains free to invoke the state secrets doctrine. All this is to say, “[t]o deny preemption is not to grant plaintiffs free reign.” Saleh, 580 F.3d at 29 (Garland, J., dissenting).10
The majority expresses its fear that lawsuits will “undermine the flexibility that military necessity requires in determining the methods for gathering intelligence.” Ante at 8. Such a concern also proves illusory. The plaintiffs allege that the contractor personnel acted contrary to military directives and law. The asserted basis of liability, then, is not one that would hamper the flexibility the military needs in determining how to gather intelligence, but rather one that would hold contractors to account for violating the bounds already set by the military.
Because the majority erroneously strains to discover a new form of preemption unjustified by Supreme Court precedent, and, more fundamentally, because we lack jurisdiction to announce this new rule, I respectfully dissent.
1. Refusing to accept Saleh as the only other case squarely on point, the dissent chooses to rely heavily on the dissenting opinion in that case and would have us create a circuit split. Post, at 31, 32, 35 n. 5, 37, 38 & n. 8, 39, 40 & n. 10.
1. The Complaint is found at J.A. 16–41. (Citations herein to “J . A. _” refer to the contents of the Joint Appendix filed by the parties to this appeal.)
2. Though it declined to dismiss the state law claims, the district court granted CACI's motion insofar as it pertained to federal claims asserted by the plaintiffs pursuant to the Alien Tort Statute, 28 U.S.C. § 1350. See Al Shimari, 657 F.Supp.2d at 726–28.
3. I address Saleh preemption on the merits because there is much in the majority's provocative analysis of the issue that should not be left unanswered. Inasmuch as CACI's derivative sovereign immunity and political question defenses are not addressed in the majority opinion, but discussed only in Judge Niemeyer's separate, nonprecedential opinion, I believe it would be unhelpful and confusing to debate them here. Left to my own devices, I would not resolve any of CACI's arguments on the merits as we lack jurisdiction to consider them. In Taylor v. Kellogg, –––F.3d ––––, No. 10–1543 (4th Cir. Sept. 21, 2011), also decided today, I authored the opinion of the Court in which, as Judge Niemeyer points out, ante at 15, we affirmed the district court's judgment on the ground that the dispute in that case presented a nonjusticiable political question. Our jurisdiction in Taylor was unquestioned, however, in that the appeal was taken from the district court's indisputably final decision dismissing the plaintiff's case. See 28 U.S.C. § 1291.
4. More recently, the Supreme Court has reiterated the narrow scope of the Boyle preemption defense, as well as its grounding in a contractor's compliance with government instructions. For example, the Court has referred to Boyle as presenting a “special circumstance” in which “the government has directed a contractor to do the very thing that is the subject of the claim.” Correctional Servs. Corp. v. Malesko, 534 U.S. 61, 74 n. 6 (2001). As the Fifth Circuit recently explained, “[t]he government contractor defense in Boyle, stripped to its essentials, is fundamentally a claim that the Government made me do it.” Katrina Canal Breaches Litig. Steering Comm. v. Wash. Group Int'l, Inc., 620 F.3d 455, 465 (5th Cir.2010).
5. The Army Field Manual provides that “[c]ommanders do not have direct control over contractors or their employees ․; only contractors manage, supervise, and give directions to their employees.” U.S. Dep't of the Army, Field Manual 3–100.21, Contractors on the Battlefield § 1–22 (2003). In turn, the contractors must adhere to their contractual obligations without regard to the chain of command. As the Field Manual emphasizes, “the terms and conditions of the contract establish the relationship between the military (U.S.Government) and the contractor․ Only the contractor can directly supervise its employees. The military chain of command exercises management control through the contract.” Id. at 3–100.21, § 1–25. As such, the government has “no more control than any contracting party has over its counterparty. And that—without more—is not enough to make the conduct of a contractor ‘the combatant activities of the military or naval forces.’ “ Saleh, 580 F.3d at 34 (Garland, J., dissenting) (quoting 28 U.S.C. § 2680(j)).
6. The majority seizes upon the plaintiffs' allegation of a conspiracy between CACI and military personnel, see ante at 9–10, in support of its irrelevant supposition that CACI employees were integrated into the mission at Abu Ghraib. Whatever the military “mission” was at Abu Ghraib, it did not include torturing the plaintiffs. In any event, regardless of the relationship between the soldiers and civilians at the prison, the duties of the latter were defined exclusively by CACI's contract with the government. We do not know whether governmental authority to amend the contract resided at the Pentagon or elsewhere, but we may be fairly certain that such authority did not reside at Abu Ghraib. That relatively low-level military personnel may have violated their orders and encouraged their civilian counterparts to act outside the bounds of the contract—and settled legal principles—in no way translates to a conclusion that CACI should summarily escape liability on the ground that the actions imputed to it were somehow consistent with the government's interests.
7. By enacting the combatant activities exception to the FTCA, Congress expressly reserved the sovereign immunity of the United States with respect to “[a]ny claim arising out of the combatant activities of the military or naval forces, or the Coast Guard, during time of war.” 28 U.S.C. § 2680(j).
8. Inasmuch as the FTCA contains other potentially applicable exceptions—for “[a]ny claim arising in a foreign country,” and for “[a]ny claim arising out of assault [and] battery” regardless of where it occurs, 28 U.S.C. §§ 2680(h), (k)—it is baffling that the majority can correctly identify the combatant activities exception as the one that decrees the relevant federal policy. This is particularly so absent any meaningful discussion by the majority of what constitutes a “combatant activity,” whether such activities may take place domestically, or how they may be distinguished from an ordinary assault or battery. The difficulties in identifying the relevant FTCA exception makes it almost impossible to articulate why the one for combatant activities matters at all. As Judge Garland observes, “the ‘degree of integration’ test ․ seems wholly beside the point” once these other exceptions are considered. Saleh, 580 F.3d at 23 (Garland, J., dissenting). Inevitably, “[o]nce we depart from the limiting principle of Boyle, it is hard to tell where to draw the line.” Id.
9. In Richardson v. McKnight, 521 U.S. 399 (1997), the Supreme Court declined to extend qualified immunity to privately employed prison guards in an action under 42 U.S.C. § 1983. The Court reasoned that, because contractors performing service contracts are subject to “competitive market pressures,” the threat of tort liability encourages them to comply with contractual obligations to screen, train, and supervise their employees, so as to promote effectiveness while preventing and deterring contractors and their employees from taking unlawful actions. See Richardson, 521 U.S. at 409 (“Competitive pressures mean not only that a firm whose guards are too aggressive will face damages that raise costs, thereby threatening its replacement, but also that a firm whose guards are too timid will face threats of replacement by other firms with records that demonstrate their ability to do both a safer and a more effective job.”). As in the Richardson litigation, the potential for tort liability and competition between contractors may well facilitate the government's selection of contractors who will perform in a more effective, lawful, and inexpensive manner.
10. Moreover, the majority's approach brings about the very problems it seeks to avert. That is, if the courts “ignore the military's own description of its chain of command” by looking to the “degree of integration that, in fact, existed between the military and [contractor] employees,” then they thereby “invite the wide-ranging judicial inquiry—with affidavits, depositions, and conflicting testimony—that the court rightly abjures.” Saleh, 580 F.3d at 34 (Garland, J., dissenting).
Reversed and remanded with instructions by published opinion. Judge NIEMEYER wrote the opinion, in which Judge SHEDD joined. Judge NIEMEYER wrote a separate opinion giving additional reasons for reversing and remanding. Judge KING wrote a dissenting opinion.