Donald Vance v. Donald Rumsfeld

EASTERBROOK, Chief Judge. This appeal presents the question whether the federal judiciary should create a right of action for damages against soldiers (and others in the chain of command) who abusively interrogate or mistreat military prisoners, or fail to prevent improper detention and interrogation. Both other courts of appeals that have resolved this question have given a negative answer. Lebron v.

Rumsfeld, 670 F.3d 540 (4th Cir.2012); Doe v. Rumsfeld, 683 F.3d 390 (D.C.Cir.2012); Ali v. Rumsfeld, 649 F.3d 762 (D.C.Cir.2011). Another circuit declined to create a damages remedy against intelligence officials who turned a suspected terrorist over to another nation for interrogation.

Arar v. Ashcroft, 585 F.3d 559, 571-81 (2d Cir.2009) (en banc). We agree with those decisions. I In 2005 and 2006 Donald Vance and Nathan Ertel worked in Iraq for Shield Group Security (later known as National Shield Security), a private firm that provided protective services to businesses and governmental organizations. (This factual narration comes from the complaint, whose allegations we must accept for current purposes.) Vance came to suspect that Shield was supplying weapons to groups opposed to the United States.

He reported his observations to the FBI. Ertel furnished some of the information that Vance relayed. Persons who Vance and Ertel suspected of gun-running retaliated by accusing Vance and Ertel of being arms dealers themselves. Military personnel arrested them in mid-April 2006. (The complaint does not specify which day the arrests occurred.)

According to the complaint, plaintiffs were held in solitary confinement and denied access to counsel. Their interrogators used “threats of violence and actual violence, sleep deprivation and alteration, extremes of temperature, extremes of sound, light manipulation, threats of indefinite detention, denial of food, denial of water, denial of needed medical care, yelling, prolonged solitary confinement, incommunicado detention, falsified allegations and other psychologically-disruptive and injurious techniques.” Vance and Er-tel were provisionally classified as “security internees” and called before a Detainee Status Board, but they were not allowed to present evidence — and the military officials running the proceedings refused to look at files on their computers that Vance and Ertel say would have established their innocence of arms-dealing charges. Nor did the Board contact the FBI, even though Vance and Ertel said that agents would verify their story. The Board concluded on April 29, 2006, that Ertel should be released.

Nonetheless he was held for another 18 days, during which interrogators continued to use harsh techniques. He was released on May 17, 2006. Vance remained in solitary confinement until his release on July 20, 2006, and was subjected to sleep deprivation, prolonged exposure to cold, intolerably loud music, “hooding,” “walling” (placing a person’s heels against a wall and slamming his body backward into that wall), threats of violence, and other techniques that caused physical or mental pain. The Army Field Manual forbids several of these techniques, which it classifies as “physical torture,” “mental torture,” or “coercion.” See Army Field Manual: Intelligence Interrogation 1-8 (1992).

Whether any of the techniques constitutes “torture” within the meaning of 18 U.S.C. § 2340(1), which makes torture by interrogators a crime, is a subject on which the parties’ briefs do not join issue, and which we therefore do not address. The Detainee Status Board eventually concluded that both Vance and Ertel are innocent of the allegations that had been made against them. Neither was charged with a crime. In December 2006 Vance and Ertel filed this suit against persons who conducted or approved their detention and interrogation, and many others who had supervisory authority over those persons.

The defendants included Secretary of Defense Donald Rumsfeld. Plaintiffs alleged that Secretary Rumsfeld had authorized the use of harsh interrogation methods in Iraq and contended that he is personally liable in damages — even though plaintiffs also alleged that they had never been accused of being enemy combatants and therefore were not within the scope of Secretary Rumsfeld’s authorization. They also sued the United States, seeking the return of all property that had been seized from them in Iraq. Rumsfeld asked the district court to dismiss the complaint, presenting three principal arguments: that federal law does not establish an action for damages on account of abusive military interrogation; that the complaint does not plausibly allege his personal involvement in plaintiffs’ detention and interrogation; and that he is entitled to qualified immunity.

