Terrorism and the Philosophers

Can the ends ever justify the means?

Are acts of terror always evil? The answer to this question, oddly enough, depends on a rather ill-understood moral principle, invented by Catholic casuists in the Middle Ages, called the doctrine of double effect.

Before getting to that, a few preliminaries. What constitutes an act of terror? The broadest definition is the deliberate killing of noncombatants. That, for example, is how Caleb Carr characterizes terrorism in his recent book The Lessons of Terror. For this he was taken to task in the New York Times Book Review by Michael Ignatieff, who insisted that if the slaughter is carried out by “a state army under regular command, as part of a formally declared campaign to defeat another state,” then it ought not to be called terrorism. But then in the current issue of TheNation, Richard Falk complains about the Bush administration’s “narrowing of terrorism to apply only to violence by nonstate movements and organizations, thereby exempting state violence against civilians from the prohibition on terrorism.”

The decision to reserve “terrorism” for nonstate acts of terror, or to extend it to state acts, is a semantic one. Morally, it does not matter whether the murderers of civilians are wearing uniforms or not. What might be morally relevant, though, is the cause behind the act of terror. If the cause is a bad one—revenge, say, or a war of aggression—then the act of terror is obviously bad: evil in pursuit of evil. But it is logically possible for the cause to be a good one. Ignatieff cites General Sherman’s murderous march through Georgia, which was intended “to bring the Civil War to a speedy conclusion.” Since Sherman’s intention was good, Ignatieff argues, he was not a terrorist. But this ends the debate too hastily, making terrorism evil by definition. The interesting question remains: Can the use of terror in a good cause, whether by a state or nonstate agents, ever be morally justified?

If the killing of noncombatants were absolutely forbidden, then almost any military action (excepting sea battles and desert wars) would be morally out of the question. When naval dockyards, munitions factories, and supply lines are bombed, civilian carnage is inevitable. That is where the doctrine of double effect comes in. Though it is always wrong to kill innocents deliberately, this doctrine says, it is sometimes permissible to attack a military target with the certain foreknowledge that some noncombatants will die as a side effect. It may even be permissible to bomb a hospital where Hitler is lying ill.

The doctrine of double effect, which can be traced back to Thomas Aquinas’ Summa Theologiae, applies to any sort of act that has two kinds of effects, good and evil. By its logic, such an act is morally allowable only when the following conditions are fulfilled: 1) The agent is aiming only at the good effect; the evil effect is not one of his ends, nor is it a means to his ends; and 2) the consequences of the act are good on balance; that is, the goodness of the good effect outweighs the evil of the evil effect. (The philosopher Michael Walzer has argued that a third condition should be added: that the actor seek to minimize the evil effect, accepting costs to himself. This is a condition that, for example, the U.S. bombing campaign over Kosovo failed to meet, since pilots flew high to protect themselves and dropped bombs inaccurately, which resulted in greater civilian death.)

Acts of terror involve the deliberate killing of noncombatants as a means to an end. Thus they are forbidden by the law of double effect, specifically by condition 1). But is this doctrine really valid? Why not leave out intentions and simply judge the rightness or wrongness of an act by its consequences, the way utilitarians do? Take the bombings of Hiroshima and Nagasaki. Between 120,000 and 250,000 civilians were deliberately executed en masse to cow the rulers of Japan into submission. Truman claimed in his memoirs that the alternative, an invasion, would have cost half a million American lives. If he was right (a big “if,” as revisionist historians will tell you), the net savings from these acts of terror amounted to around a quarter of a million lives. Why, one might ask, let medieval scruples about intentions get in the way of that?

There are plenty of contemporary moral philosophers who are unhappy with the doctrine of double effect. Some (Peter Singer, for example) reject it because they think the distinction between directly intended effects and inevitable side effects is a contrived one. If you ask the terror bomber why he is killing civilians, he will say, “To win a just war.” He might even say that he does not need the civilians actually to be dead, but only to be thought to be dead until the war is over to demoralize the other side. If his victims could be miraculously brought back to life after the end of the struggle, he would not object. In this sense, he does not really intend their deaths. (This point has been made by the philosopher Jonathan Bennett.) To take a different case: If I can kill Saddam Hussein only by shooting him through an innocent human shield, do I intend harm to the innocent shield or not?

A second problem with double effect (raised by the English philosopher Jonathan Glover) is identifying the class of evil acts that can never be justified by their good effects. It is not hard to imagine a case in which lying to an innocent person might save a life. Would the doctrine of double effect forbid this? Presumably not, but then why does it forbid the deliberate killing of an innocent person? Even supposing that killing is a million times worse than lying, if you can lie to save a life, why can’t you kill to save a million lives?

Finally, there is irrelevance of intentions to the harms caused by an act. The incidental victims of a strategic bomber are just as dead as the intended victims of a terror bomber; their surviving family members are no less grief-stricken. A civilian threatened by a strategic bomber has just as strong a moral claim to be rescued as a civilian threatened by a terror bomber.

Defenders of the doctrine of double effect appeal to Kant’s categorical imperative: A person is always to be treated as an end, never merely as a means. And treating people as a means—to send a message, to create wider panic, to demoralize the enemy nation they are a part of—is precisely what the perpetrator of an act of terror does. “He sees them as material to be strategically shaped or framed by his agency,” the moral philosopher Warren S. Quinn writes. “He must treat them as if they were then and there for his purposes.”

The debate over the doctrine of double effect, though philosophically interesting, is mostly moot. It is possible to think up hypothetical cases where an evil act is a means to a disproportionately good outcome—”If you boil this baby it will save 10 lives.” But in the real world, acts of terror are rarely efficacious in a good cause (somewhat less rarely in a bad one). That is not surprising when you think about it. Deliberately killing noncombatants does not weaken an enemy militarily, precisely because they are noncombatants—children, the aged, and so forth. Far from having a demoralizing effect, the bombing of civilians both by the Germans and by the British during World War II seemed to stiffen the sinews of each side. In The Lessons of Terror, Caleb Carr produces ample historical evidence that terror is especially impotent when it comes to fighting terror—witness the sorry experience of the French in Algeria. All of which suggests an empirical counterpart to the doctrine of double effect: To act knavishly in a good cause is to act foolishly.