Resistance to Military Occupation
Surprise is the essential feature of guerrilla war; thus the ambush is the classic guerrilla tactic. It is also, of course, a tactic in conventional war; the concealment and camouflage that it involves, though they were once repugnant to officers and gentlemen, have long been regarded as legitimate forms of combat. But there is one kind of ambush that is not legitimate in conventional war and that places in sharp focus the moral difficulties guerrillas and their enemies regularly encounter. This is the ambush prepared behind political or moral rather than natural cover. An example is provided by Captain Helmut Tausend, of the German Army, in Marcel Ophuls’ documentary film The Sorrow and the Pity. Tausend tells of a platoon of soldiers on a march through the French countryside during the years of the German occupation. They passed a group of young men, French peasants, or so it seemed, digging potatoes. But these were not in fact peasants; they were members of the Resistance. As the Germans marched by, the “peasants” dropped their shovels, picked up guns hidden in the field, and opened fire. Fourteen of the soldiers were hit. Years later, their captain was still indignant. “You call that ‘partisan’ resistance? I don’t. Partisans for me are men that can be identified, men who wear a special armband or a cap, something with which to recognize them. What happened in that potato field was murder.”1
The captain’s argument about armbands and caps is simply a citation from the international law of war, from the Hague and Geneva conventions, and I shall have more to say about it later on. It is important to stress first that the partisans had here taken on a double disguise. They were disguised as peaceful peasants and also as Frenchmen, that is, citizens of a state that had surrendered, for whom the war was over (just as guerrillas in a revolutionary struggle disguise themselves as unarmed civilians and also as loyal citizens of a state that is not at war at all). It was because of this second disguise that the ambush was so perfect. The Germans thought they were in a rear area, not at the front, and so they were not battle-ready; they were not preceded by a scouting party; they were not suspicious of the young men in the field. The surprise achieved by the partisans was of a kind virtually impossible in actual combat. It derived from what might be called the protective coloration of national surrender, and its effect was obviously to erode the moral and legal understandings upon which surrender rests.
Surrender is an explicit agreement and exchange: the individual soldier promises to stop fighting in exchange for benevolent quarantine for the duration of the war; a government promises that its citizens will stop fighting in exchange for the restoration of ordinary public life. The precise conditions of “benevolent quarantine” and “public life” are specified in the law books; I need not go into them here.2 The obligations of individuals are also specified: they may try to escape from the prison camp or to flee occupied territory, and if they succeed in their escape or flight, they are free to fight again; they have regained their war rights. But they may not resist their quarantine or occupation. If a prisoner kills a guard in the course of his escape, the act is murder; if the citizens of a defeated country attack the occupation authorities, the act has, or once had, an even grimmer name: it is, or was, “war treason” (or “war rebellion”), a breaking of political faith, punishable, like the ordinary treason of rebels and spies, by death.
But “traitor” does not seem the right name for those French partisans. Indeed, it is precisely their experience, and that of other guerrilla fighters in World War II, that has led to the virtual disappearance of “war treason” from the law books and of the idea of breaking faith from our moral discussions of wartime resistance (and of peacetime rebellion also, when it is directed against alien or colonial rule). We tend to deny, today, that individuals are automatically subsumed by the decisions of their government or the fate of its armies. We have come to understand the moral commitment they may feel to defend their homeland and their political community, even after the war is officially over.3 A prisoner of war, after all, knows that the fighting will go on despite his own capture; his government is in place, his country is still being defended. But after national surrender the case is different, and if there are still values worth defending, no one can defend them except ordinary men and women, citizens with no political or legal standing. I suppose it is some general sense that there are such values, or often are, that leads us to grant these men and women a kind of moral authority.
But though this grant reflects new and valuable democratic sensibilities, it also raises serious questions. For if citizens of a defeated state still have a right to fight, what is the meaning of surrender? And what obligations can be imposed on conquering armies? There can be no ordinary public life in occupied territory if the occupation authorities are subject to attack at any time and at the hands of any citizen. And ordinary life is a value, too. It is what most of the citizens of a defeated country most ardently hope for. The heroes of the resistance put it in jeopardy, and we must weigh the risks they impose on others in order to understand the risks they must accept themselves. Moreover, if the authorities actually do aim at the restoration of everyday peacefulness, they seem entitled to enjoy the security they provide; and then they must also be entitled to regard armed resistance as a criminal activity. So the story with which I began might end this way (in the film it has no end): the surviving soldiers rally and fight back; some of the partisans are captured, tried as murderers, condemned, and executed. We would not, I think, add those executions to the list of Nazi war crimes. At the same time, we would not join in the condemnation.
