This week, U.S. officials announced that they had killed al Qaeda’s second-in-command with a drone strike. The news came soon after the New York Times published the fullest account to date of the process by which the United States selects lower-profile targets for drone attacks in Pakistan and Yemen. The most startling revelation was that President Obama personally supervises the “nomination” of potential targets and gives final approval for killing them.

The lengthy Times story isn’t based on a leak—it is clearly news that the White House wants the world to know. The reporters interviewed three dozen current and former Obama advisors to assemble their picture of the target-selection process. In his new book, Kill or Capture: The War on Terror and the Soul of the Obama Presidency, Newsweek reporter Daniel Klaidman notes that Obama’s presidential campaign “is painting a portrait of a steely commander who pursues the enemy without flinching.” Three days after the drones article, the Times ran an equally detailed article, again with obvious White House consent, about U.S. cyberattacks against Iran, reporting that “Mr. Obama, according to participants in the many Situation Room meetings . . . was acutely aware that with every attack he was pushing the United States into new territory.” This image of the president firmly in command of the drone campaign is precisely what the White House wishes to convey in the run-up to the election.

So why did the president put his hand on the helm? The Times reports:

Aides say Mr. Obama has several reasons for becoming so immersed in lethal counterterrorism operations. A student of writings on war by Augustine and Thomas Aquinas, he believes that he should take moral responsibility for such actions.

This image of a president schooled in just war theory is remarkable. At least one Catholic Web site has poured scorn on “the wise, judicious philosopher-king consulting Aquinas and Augustine before sending a drone missile on a ‘signature strike’ on a group of picnickers in Yemen or farmers in Pakistan.” (Perhaps the sarcasm is deserved—there have indeed been catastrophic mistakes in targeting—but Abu Yahya al-Libi, the al Qaeda second-in-command, was no picnicker or farmer.)

We of course have no idea how serious a student of just war theory the president is, but there is no reason to suspect that his aides are making it up. That entitles us to ask what the president may have taken from these two Christian writers and, more important, whether their arguments in fact support the morality of the president’s actions. What we find is a messy mix of insights and errors, by the saints as well as the president. The central themes of just war theory are easy to grasp: that war is a proper subject of moral judgment and that no leader should duck responsibility for making these judgments. Obama seems to understand that much. But today’s debate about drones centers on more specific questions about targeting, civilian deaths, and who should make the crucial decisions. On these issues Augustine and Aquinas offer scant guidance. Furthermore, the Times reports that the CIA uses dishonest rules for counting civilian casualties. If Obama acquiesces to such deceits, all the just war theory in the world will make no difference.

The verdict on Obama turns on the morality of targeted killings themselves. In my view, they are no different in principle from other wartime killings, and they have to be judged by the same standards of necessity and proportionality applied to warfare in general: sometimes they are justified, sometimes not. There are no simple answers.

• • •

Augustine and Aquinas, like all just war theorists, believe that some wars and tactics are morally permissible but others are not. This is the first, and far and away the most important, point that Obama would have taken from just war theory. Realists, by contrast, agree with the English philosopher Thomas Hobbes that in war “the notions of right and wrong, justice and injustice have there no place” while “force, and fraud, are in war the two cardinal virtues.” (Leviathan, ch. 13.) Michael Walzer begins Just and Unjust Wars, the greatest twentieth-century book on just war, with a chapter titled “Against ‘Realism’.” His point is that people do make moral judgments about war, and that “without them we would have no coherent way of talking about war.” In that sense, “realism” is quite unrealistic.

The desire to punish our enemies through war is never far below the surface.

A second fundamental tenet of Christian just war theory is that peace is the natural and desirable state for human beings and that the sole aim of just war is “a certain earthly peace.” (Augustine, City of God, XIV.4) Statecraft that regards war as nothing more than a gambit in the great game of geopolitics or a tool of national power is unjust and immoral. The Times article quotes National Security Advisor Thomas Donilon’s view that Obama is “a president who is quite comfortable with the use of force on behalf of the United States.” By contrast, just war theory is not comfortable with the use of force, and regards it as at best a necessary evil. As for the view that combativeness is part of human nature, like head-butting among rams, Augustine would concede it only because we are fallen, sinful creatures. In his view, “if the earthly city observes Christian principles, even its wars will be waged with the benevolent purpose that better provision might be made for the defeated to live harmoniously together in justice.” (Letter 138 to Marcellinus, ch. 2, sec. 14; translation slightly altered.)