The district court ruled against all of these contentions. 694 F.Supp.2d 957 (N.D.Ill.2010). Rumsfeld has appealed under the doctrine of Mitchell v. Forsyth, 472 U.S. 511, 105 S.Ct. 2806, 86 L.Ed.2d 411 (1985), which treats the rejection of an immunity defense as a final decision for the purpose of 28 U.S.C. § 1291. The United States also moved to dismiss the complaint, contending that the “military authority exception” to the Administrative Procedure Act, 5 U.S.C. § 701(b)(1)(G), bars the suit against it.

Section 701(b)(1)(G) prohibits judicial review of “military authority exercised in the field in time of war or in occupied territory”. The district court concluded that this language does not apply — at least, does not prevent Vance and Ertel from engaging in discovery that they contend would show the statute’s inapplicability — and denied the motion to dismiss. 2009 WL 2252258, 2009 U.S. Dist. Lexis 67349 (N.D.Ill.

July 29, 2009). The district court later certified this order for interlocutory appeal under 28 U.S.C. § 1292(b), see 2010 WL 2136657, 2010 U.S. Dist. Lexis 51973 (N.D.Ill.

May 26, 2010), and a motions panel accepted the appeal. A merits panel reversed the district court’s decision with respect to the United States but affirmed with respect to Rumsfeld’s claim of immunity. 653 F.3d 591 (7th Cir.2011). We granted Rumsfeld’s request for rehearing en banc and vacated the panel’s opinion and judgment; this set aside both aspects of its decision. II Both the district court and the panel concluded that it is appropriate to create a private right of action for damages against persons in the military chain of command. See generally Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971).

The lead argument in former Secretary Rumsfeld’s brief contests this conclusion. Because the basis of appellate jurisdiction is the district court’s rejection of an immunity defense, however, we must consider whether we are authorized to address the merits. The answer is yes. The Supreme Court held in Siegert v.

Gilley, 500 U.S. 226, 232, 111 S.Ct. 1789, 114 L.Ed.2d 277 (1991), that when evaluating an argument that a right is not “clearly established” — the essential ingredient in any invocation of qualified immunity — a court may conclude that the right has not been “clearly” established because it has not been established at all. The Court followed up in Saucier v. Katz, 533 U.S. 194, 121 S.Ct. 2151, 150 L.Ed.2d 272 (2001), by holding that a court of appeals must decide both whether the right in question exists and whether its existence had been “clearly established” before the time of the challenged acts. Pearson v.

Callahan, 555 U.S. 223, 129 S.Ct. 808, 172 L.Ed.2d 565 (2009), overruled that portion of Saucier and held that a court of appeals may use sound discretion when deciding whether to reach the merits ahead (or instead) of the immunity question. But the Court did not doubt that, on an interlocutory appeal under Mitchell, one potential ground of decision is a conclusion that the plaintiff does not have a legally sound claim for relief. Wilkie v. Robbins, 551 U.S. 537, 548-50, 127 S.Ct. 2588, 168 L.Ed.2d 389 (2007), applies this approach to Bivens claims in particular.

Robbins sued some federal officials, asserting extra-statutory claims for damages and contending that reasoning along the lines of Bivens allowed the federal judiciary to recognize such a remedy. Defendants took an interlocutory appeal, contending that they enjoyed qualified immunity. The Supreme Court ruled in defendants’ favor — not because of immunity, but because it concluded that it should not create a new Bivens remedy. Similarly, in Ashcroft v. Iqbal, 556 U.S. 662, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009), the Supreme Court resolved a qualified-immunity appeal by deciding that the complaint did not state a plausible claim on the facts.

We have jurisdiction to decide this case on the same grounds the Supreme Court employed in Wilkie and Iqbal. See also Levin v. Madigan, 692 F.3d 607, 610-11 (7th Cir.2012). The appeal by the United States does not present any jurisdictional problem, given the court’s decision to accept the appeal certified under § 1292(b).