So the situation can be summed up: resistance is legitimate, and the punishment of resistance is legitimate. That may seem like a simple standoff and an abdication of ethical judgment. It is actually a precise reflection of the moral realities of military defeat. I want to stress again that our understanding of these realities has nothing to do with our view of the two sides. We can deplore the resistance, without calling the partisans traitors; we can hate the occupation, without calling the execution of the partisans a crime. If we alter the story or add to it, of course, the case is changed. If the occupation authorities do not live up to their obligations under the surrender agreement, they lose their entitlements. And once the guerrilla struggle has reached a certain point of seriousness and intensity, we may decide that the war has effectively been renewed, notice has been given, the front has been re-established (even if it is not a line), and soldiers no longer have a right to be surprised even by a surprise attack. Then guerrillas captured by the authorities must be treated as prisoners of war—provided, that is, they have themselves fought in accordance with the war convention.
But guerrillas don’t fight that way. Their struggle is subversive not merely with reference to the occupation or to their own government, but with reference to the war convention itself. Wearing peasant clothes and hiding among the civilian population, they challenge the most fundamental principle of the rules of war. For it is the purpose of those rules to specify for each individual a single identity; he must be either a soldier or a civilian. The British Manual of Military Law makes the point with special clarity: “Both these classes have distinct privileges, duties, and disabilities. . . . An individual must definitely choose to belong to one class or the other, and shall not be permitted to enjoy the privileges of both; in particular . . . an individual [shall] not be allowed to kill or wound members of the army of the opposed nation and subsequently, if captured or in danger of life, pretend to be a peaceful citizen.”4 That is what guerrillas do, however, or sometimes do. So we can imagine another conclusion to the story of the partisan attack. The partisans successfully disengage, disperse to their homes, and go about their ordinary business. When German troops come to the village that night, they cannot distinguish the guerrilla fighters from any other of the villagers. What do they do then? If, through searches and interrogations—police, not soldier’s, work—they seize one of the partisans, should they treat him as a captured criminal or a prisoner of war (leaving aside now the problems of surrender and resistance)? And if they seize no one, can they punish the whole village? If the partisans don’t maintain the distinction of soldiers and civilians, why should they?
The Rights of Guerrilla Fighters
As this example suggests, the guerrillas don’t subvert the war convention by themselves attacking civilians; at least, it is not a necessary feature of their struggle that they do that. Instead, they invite their enemies to do it. By refusing to accept a single identity, they seek to make it impossible for their enemies to accord to combatants and noncombatants their “distinct privileges . . . and disabilities.” The political creed of the guerrillas is essentially a defense of this refusal. The people, they say, are no longer being defended by an army; the only army in the field is the army of the oppressors; the people are defending themselves. Guerrilla war is “people’s war,” a special form of the levée en masse, authorized from below. “The war of liberation,” according to a pamphlet of the Vietnamese National Liberation Front, “is fought by the people themselves; the entire people . . . are the driving force. . . . Not only the peasants in their rural areas, but the workers and laborers in the city, along with intellectuals, students, and businessmen have gone to fight the enemy.”5 And the NLF drove the point home by naming its paramilitary forces Dan Quan, literally, civilian soldiers. The guerrilla’s self-image is not of a solitary fighter hiding among the people, but of a whole people mobilized for war, himself a loyal member, one among many. If you want to fight against us, the guerrillas say, you are going to have to fight civilians, for you are not at war with an army but with a nation. Therefore, you should not fight at all, and if you do, you are the barbarians, killing women and children.
In fact, the guerrillas mobilize only a small part of the nation—a very small part, when they first begin their attacks. They depend upon the counter-attacks of their enemies to mobilize the rest. Their strategy is framed in terms of the war convention: they seek to place the onus of indiscriminate warfare on the opposing army. The guerrillas themselves have to discriminate, if only to prove that they are really soldiers (and not enemies) of the people. It is also and perhaps more importantly true that it is relatively easy for them to make the relevant discriminations. I don’t mean that guerrillas never engage in terrorist campaigns (even against their fellow countrymen) or that they never take hostages or burn villages. They do all those things, though they generally do less of them than the anti-guerrilla forces. For the guerrillas know who their enemies are, and they know where they are. They fight in small groups, with small arms, at close quarters—and the soldiers they fight against wear uniforms. Even when they kill civilians, they are able to make distinctions: they aim at well-known officials, notorious collaborators, and so on. If the “entire people” are not really the “driving force,” they are also not the objects of guerrilla attack.