At the same time, traditional just war theory is a long way from pacifism. One of Augustine’s most important works on just war theory is a letter to his friend Marcellinus, a Roman official in North Africa and a devout Christian. He aimed to persuade Marcellinus that the pacifism of some early Church Fathers, grounded in Christ’s injunction to turn the other cheek, was a mistake. Just war theory may be about justice, but it is also very much about war, and it never says that all war is unjust. It limits violence, but it also licenses it.

The most important criterion of just war is possession of a just cause, the paradigmatic example being self-defense. In the wake of the September 11 attacks, there can be little doubt that the United States could morally use force in self-defense. There is a strong case that the U.S. drone campaign against al Qaeda in the Arabian Peninsula is genuine self-defense, given AQAP’s relentless series of (fortunately thwarted) attacks against the American homeland—the underwear bomber, the printer-cartridge bombs, the attempt at a second underwear bombing in May with explosives designed to evade airline security. Even those who favor the use of law enforcement rather than military action in the struggle against al Qaeda accept that force employed to defend against terrorist attacks is justified when lesser measures don’t work.

Another condition of just war, emphasized by Aquinas, is “rightful authority.” Not just anyone can launch a war; only a legitimate ruler can. It follows that non-state actors such as al Qaeda by definition are unjust warriors. (Granted, by this criterion the same would be true of the American revolutionaries, and those who believe that freedom fighters can wage just struggles will reject the “rightful authority” criterion for its bias in favor of the status quo.) One of the most controversial aspects of the use of military commissions at the Guantánamo Bay detention facilities was treating mere membership in al Qaeda as an offense against the laws of war. In my view this treatment is a legal and moral mistake, but the “rightful authority” strand of just war theory supports it.

From Augustine, Obama might also have learned this formula:

As a rule just wars are defined as those which avenge injuries, if some nation or state against whom one is waging war has neglected to punish a wrong committed by its citizens.

This brief text, from a commentary on the Book of Joshua (Questions on the Heptateuch, VI.10) was cited as authority by the most influential Christian writers, including Aquinas, for a thousand years after it was written. Strikingly, the formula refers specifically to nations that don’t control misdeeds by their private citizens, not to nations that are themselves the aggressor. This can easily be read as a precursor to the key U.S. doctrine articulated by National Security Advisor John Brennan last September: to defend itself from al Qaeda, the United States reserves “the right to take unilateral action if or when other governments are unwilling or unable to take the necessary actions themselves.” In other words, the U.S. claims the right to use force in Pakistan and Yemen, even without their governments’ consent, if they can’t or won’t suppress al Qaeda. In a speech last March, Attorney General Eric Holder echoed the “unable or unwilling” doctrine, in the fullest public explanation to date of the Obama administration’s legal justification for drone strikes. The doctrine doesn’t come directly from Augustine, and scholars disagree about its modern-day pedigree. But if the president truly studied and applied traditional just war theory, he might have believed that the doctrine is supported by Augustine’s formula. The doctrine seems like a legitimate corollary of a state’s right to defend itself against unjust attacks. But since it carries deadly risks of civilian casualties and escalation, it can only be justified as a last resort.

Far less appealing is the other half of Augustine’s formula, that “just wars are defined as those which avenge injuries.” By approving wars of punishment, Augustine and Aquinas took a dangerously wrong turn, in effect permitting states to be prosecutor, judge, and executioner in their own cause. Wars of vengeance are immoral and, thankfully, illegal. Under the United Nations Charter, self-defense is the only acceptable basis for states to use force without the consent of the Security Council. To be fair both Brennan and Holder made it clear that they were justifying the use of force against al Qaeda and its allies as self-defense, not punishment. Nevertheless, the desire to punish our enemies through war is never far below the surface. On the anniversary of the killing of Osama bin Laden, President Obama said, “I think that the American people rightly remember what we as a country accomplished in bringing to justice someone who killed over 3,000 of our citizens.” Obama’s boast is wrongheaded but revealing: to call the killing of bin Laden “bringing him to justice” is to mistake a military operation for an execution, with the president as self-appointed judge and jury. If Obama learned this mistake from Augustine or Aquinas, it would be better if he had never read them.

If the president has adopted the CIA’s dishonest accounting of casualties, that would be a disgraceful moral cheat.

Strikingly, what Obama would not have gotten directly from Augustine or Aquinas is the most basic rule of modern warfare: thou shalt not target non-combatants. It makes no appearance in Augustine, and the nearest we find in Aquinas is a general prohibition against killing the innocent. (Aquinas, Summa Theologica, II-II, question 64, article 6) But this prohibition is not in his writing on warfare, and he nowhere clarifies that non-combatants are innocent.