Neither does it present a difficult question. The panel held that § 701(b)(1)(G) prevents any relief against the United States. 653 F.3d at 626-27. We agree with that conclusion, for the reasons the panel gave. Further discussion of the subject is unnecessary.

Ill When considering whether to create an extra-statutory right of action for damages against military personnel who mistreat detainees, we assume that at least some of the conditions to which plaintiffs were subjected violated their rights. Although the Constitution’s application to interrogation outside the United States is not settled, see United States v. Verdugo-Urquidez, 494 U.S. 259, 268-69, 110 S.Ct. 1056, 108 L.Ed.2d 222 (1990), Rumsfeld concedes (for current purposes at least) that it governs. The conduct alleged in the complaint appears to violate the Detainee Treatment Act, 10 U.S.C. § 801 note and 42 U.S.C. §§ 2000dd to 2000dd-l, and may violate one or more treaties.

The source of the substantive right does not matter for the analysis that follows. Unless there is a right of action against soldiers and their immediate commanders, however, there cannot be a right of action for damages against remote superiors such as former Secretary Rumsfeld. And neither the Detainee Treatment Act nor any other statute creates a private right of action for damages under the circumstances narrated by plaintiffs’ complaint. This much, at least, is common ground among the parties.

Plaintiffs therefore ask us to create a right of action under federal common law. Bivens was the first time the Supreme Court created a non-statutory right of action for damages against federal employees. Since then the Court has created two others: for unconstitutional discrimination in public employment, see Davis v. Passman, 442 U.S. 228, 99 S.Ct. 2264, 60 L.Ed.2d 846 (1979), and for violations of the eighth amendment by prison guards, see Carlson v.

Green, 446 U.S. 14, 100 S.Ct. 1468, 64 L.Ed.2d 15 (1980). It has not created another during the last 32 years — though it has reversed more than a dozen appellate decisions that had created new actions for damages. Whatever presumption in favor of a Bivens-like remedy may once have existed has long since been abrogated. The Supreme Court has never created or even favorably mentioned the possibility of a non-statutory right of action for damages against military personnel, and it has twice held that it would be inappropriate to create such a claim for damages.

See Chappell v. Wallace, 462 U.S. 296, 103 S.Ct. 2362, 76 L.Ed.2d 586 (1983); United States v. Stanley, 483 U.S. 669, 107 S.Ct. 3054, 97 L.Ed.2d 550 (1987). The Court has never created or even favorably mentioned a non-statutory right of action for damages on account of conduct that occurred outside the borders of the United States.

Yet plaintiffs propose a novel damages remedy against military personnel who acted in a foreign nation— and in a combat zone, no less. The Court’s most recent decision declining to extend Bivens is Minneci v. Pollard, — U.S. —, 132 S.Ct. 617, 181 L.Ed.2d 606 (2012). Minneci treated Wilkie as a restatement of the governing principles, 132 S.Ct. at 621.

Wilkie tells us: our consideration of a Bivens request follows a familiar sequence, and on the assumption that a constitutionally recognized interest is adversely affected by the actions of federal employees, the decision whether to recognize a Bivens remedy may require two steps. In the first place, there is the question whether any alternative, existing process for protecting the interest amounts to a convincing reason for the Judicial Branch to refrain from providing a new and freestanding remedy in damages. [Bush v. Lucas, 462 U.S. 367] at 378 [103 S.Ct. 2404, 76 L.Ed.2d 648 (1983) ]. But even in the absence of an alternative, a Bivens remedy is a subject of judgment: “the federal courts must make the kind of remedial determination that is appropriate for a common-law tribunal, paying particular heed, however, to any special factors counselling hesitation before authorizing a new kind of federal litigation.” Bush, supra, at 378 [103 S.Ct. 2404]. 551 U.S. at 550, 127 S.Ct. 2588.

Congress has provided some opportunities-for compensation of persons injured by the military in combat zones. Rumsfeld does not contend that these statutes (which we discuss later) supply a “convincing reason for the Judicial Branch to refrain from creating a new and freestanding remedy in damages.” But he does contend that many factors make it inappropriate for the judiciary to create a common-law remedy for damages arising from military operations in a foreign nation. Chappell and Stanley hold that it is inappropriate for the judiciary to create a right of action that would permit a soldier to collect damages from a superior officer. Plaintiffs say that these decisions are irrelevant because they were not soldiers.