For this reason, guerrilla leaders and publicists are able to stress the moral quality not only of the goals they seek but also of the means they employ. Consider for a moment Mao Tse-tung’s famous “Eight Points for Attention.” Mao is by no means committed to the notion of noncombatant immunity (as we shall see), but he writes as if, in the China of the warlords and the Kuomintang, only the communists respect the lives and property of the people. The “Eight Points” are meant to mark off the guerrillas first of all from their predecessors, the bandits of traditional China, and then from their present enemies, who ravage the countryside. They suggest how the military virtues can be radically simplified for a democratic age.6
The last of these is particularly problematic, for in the conditions of guerrilla war it must often involve releasing prisoners, something most guerrillas are no doubt loath to do. Yet it is at least sometimes done, as an account of the Cuban revolution, originally published in the Marine Corps Gazette, suggests:7
That same evening, I watched the surrender of hundreds of Batistianos from a small-town garrison. They were gathered within a hollow square of rebel Tommy-gunners and harangued by Raul Castro:
“We hope that you will stay with us and fight against the master who so ill-used you. If you decide to refuse this invitation—and I am not going to repeat it—you will be delivered to the custody of the Cuban Red Cross tomorrow. Once you are under Batista’s orders again, we hope that you will not take arms against us. But, if you do, remember this:
“We took you this time. We can take you again. And when we do, we will not frighten or torture or kill you. . . . If you are captured a second time or even a third . . . we will again return you exactly as we are doing now.”
Even when guerrillas behave this way, however, it is not clear that they are themselves entitled to prisoner of war status when captured, or that they have any war rights at all. For if they don’t make war on noncombatants, it also appears that they don’t make war on soldiers: “What happened in that potato field was murder.” They attack stealthily, deviously, without warning, and in disguise. They violate the implicit trust upon which the war convention rests: soldiers must feel safe among civilians if civilians are ever to be safe from soldiers. It is not the case, as Mao once suggested, that guerrillas are to civilians as fish to the ocean. The actual relation is rather of fish to other fish, and the guerrillas are as likely to appear among the minnows as among the sharks.
That, at least, is the paradigmatic form of guerrilla war. I should add that it is not the form such war always or necessarily takes. The discipline and mobility required of guerrilla fighters often preclude a domestic retreat. Their main forces commonly operate out of base camps located in remote areas of the country. And, curiously enough, as the guerrilla units grow larger and more stable, their members are likely to put on uniforms. Tito’s partisans in Yugoslavia, for example, wore distinctive dress, and this was apparently no disadvantage in the kind of war they fought.8 All the evidence suggests that quite apart from the rules of war, guerrillas, like other soldiers, prefer to wear uniforms; it enhances their sense of membership and solidarity. In any case, soldiers attacked by a guerrilla main force know who their enemies are as soon as the attack begins; ambushed by uniformed men, they would know no sooner. When the guerrillas “melt away” after such an attack, they more often disappear into jungles or mountains than into villages, a retreat that raises no moral problems. Battles of this sort can readily be assimilated to the irregular combat of army units like Wingate’s “Chindits” or “Merrill’s Marauders” in World War II.9 But this is not what most people have in mind when they talk about guerrilla war. The paradigm worked out by guerrilla publicists (together with their enemies) focuses precisely on what is morally difficult about guerrilla war—and also, as we shall see, about anti-guerrilla war. In order to deal with these difficulties, I shall simply accept the paradigm and treat guerrillas as they ask to be treated, as fish among the ocean’s fish. What then are their war rights?