The most agonizing issue in the drone program is figuring out who is an enemy combatant, who is not, and how one knows. The modern law of war is clear, however, that no matter how difficult the inquiry is, it must be undertaken: parties “shall at all times distinguish between the civilian population and combatants and between civilian objects and military objectives and accordingly shall direct their operations only against military objectives.” (Additional Protocol I to the Geneva Conventions, art. 48) There is nothing quite so clear or straightforward in the just war classics, and they would offer no guidance to the president on this score. The administration claims that it faithfully abides by this rule, and its spokespeople hammer that point home in almost every speech. The tricky part is that the government also asserts that al Qaeda “has no ‘non-military’ wing,” and so every member can be targeted. This conclusion rests on a factual assessment of al Qaeda that has never been released or explained.

What about unintended civilian deaths, so-called collateral damage? The rule in the modern law of war is that militaries must do everything feasible to minimize collateral damage, and must never launch an attack at a military target if the expected civilian damage “would be excessive in relation to the concrete and direct military advantage anticipated.” (AP I, art. 51(4) and 51(5)(b); art. 57(2)(a)(iii)) This rule permits proportionate collateral damage so long as harming civilians was not the intention. This principle also has no clear counterpart in medieval just war theory, but it traces its roots to Aquinas’s “doctrine of double effect”:

Nothing hinders a single act from having two effects, only one of which is intended, while the other is beside the intention. Now moral acts get their character in accordance with what is intended, but not from what is beside the intention, since the latter is incidental. (Summa Theologiae, II-II, question 64, article 7)

In the war context, striking the military target is the intended effect, the collateral damage is the unintended effect, and it is the former, not the latter, that determines the moral character of the action. The requirement of proportionality between the intended and unintended effect is a modern refinement of this principle. The Times article reports that President Obama and his advisors have cancelled targeted killings for fear of excessive collateral damage. It was reportedly for this very reason that the president vetoed his advisors’ recommendation of an airstrike on the bin Laden compound in favor of the riskier SEAL operation.

• • •

The most troubling point in the Times article is that the CIA has apparently counted civilian casualties in a self-serving, dishonest way. Any dead military-age male in the vicinity of a targeted strike is presumed a “militant” unless proven otherwise. That sounds wrong, and the more you think about it, the worse it gets.

What would constitute proof that a dead young man is not a militant? And how diligently is the CIA looking for evidence of its own fatal mistakes? After all, this is the same CIA that is investigating itself for illicitly censoring its own critics because it doesn’t want knowledge of its misdeeds and mistakes to see the light of day.

For that matter, what is a military-age man? Any male between 15 and 50? An honest casualty count would presume that women, children, and the elderly are civilians; that only previously identified militants count as combatants; and that, absent further evidence, everyone else is a question-mark who must be counted as a civilian in the proportionality analysis. One official says, “Al Qaeda is an insular, paranoid organization—innocent neighbors don’t hitchhike rides in the back of trucks headed for the border with guns and bombs.” That may well be true, and perhaps it is fair to conclude that all the men in the back of an al Qaeda-identified truck headed for the border with guns and bombs are combatants. (Such a truck would itself be a legitimate target, in any event.) But that is a far cry from assuming that all men of a certain age are combatants whenever they are near an al Qaeda operative.

The Times article also disturbingly suggests (without providing evidence) that the president has adopted the CIA’s dishonest accounting. Does that mean that he takes at face value a post-mortem report from the CIA saying, in effect, “Got the target! And a bunch of militants who were in the house with him! No civilian casualties!” (wink wink, nudge nudge)? Does he then use this marvelous success record in his deliberations about the next operation?—“We are really good at not killing civilians, so there’s little need to be concerned about ‘collateral damage’ in considering the next drone strike.” That would be a disgraceful moral cheat. But the Times has nothing to say about how the president reviews the CIA’s casualty figures, and we don’t really know anything except what the president’s aides want us to know.

• • •

The Times suggests that the main point President Obama took from Augustine and Aquinas is that he, as commander in chief, ought to personally assume the moral responsibility for targeted killings. In Shakespeare’s words, “if the cause be not good, the King himself hath a heavy reckoning to make.” (Henry V, act 4, sc. 1, l. 134) Moral responsibility is central to just war theory. If war is a fit subject for moral judgment, commanders must take moral responsibility for the choices they make. Kudos to Obama for shouldering his responsibility.

That said, the president would not find in either Augustine or Aquinas a requirement that he take personal charge of targeting decisions. That would be an impossible requirement to impose on a battlefield leader who can’t be everywhere at once. But the doctrine of command responsibility, according to which a commander is responsible for crimes by his subordinates that he knows about but fails to prevent or punish, is central to the law of war, and it has impeccable American credentials: it appears in the Articles of War of George Washington’s armies. (1775 Articles of War, art. 12; 1776 Articles of War, sec. IX, art. 1) The point of this rule was to encourage officers to supervise and discipline their men, and Obama has evidently chosen the most straightforward way to do it: running the operation himself.