That is not so clear. They were security contractors in a war zone, performing much the same role as soldiers. Some laws treat employees of military contractors in combat zones the same as soldiers. See, e.g., 18 U.S.C. § 3261 and § 3267(1) (A) (iii), parts of the Military Extraterritorial Jurisdiction Act discussed in United States v.

Brehm, 691 F.3d 547 (4th Cir.2012). See also United States v. Ali, 71 M.J. 256 (C.A.A.F.2012) (holding that a civilian employee of a security contractor in Iraq is treated as a soldier for the purpose of prosecution under the Uniform Code of Military Justice). But we need not decide whether civilians doing security work in combat zones are soldiers by another name, because Chappell and Stanley did not entirely depend on the relation between the soldier and the superior officer.

The Supreme Court’s principal point was that civilian courts should not interfere with the military chain of command — not, that is, without statutory authority. Chappell observed that military efficiency depends on a particular command structure, which civilian judges could mess up without appreciating what they were doing. 462 U.S. at 300, 103 S.Ct. 2362. The Court observed that Congress has ample authority, under its constitutional power to “make Rules for the Government and Regulation of the land and naval Forces” (Art. I § 8 cl. 14), to provide for awards of damages and other kinds of judicial review of military decisions. When Congress does not exercise that power — or when, as we explain in a moment, it exercises that power without providing for damages against military wrongdoers — the judiciary should leave the command structure alone. “Matters intimately related to ... national security are rarely proper subjects for judicial intervention.” Haig v.

Agee, 453 U.S. 280, 292, 101 S.Ct. 2766, 69 L.Ed.2d 640 (1981). Stanley tried to circumvent Chappell by suing some civilians and contending that the officers he had named were not his superiors but had been in a different branch of the military hierarchy. Stanley also observed that the plaintiff in Chappell had at least some monetary remedy through legislation, while he had none. The Court wrote in response: “The ‘special facto[r]’ that ‘counsels] hesitation’ [in creating a common-law remedy] is not the fact that Congress has chosen to afford some manner of relief in the particular case, but the fact that congressionally uninvited intrusion into military affairs by the judiciary is inappropriate.” 483 U.S. at 683, 107 S.Ct. 3054.

That’s equally true of our plaintiffs’ situation. The fourth circuit addressed this subject in detail in Lebron, 670 F.3d at 548-52, and we agree with its evaluation. What plaintiffs want is an award of damages premised on a view that the military command structure should be different— that, for example, the Secretary of Defense must do more (or do something different) to control misconduct by interrogators and other personnel on the scene in foreign nations. They want a judicial order that would make the Secretary of Defense care less about the Secretary’s view of the best military policy, and more about the Secretary’s regard for his own finances.

Plaintiffs believe that giving the Secretary of Defense a financial stake in the conduct of interrogators would lead the Secretary to hold the rights of detainees in higher regard — which surely is true, but that change would come at an uncertain cost in national security. If the judiciary never erred, damages awards against soldiers and their civilian supervisors would be all gain and no loss. But judges make mistakes: They may lack vital knowledge, may accept claims that should be rejected on the facts or the law, or may award excessive damages on justified claims or create supervisory liability when they shouldn’t. See Stanley, 483 U.S. at 682-83, 107 S.Ct. 3054; see also Ashcroft v. al-Kidd, — U.S. —, 131 S.Ct. 2074, 2087, 179 L.Ed.2d 1149 (2011) (Kennedy, J., concurring).

Accounting for human fallibility is an important part of the design of a legal system. Military prosecutors (or civilian prosecutors acting under the President’s direction) can consider the needs of effective military action when exercising prosecutorial discretion. Judges lack information that executive officials possess, and in civil litigation there is no source of discretion comparable to a prosecutor’s. The Justices concluded in Chappell and Stanley that Congress and the Commander-in-Chief (the President), rather than civilian judges, ought to make the essential tradeoffs, not only because the constitutional authority to do so rests with the political branches of government but also because that’s where the expertise lies.