The legal rules are simple and clearcut, though not without their own problems. To be eligible for the war rights of soldiers, guerrilla fighters must wear “a fixed distinctive sign visible at a distance” and must “carry their arms openly.”10 It is possible to worry at length about the precise meaning of distinctiveness, fixity, and openness, but I do not think we would learn a great deal by doing so. In fact, these requirements are often suspended, particularly in the interesting case of a popular rising to repel invasion or resist foreign tyranny. When the people rise en masse, they are not required to put on uniforms. Nor will they carry arms openly, if they fight, as they usually do, from ambush: hiding themselves, they can hardly be expected to display their weapons. Francis Lieber, in one of the earliest legal studies of guerrilla war, cites the case of the Greek rebellion against Turkey, where the Turkish government killed or enslaved all prisoners: “But I take it,” he writes, “that a civilized government would not have allowed the fact that the Greeks . . . carried on mountain guerrilla [war] to influence its conduct toward prisoners.”11
The key moral issue, which the law gets at only imperfectly, does not have to do with distinctive dress or visible weapons, but with the use of civilian clothing as a ruse and a disguise.a The French partisan attack perfectly illustrates this, and it has to be said, I think, that the killing of those German soldiers was more like assassination than war. That is not because of the surprise, simply, but because of the kind and degree of deceit involved: the same sort of deceit that is involved when a public official or party leader is shot down by some political enemy who has taken on the appearance of a friend and supporter or of a harmless passer-by. Now it may be the case—I am more than open to this suggestion—that the German army in France had attacked civilians in ways that justified the assassination of individual soldiers, just as it may be the case that the public official or party leader is a brutal tyrant who deserves to die. But assassins cannot claim the protection of the rules of war; they are engaged in a different activity. Most of the other enterprises for which guerrillas require civilian disguise are also “different.” These include all the possible varieties of espionage and sabotage; they can best be understood by comparing them to acts carried out behind enemy lines by the secret agents of conventional armies. It is widely agreed that such agents possess no war rights, even if their cause is just. They know the risks their efforts entail, and I see no reason to describe the risks of guerrillas engaged in similar projects any differently. Guerrilla leaders claim war rights for all their followers, but it makes sense to distinguish, if this is possible, between those guerrillas who use civilian dress as a ruse and those who depend upon camouflage, the cover of darkness, tactical surprise, and so on.
The issues posed by the guerrilla war paradigm, however, are not resolved by this distinction. For guerrillas don’t merely fight as civilians; they fight among civilians, and this in two senses. First, their day-to-day existence is much more closely connected with the day-to-day existence of the people around them than is ever the case with conventional armies. They live with the people they claim to defend, whereas conventional troops are usually billeted with civilians only after the war or the battle is over. And second, they fight where they live; their military positions are not bases, posts, camps, forts, or strongholds, but villages. Hence they are radically dependent on the villagers, even when they don’t succeed in mobilizing them for “people’s war.” Now, every army depends upon the civilian population of its home country for supplies, recruits, and political support. But this dependence is usually indirect, mediated by the bureaucratic apparatus of the state or the exchange system of the economy. So food is passed from the farmer to the marketing co-op, to the food processing plant, to the trucking company, to the army commissary. But in guerrilla war, the dependence is immediate: the farmer hands the food to the guerrilla, and whether it is received as a tax or paid for in accordance with Mao’s Second Point for Attention, the relation between the two men is face-to-face. Similarly, an ordinary citizen may vote for a political party that in turn supports the war effort and whose leaders are called in for military briefings. But in guerrilla war, the support a civilian provides is far more direct. He doesn’t need to be briefed; he already knows the most important military secret; he knows who the guerrillas are. If he doesn’t keep this information to himself, the guerrillas are lost.
Their enemies say that the guerrillas rely on terror to win the support or at least the silence of the villagers. But it seems more likely that when they have significant popular support (which they don’t always have), they have it for other reasons. “Violence may explain the cooperation of a few individuals,” writes an American student of the Vietnamese war, “but it cannot explain the cooperation of a whole social class [the peasantry].”12 If the killing of civilians were sufficient to win civilian support, the guerrillas would always be at a disadvantage, for their enemies possess far more fire power than they do. But killing will work against the killer “unless he has already pre-empted a large part of the population and then limits his acts of violence to a sharply defined minority.” When the guerrillas succeed, then, in fighting among the people, it is best to assume that they have some serious political support among the people. The people, or some of them, are complicitous in guerrilla war, and the war would be impossible without their complicity. That doesn’t mean that they seek out opportunities to help. Even when he sympathizes with the goal of the guerrillas, we can assume that the average civilian would rather vote for them than hide them in his house. But guerrilla war makes for enforced intimacies, and the people are drawn into it in a new way even though the services they provide are nothing more than functional equivalents of the services civilians have always provided for soldiers. For the intimacy is itself an additional service, which has no functional equivalent. Whereas soldiers are supposed to protect the civilians who stand behind them, guerrillas are protected by the civilians among whom they stand.