The opacity and unaccountability of the drone program are threats to the rule of law.

How should we judge Obama’s personal involvement in these life-and-death decisions? Anyone who thinks the targeted killings are immoral or illegal will be repelled. I don’t think that killing deadly enemies is in principle either immoral or illegal. On the assumption that the targeted killing program is a legitimate way for the nation to defend itself against an all-too-real threat, I support presidential involvement. The alternative, after all, is presidential non-involvement—passing the buck to someone else.

The Times editors offered a far more negative view in a strongly worded editorial entitled “Too Much Power for a President.” The Times objects to giving a politician the power over life and death with no independent oversight or judicial review. A fair-enough point, but it blurs two issues: first, whether the decision to strike should be decided exclusively within the executive branch and, second, whether the president should insist on being part of that decision-making. The Times says “no” to the first; I say “yes” to the second. Passing the buck to lower-level bureaucrats and to military and intelligence officials who do not share the president’s electoral accountability will not make the process less political or more accountable, nor will it better protect against tragic and horrifying mistakes. Apparently decisions only reach the president after other officials have approved the use of force. Adding his possible veto, then, can only be for the good, checking unwise uses of force that otherwise would likely have occurred.

I am skeptical about the Times’s preferred alternative, judicial review of targeted killing decisions. What evidence would judges hear that the president doesn’t? And why think that judges would make better decisions on the purely factual question of who is a member of Aa Qaeda and who poses an imminent threat to the United States? Judges don’t face the political pressures on the president, but the last decade has shown that they are seldom profiles in courage when officials tell them that national security is at risk. Ultimately, it’s easier for the president to veto a targeted killing than it would be for a cautious judge.

The real issue is not about decision-making within the executive branch versus outside review. It’s about whether the decision-making process is based on a genuinely skeptical, probing structure, with a heavy burden of proof on those proposing a killing and an institutionalized “devil’s advocate” to argue against each and every deadly “nomination.” The Times article suggests that the actual process over which Obama presides includes some skeptical voices, although former National Security Advisor Dennis Blair, a skeptic of the entire program, was fired. But the article is light on detail, and the process still remains essentially shrouded in fog.

Fog is one of the besetting problems with the Obama drone program. For months the administration did not even admit its existence, even while selectively leaking its supposed successes. Brennan finally confirmed it in April. The administration still has not released the 50-page Justice Department memo offering the legal justification for targeting Anwar al-Awlaki, although Times reporter Charlie Savage reported last fall that it exists. Philip Alston, the New York University law professor who served as the UN’s Special Rapporteur on Extrajudicial, Summary, or Arbitrary Executions, has written an important and scathing article summarizing everything we know about the CIA’s targeted killing program, including wildly varying estimates of collateral damage from credible sources, some of which suggest that civilian casualties are numerous and quite possibly disproportionate. Alston has concluded that in fact we know next to nothing and that the opacity and unaccountability of the program are, in and of themselves, threats to the rule of law. I agree. The decision to provide information for the Times article, like Attorney General Holder’s speech, is a step toward transparency, but a baby step.

Kenneth Anderson raises a more subtle concern about the president’s personal involvement. Is Obama trying to signal that hands-on involvement in targeted killing decisions is a moral imperative for any president? Or is he signaling “that these actions are legitimate only because he is personally trusted to do the right thing, just because he is Barack Obama”? Anderson accepts the former and rejects the latter. Surely he is right. Ultimately, the issue of personal involvement by the president isn’t whether Obama reads Augustine and Aquinas, or whether George W. Bush studied biographies of Lincoln and Churchill, or whether the next president prefers The Girl With the Dragon Tattoo. Of course we want the ultimate decision-maker on matters of life and death to be wise and smart and morally decent, and I still like to think Obama is all those things. We won’t always get that in a president. But we want the decision-maker to be the president for a different reason: to make sure that the buck stops at the top, with no plausible deniability.

• • •

So far I have said very little about the issue that, for the majority of observers, matters most—the drone strikes themselves—beyond my basic point that killing deadly enemies is in principle neither immoral nor illegal. I say “in principle” because so much turns on the details: the expected collateral damage, how much care has been taken to verify the target and the danger he poses, whether the target was trying to surrender, whether the foreign state is truly unwilling or unable to suppress the target, what the non-lethal alternatives were. The wrong answer on any of these issues means the decision to kill from the air flunks the test of morality.