That is as true here as it was in Chappell and Stanley. Accord, Doe, 683 F.3d at 394 (“Doe is a contractor and not an actual member of the military, but we see no way in which this affects the special factors analysis.”). The political branches have not been indifferent to detainees’ interests. To the contrary, the treatment of military detainees has occasioned extended debate and led to a series of statutes.

The Detainee Treatment Act is one. Others enacted or amended in the past decade include the Torture Victim Protection Act, 28 U.S.C. § 1350 note; the Military Claims Act, 10 U.S.C. § 2733; the Foreign Claims Act, 10 U.S.C. § 2734; the Military Commissions Act, 10 U.S.C. § 948a et seq.; the federal torture statute, 18 U.S.C. §§ 2340-2340A; the War Crimes Act, 18 U.S.C. § 2441; and the Uniform Code of Military Justice, 10 U.S.C. § 801 et seq. Lebrón summarizes the ways in which the political branches have addressed the appropriate design of policies about interrogation. 670 F.3d at 548-52. These statutes have one thing in common: none provides for damages against military personnel or their civilian superiors.

Some, such as the Detainee Treatment Act, expressly block damages liability. (We return to this shortly.) Others provide compensation to victims of military errors or misconduct, but the compensation comes from the public fisc rather than private pockets. For example, the Military Claims Act provides that the Judge Advocate General of each service may award up to $100,000 from the Treasury to any person injured by the military. The Foreign Claims Act provides that a claims commission may award up to $100,000 of public money to a person injured by the U.S. military in a foreign nation. (These options are mutually exclusive; when the Foreign Claims Act or the Federal Tort Claims Act applies, the Military Claims Act does not.

See 10 U.S.C. § 2733(b)(2).) We asked plaintiffs’ counsel at oral argument whether they had applied for awards under either statute. Counsel said no, telling us that $100,000 is too little for their injuries and that the persons charged with implementing these laws enjoy too much discretion for plaintiffs’ liking. (Plaintiffs have not argued that 32 C.F.R. § 536.45(h), which provides that the military will not make awards under either statute for assault and battery, would make these statutes useless to them. Section 536.45(h) allows awards for intentional torts related to an investigation; because the briefs do not discuss the effect of § 536.45(h), we do not consider whether plaintiffs’ losses would come within the “investigation” clause.)

We are willing to assume that the cap on awards, and the existence of discretion about when to award compensation (and how much to provide), means that these statutes are not full substitutes for a Bivens remedy. See Minneci, the Court’s most recent discussion of that subject. Still, the fact that Congress has provided for compensation tells us that it has considered how best to address the fact that the military can injure persons by improper conduct. We take two things from the Military Claims Act and the Foreign Claims Act: first, Congress has decided that compensation should come from the Treasury rather than from the pockets of federal employees; second, plaintiffs do not need a common-law damages remedy in order to achieve some recompense for wrongs done them.

Unlike Webster Bivens, they are not without recourse. Vance and Ertel maintain, however, that through the Detainee Treatment Act Congress has decided that they are entitled to damages from the Secretary of Defense and his subordinates. A portion of the Detainee Treatment Act codified at 42 U.S.C. § 2000dd-l(a) provides that in both civil suits and criminal prosecutions, military interrogators and their superiors are protected from liability if “such officer, employee, member of the Armed Forces, or other agent did not know that the practices were unlawful and a person of ordinary sense and understanding would not know the practices were unlawful. Good faith reliance on advice of counsel should be an important factor, among others, to consider in assessing whether a person of ordinary sense and understanding would have known the practices to be unlawful.” Of course a defense to damages liability does not create damages liability, but plaintiffs contend that § 2000dd-l(a) assumes that this liability already exists, so personal liability must have Congress’s blessing. That assumption is unwarranted.