But the fact that they accept this protection, and depend upon it, doesn’t seem to me to deprive the guerrillas of their war rights. Indeed, it is more plausible to make exactly the opposite argument: that the war rights the people would have were they to rise en masse are passed on to the irregular fighters they support and protect—assuming that the support, at least, is voluntary. For soldiers acquire war rights not as individual warriors but as political instruments, servants of a community that in turn provides services for its soldiers. Guerrillas take on a similar identity whenever they stand in a similar or equivalent relationship, that is, whenever the people are helpful and complicitous in the ways I have described. When the people do not provide this recognition and support, guerrillas acquire no war rights, and their enemies may rightly treat them when captured as “bandits” or criminals. But any significant degree of popular support entitles the guerrillas to the benevolent quarantine customarily offered prisoners of war (unless they are guilty of specific acts of assassination or sabotage, for which soldiers, too, can be punished).b
This argument clearly establishes the rights of the guerrillas; it raises the most serious questions, however, about the rights of the people; and these are the crucial questions of guerrilla war. The intimacies of the struggle expose the people in a new way to the risks of combat. In practice, the nature of this exposure, and its degree, are going to be determined by the government and its allies. So the burdens of decision are shifted by the guerrillas onto their enemies. It is their enemies who must weigh (as we must) the moral significance of the popular support the guerrillas both enjoy and exploit. One can hardly fight against men and women who themselves fight among civilians without endangering civilian lives. Have these civilians forfeited their immunity? Or do they, despite their wartime complicity, still have rights vis-à-vis the anti-guerrilla forces?
The Rights of Civilian Supporters
If civilians had no rights at all, or were thought to have none, it would be a small benefit to hide among them. In a sense, then, the advantages the guerrillas seek depend upon the scruples of their enemies—though there are other advantages to be had if their enemies are unscrupulous: that is why anti-guerrilla warfare is so difficult. I shall want to argue that these scruples in fact have a moral basis, but it is worth suggesting first that they also have a strategic basis. It is always in the interest of the anti-guerrilla forces to insist upon the soldier/civilian distinction, even when the guerrillas act (as they always will if they can) so as to blur the line. All the handbooks on “counter-insurgency” make the same argument: what is necessary is to isolate the guerrillas from the civilian population, to cut them off from their protection and at the same time to shield civilians from the fighting.13 The last point is more important in guerrilla than in conventional war, for in conventional war one assumes the hostility of “enemy civilians,” while in a guerrilla struggle one must seek their sympathy and support. Guerrilla war is a political, even an ideological conflict. “Our kingdoms lay in each man’s mind,” wrote T. E. Lawrence of the Arab guerrillas he led in World War I. “A province would be won when we had taught the civilians in it to die for our ideal of freedom.”14 And it can be won back only if those same civilians are taught to live for some counter-ideal (or in the case of a military occupation, to acquiesce in the re-establishment of order and ordinary life). That is what is meant when it is said that the battle is for the “hearts and minds” of the people. And one cannot triumph in such a battle by treating the people as so many enemies to be attacked and killed along with the guerrillas who live among them.
But what if the guerrillas cannot be isolated from the people? What if the levée en masse is a reality and not merely a piece of propaganda? Characteristically, the military handbooks neither pose nor answer such questions. There is, however, a moral argument to be made if this point is reached: the anti-guerrilla war can then no longer be fought—and not just because, from a strategic point of view, it can no longer be won. It cannot be fought because it is no longer an anti-guerrilla but an anti-social war, a war against an entire people, in which no distinctions would be possible in the actual fighting. But this is the limiting case of guerrilla war. In fact, the rights of the people come into play earlier on, and I must try now to give them some plausible definition.
Consider again the case of the partisan attack in occupied France. If, after the ambush, the partisans hide in a nearby peasant village, what are the rights of the peasants among whom they hide? German soldiers arrive that night, let’s say, seeking the men and women directly involved or implicated in the ambush and looking also for some way of preventing future attacks. The civilians they encounter are hostile, but that doesn’t make them enemies in the sense of the war convention, for they don’t actually resist the efforts of the soldiers. They behave exactly as citizens sometimes do in the face of police interrogations: they are passive, blank, evasive. We must imagine a domestic state of emergency and ask how the police might legitimately respond to such hostility. Soldiers can do no more when what they are doing is police work; for the status of the hostile civilians is no different. Interrogations, searches, seizures of property, curfews—all these seem to be commonly accepted (I will not try to explain why); but not the torture of suspects or the taking of hostages or the internment of men and women who are or might be innocent.15 Civilians still have rights in such circumstances. If their liberty can be temporarily abridged in a variety of ways, it is not entirely forfeit; nor are their lives at risk. The argument would be much harder, however, had the troops been ambushed as they marched through the village itself, shot at from the cover of peasant homes and barns. To understand what happens then, we must look at another historical example.