But if the killing is legitimate, the fact that it was targeted, or done by a drone rather than a bomb or a gun, makes no difference. If anything, targeted killing is better than untargeted killing, which the laws of war call “indiscriminate” and a war crime.

A counterproductive campaign is an unjust campaign, because it sheds blood and shatters peace to no good end.

The drone issue, however, seems more complicated for many reasons. To start, there is the dystopic, Terminator-like image of machines relentlessly hunting humans. There is also the jarring fact that the drone operators sit in safety thousands of miles away, insulated from the destruction they cause. And finally there is the nagging sense that somehow it is worse to kill someone whose name you know and whose photo sits on the table before you.

But manned aircraft also kill from a distance, sometimes in a risk-free way. Drone operators watch hours of video of the people they have killed, so they are not insulated (although a study shows that they experience little stress from seeing the video).

What about named versus anonymous killing? It looks like assassination—sneaky, underhanded, and prohibited. But the government holds that wartime killing of the enemy is not the same as assassination, and that includes targeting the enemy commander whose name you know. That sounds right.

Perhaps what really troubles us is the suspicion that if U.S. forces know so much about the target and his location, they could capture him rather than killing him. This worry originated among those who believe that the struggle with al Qaeda should be a criminal justice matter, not a military one. But it has also become a taunt from the pro-torture, pro-Guantánamo right: Obama went soft on detention and torture, so now he has to kill them. In fact, there is no evidence that any of the people targeted by drones could have been captured. Realistically, dropping special forces into Yemen to try to capture an al Qaeda leader (who in any case will have moved on by the time the SEALs arrive) poses greater risk to innocent civilians than a drone strike, given the possibility that the special forces would have to shoot their way out.

There are also legal issues. Starting with the least important, the drone program is run by the CIA, not the military, and that makes the program less accountable and the operators unprivileged combatants. This issue is a legalistic sideshow. Would we really think differently about the drones if they were piloted by uniformed military? Surely not. As for accountability, military combat operations and the rules of engagement that govern them are hardly models of transparency, and are not meant to be.

Other issues are harder. Drone strikes can violate other nations’ sovereignty. Reportedly, Yemen and Pakistan have both consented to drone operations, but I am not convinced that human rights law permits a state to consent to the killing of its own citizens on its territory. There is also the difficult question of what makes Yemen a war zone. Lawyers debate these questions vigorously. All the more reason for the Obama administration to release its detailed legal reasoning.

Then there is the drone killing of American citizens such as Anwar al-Awlaki. The U.S. debate about drone strikes has focused obsessively and almost exclusively on whether, in the words of the ACLU’s Anthony Romero, “President Obama is allowed to execute American citizens without judicial review and outside the theater of war.” It sounds terrible, doesn’t it? Of course, calling the killings “executions” assumes that targeted killings cannot be legitimate acts of war. But leave that vexed issue to one side and focus on the phrase “American citizens.” Is it worse, in Mr. Romero’s eyes, to kill American citizens than to kill foreigners? Hopefully not.

From the point of view of just war theory, the nationality of casualties is irrelevant. If they are enemy belligerents, they can be targeted, regardless of their nationality; if they are not enemy belligerents, they can’t be, regardless of their nationality. (AP I to the Geneva Conventions, articles 51 and 57) To focus only on the lives of Americans is parochial in a way that the morality of war is not.

Among this welter of arguments about targeted killing, the genuine issues of principle are whether self-defense requires it and proportionality permits it. The question of where the zone of combat ends and civilian rules begin is important, but it is a question of line-drawing, not of moral principle. If self-defense is a just cause of war, and if killing is necessary for self-defense (a big if), then targeted killing is permissible—provided that it targets only enemy fighters, keeps civilian casualties low, and actually does more good than harm in defending ourselves.

But whether targeted killing by drones does more harm than good is far from a settled question. Just a day after the Times article, the Washington Post reported that drone strikes in Yemen are infuriating and radicalizing Yemenis, turning them into Al Qaeda sympathizers, and enabling Al Qaeda to expand its membership and the territory under its control. The Post raises the terrible thought that the president and his advisors, focused on their Power Point presentations and “baseball cards” of suspected militants, are missing something vital and conducting a campaign that may undermine national self-defense rather than secure it. That would be a moral failing, not merely a failure of self-interest. One of the classical criteria of just war, dating from the seventeenth century, is the requirement of reasonable likelihood of success. A counterproductive campaign is an unjust campaign, because it sheds blood and shatters peace to no good end. Whether the president’s strategy satisfies this criterion remains to be seen. He has a heavy reckoning to make.