Congress often legislates to make doubly sure that federal employees will not be personally liable. The Westfall Act, 28 U.S.C. § 2679, is an example of that strategy. (Gutierrez de Martinez v. Lamagno, 515 U.S. 417, 115 S.Ct. 2227, 132 L.Ed.2d 375 (1995), and Ali v. Rumsfeld, supra, discuss that law’s scope and effects.)

The Public Health Service Act, 42 U.S.C. § 233(a), is another. See Hui v. Castaneda, 559 U.S. 799, 130 S.Ct. 1845, 176 L.Ed.2d 703 (2010). Section 7(a) of the Military Commissions Act, 28 U.S.C. § 2241(e)(2), is a third.

It forbids awards of damages to aliens detained as enemy combatants. See Al-Zahrani v. Rodriguez, 669 F.3d 315 (D.C.Cir.2012). The existence of safeguards against personal liability does not imply legislative authorization for the judiciary to create personal liability.

Section 2000dd-l(a) applies only to suits by aliens and therefore does not affect suits by citizens such as plaintiffs. Plaintiffs treat the restricted coverage of § 2000dd-l as a glitch, but we think it is more likely that the coverage reflects an assumption behind the statute. Aliens detained by U.S. military personnel might invoke multiple sources of authorization to award damages: one is the Torture Victim Protection Act; a second is the Alien Tort Act, 28 U.S.C. § 1350; and the third is the law of the nation in which the detention occurred (here, the law of Iraq). Congress may have wanted to make sure that military personnel enjoy some protection against suits by persons who have an express right of action.

Vance and Ertel cannot use (at least, have not tried to use) the Torture Victim Protection Act, the Alien Tort Act, or the law of Iraq as a basis for the remedy they seek. That Congress has put an obstacle in the way of persons who could use those bodies of law does not imply that persons who cannot use them must have a common-law damages remedy. The Detainee Treatment Act can be— and has been — enforced by criminal prosecutions. The Department of Defense has procedures for reporting claims of abuse; these procedures require all reports to be investigated and require prosecution to follow substantiated reports.

See Army Regulation 190-8 at §§ 1-5, 3-16, 6-9; DoD Directives 5100.77, 2311.01E. Failure by military personnel to follow these procedures is a court-martial offense. 10 U.S.C. § 892. Abusive interrogation in Iraq and Afghanistan has led to courts-martial. Injunctions that enforce the Detainee Treatment Act prospectively may be possible under the doctrine of Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908), or the waiver of sovereign immunity in 5 U.S.C. § 702.

But Congress has not authorized awards of damages against soldiers and them superiors, and creating a right of action in common-law fashion would intrude inappropriately into the military command structure. A Bivens-like remedy could cause other problems, including diverting Cabinet officers’ time from management of public affairs to the defense of their bank accounts. See Doe, 683 F.3d at 396. Then there are problems with evidence. See Lebron, 670 F.3d at 555-56.

When the state-secrets privilege did not block the claim, a court would find it challenging to prevent the disclosure of secret information. Anyone, whether or not a bona fide victim of military misconduct, could sue and then use graymail (the threat of disclosing secrets) to extract an undeserved settlement. See Arar, 585 F.3d at 578-81. That’s not a problem under the Military Claims Act and the Foreign Claims Act, which allow proceedings to be conducted in confidence.

The panel distinguished Arar and Ali v. Rumsfeld on the ground that those plaintiffs were aliens (Arar, for example, is a citizen of Canada). 653 F.3d at 620-22. More recent decisions, including Lebrón and Doe, dealt with (and rejected) Bivenslike claims by U.S. citizens. We do not think that the plaintiffs’ citizenship is dis-positive one way or the other.

See Doe, 683 F.3d at 396. Wallace and Stanley also were U.S. citizens. The Supreme Court has never suggested that citizenship matters to a claim under Bivens. It would be offensive to our allies, and it should be offensive to our own principles of equal treatment, to declare that this nation systematically favors U.S. citizens over Canadians, British, Iraqis, and our other allies when redressing injuries caused by our military or intelligence operations.

Treaties may pose a further obstacle to favoring U.S. citizens in the design of common-law remedies, but we need not decide, because the choice of remedies for military misconduct belongs to Congress and the President rather than the judicial branch. IV Even if we were to create a common-law damages remedy against military personnel and their civilian superiors, former Secretary Rumsfeld could not be held liable. He did not arrest plaintiffs, hold them incommunicado, refuse to speak with the FBI, subject them to loud noises, threaten them while they wore hoods, and so on. The most one could say about him—the most plaintiffs do say about him—is that (a) in 2002 and 2003 he authorized the use of harsh interrogation techniques when dealing with enemy combatants, (b) he received reports that his subordinates sometimes used these techniques, without authorization, on persons such as plaintiffs despite the Detainee Treatment Act of 2005, and (c) he did not do enough to bring interrogators under control. The Supreme Court held in Iqbal that liability under a Bivens-like remedy is personal. 556 U.S. at 676-77, 129 S.Ct. 1937.

Cabinet secretaries (in Iqbal the Attorney General) and other supervisory personnel are accountable for what they do, but they are not vicariously liable for what their subordinates do. The Court added that knowledge of a subordinate’s misconduct is not enough for liability. The supervisor can be liable only if he wants the unconstitutional or illegal conduct to occur. Id. at 677, 129 S.Ct. 1937.

Yet plaintiffs do not allege that Secretary Rumsfeld wanted them to be mistreated in Iraq. His orders concerning interrogation techniques concerned combatants and terrorists, not civilian contractors. What happened to plaintiffs violated both Rumsfeld’s directives of 2002 and 2003, and the Detainee Treatment Act of 2005. In an ideal world, the Secretary of Defense and the Army’s Chief of Staff would have achieved full compliance with the Detainee Treatment Act, but a public official’s inability to ensure that all subordinate federal employees follow the law has never justified personal liability.

The gist of plaintiffs’ claim against Rumsfeld is that harsh interrogation tactics were used erroneously, pointlessly, and excessively in their situation. Plaintiffs should be compensated, if their allegations are true—though it is too late for them to invoke the Foreign Claims Act, which has a two-year period of limitations. Just because it may be hard to use the statutory mechanisms of compensation, however, it does not follow that a Cabinet official must pay out of his own pocket. To see this, ignore for the moment the military and foreign-location issues and ask whether persons in the United States who are shot by federal agents or beaten by prison guards have a good claim against the Director of the FBI, the Director of the Bureau of Prisons, or the Attorney General.

They do not. Both Iqbal and alKidd, say that supervisors are not vicariously liable for their subordinates’ transgressions. The Director of the FBI allows field agents to carry guns and permits them to use deadly force. Yet if an agent shoots a fleeing suspect in the back, violating the fourth amendment, see Tennessee v.

Garner, 471 U.S. 1, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985), the Director is not liable just because the gun, issued under the Director’s policy, was a cause of the injury. Similarly for a police chief who establishes a K-9 squad, if a dog bites a bystander, or who authorizes search or arrest based on probable cause, if the police then search or arrest without probable cause. Plaintiffs’ theme is that Secretary Rumsfeld, having authorized harsh interrogation tactics for enemy combatants in 2002 and 2003, should have intervened after receiving reports that non-combatants were being subjected to these tactics and that interrogators had not properly implemented the Detainee Treatment Act of 2005. Yet the standard form of intervention would have been criminal prosecution (in the civilian courts or by court-martial).

The Department of Defense did prosecute some soldiers through courts-martial, and the Department of Justice filed some criminal prosecutions. Plaintiffs think that they should have done more, but no one can demand that someone else be prosecuted. See, e.g., Castle Rock v. Gonzales, 545 U.S. 748, 125 S.Ct. 2796, 162 L.Ed.2d 658 (2005); DeShaney v.

Winnebago County Dep’t of Social Services, 489 U.S. 189, 109 S.Ct. 998, 103 L.Ed.2d 249 (1989); Linda R.S. v. Richard D., 410 U.S. 614, 93 S.Ct. 1146, 35 L.Ed.2d 536 (1973). A court cannot say that, if there are too few prosecutions (or other enforcement), and thus too much crime, then the Attorney General or the Secretary of Defense is personally liable to victims of (preventable) crime. Yet that’s what plaintiffs’ approach entails. Iqbal held that knowledge of subordinates’ misconduct is not enough for liability.

The supervisor must want the forbidden outcome to occur. Deliberate indifference to a known risk is a form of intent. But Farmer v. Brennan, 511 U.S. 825, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994), holds that, to show scienter by the deliberate-indifference route, a plaintiff must demonstrate that the public official knew of risks with sufficient specificity to allow an inference that inaction is designed to produce or allow harm. A warden’s knowledge that violence occurs frequently in prison does not make the warden personally liable for all injuries.

See McGill v. Duckworth, 944 F.2d 344 (7th Cir.1991). Prisons are dangerous places, and misconduct by both prisoners and guards is common. Liability for wardens would be purely vicarious.

Farmer rejected a contention that wardens (or guards) can be liable just because they know that violence occurs in prisons and don’t do more to prevent it on an institution-wide basis. To get anywhere, Vance and Ertel would need to allege that Rumsfeld knew of a substantial risk to security contractors’ employees, and ignored that risk because he wanted plaintiffs (or similarly situated persons) to be harmed. The complaint does not contain such an allegation and could not plausibly do so. The head of any large bureaucracy receives reports of misconduct.

The Secretary of Defense has more than a million soldiers under his command. The Attorney General supervises thousands of FBI and DEA agents, thousands of prison guards, and so on. Many exceed their authority. People able to exert domination over others often abuse that power; it is a part of human nature that is very difficult to control.

See Philip Zimbardo, The Lucifer Effect: Understanding How Good People Turn Evil (2007). The head of an organization knows this, or should know it. Every police chief knows that some officers shoot unnecessarily or arrest some suspects without probable cause, and that others actually go over to the criminal side and protect drug rackets. But heads of organizations have never been held liable on the theory that they did not do enough to combat subordinates’ misconduct, and the Supreme Court made it clear in Iqbal that such theories of liability are unavailing.

Plaintiffs do not cite even one instance in which an Attorney General, a Director of the FBI, a Director of the Bureau of Prisons, or a municipal chief of police has been held personally liable for not ensuring that subordinates respect prisoners’ or suspects’ rights. Claims against the Secretary of Defense, who has more people under his command, and a longer chain of subordinates between him and the culpable soldiers, are weaker. Although Vance and Ertel contend that their injuries can be traced (remotely) to Secretary Rumsfeld’s policies of 2002 and 2003, as well as to the misconduct of personnel in Iraq, they do not contend that the policies authorized harsh interrogation of security detainees, as opposed to enemy combatants. It is therefore unnecessary to decide when, if ever, a Cabinet officer could be personally liable for damages caused by the proper application of an unlawful policy or regulation.

As we observed in Hammer v. Ashcroft, 570 F.3d 798, 800 (7th Cir.2009) (en banc), the normal means to handle defective policies and regulations is a suit under the Administrative Procedure Act or an equivalent statute, not an award of damages against the policy’s author. Accord, Arar, 585 F.3d at 572-73. No court has ever held the Administrator of the EPA personally liable for promulgating an invalid regulation, even if that regulation imposes billions of dollars in unjustified costs before being set aside.

Cf. Padilla v. Yoo, 678 F.3d 748 (9th Cir.2012) (Deputy Assistant Attorney General not personally liable for preparing an opinion concluding that Secretary Rumsfeld’s policies were valid). The extent to which untenable directives, policies, and regulations may support awards of damages can safely be postponed to another day. V Because we have held that a common-law right of action for damages should not be created — and that plaintiffs’ complaint would fail to state a claim against former Secretary Rumsfeld even if such a right of action were to be created — it is unnecessary to decide whether Rumsfeld violated plaintiffs’ clearly established rights.

The decisions of the district court are reversed.