The American “Rules of Engagement” in Vietnam
Here is a typical incident of the Vietnam War. “An American unit moving along Route 18 [in Long An province] received small arms fire from a village, and in reply the tactical commander called for artillery and air strikes on the village itself, resulting in heavy civilian casualties and extensive physical destruction.”16 Something like this must have happened hundreds, even thousands of times. The bombing and strafing of peasant villages was a common tactic of the American forces. It is a matter of special interest to us that it was permitted by the U.S. Army’s “rules of engagement,” worked out, so it was said, to isolate the guerrillas and minimize civilian casualties.
The attack on the village near Route 18 looks as if it was intended to minimize only army casualties. It looks like another instance of a practice I have already examined: the indiscriminate use of modern fire power to save soldiers from trouble and risk. But in this case, the trouble and risk are of a sort very different from anything encountered on the front line of a conventional war. It is most unlikely that an army patrol moving into the village would have been able to locate and destroy an enemy position. The soldiers would have found . . . a village, its population sullen and silent, the guerrilla fighters hiding, the guerrilla “fortifications” indistinguishable from the homes and shelters of the villagers. They might have drawn hostile fire; more likely, they would have lost men to mines and booby traps, the exact location of which everyone in the village knew and no one would reveal. Under such circumstances, it was not difficult for soldiers to convince themselves that the village was a military stronghold and a legitimate target. And if it was known to be a stronghold, surely it could be attacked, like any other enemy position, even before hostile fire was encountered. In fact, this became American policy quite early in the war: villages from which hostile fire might reasonably be expected were shelled and bombed before soldiers moved in and even if no movement was planned. But then how does one minimize civilian casualties, let alone win over the civilian population? It was to answer this question that the rules of engagement were developed.
The crucial point of the rules, as they are described by the journalist Jonathan Schell, was that civilians were to be given warning in advance of the destruction of their villages, so that they could break with the guerrillas, expel them, or leave themselves.17 The goal was to force the separation of combatants and noncombatants, and the means was terror. Enormous risk was attached to complicity in guerrilla war, but this was a risk that could only be imposed on whole villages; no further differentiation was possible. It is not the case that civilians were held hostage for the activities of the guerrillas. Rather, they were held responsible for their own activity, even when this activity was not overtly military. The fact that the activity sometimes was overtly military, that ten-year-old children threw hand grenades at American soldiers (the incidence of such attacks was probably exaggerated by the soldiers, in part to justify their own conduct toward civilians) blurs the nature of this responsibility. But it has to be stressed that a village was regarded as hostile not because its women and children were prepared to fight, but because they were not prepared to deny material support to the guerrillas or to reveal their whereabouts or the location of their mines and booby traps.
These were the rules of engagement: (1) A village could be bombed or shelled without warning if American troops had received fire from within it. The villagers were presumed able to prevent the use of their village as a fire base, and whether or not they actually were able, they certainly knew in advance whether it would be so used. In any case, the shooting itself was a warning, since return fire was to be expected—though it is unlikely that the villagers expected the response to be as disproportionate as it usually was, until the pattern had become familiar. (2) Any village known to be hostile could be bombed or shelled if its inhabitants were warned in advance, either by the dropping of leaflets or by helicopter loudspeaker. These warnings were of two sorts: sometimes they were specific in character, delivered immediately before an attack, so that the villagers only just had time to leave (and then the guerrillas could leave with them), or they were general, describing the attack that might come if the villagers did not expel the guerrillas.
The U.S. Marines will not hesitate to destroy immediately any village or hamlet harboring the Vietcong. . . . The choice is yours. If you refuse to let the Vietcong use your villages and hamlets as their battlefield, your homes and your lives will be saved.
And if not, not. Despite the emphasis on choice, this is not quite a liberal pronouncement, for the choice in question is very much a collective one. Exodus, of course, remained an individual option: people could move out of villages where the Vietcong had established itself, taking refuge with relatives in other villages, or in the cities, or in government-run camps. Most often, however, they did this only after the bombing had begun, either because they did not understand the warnings, or did not believe them, or simply hoped desperately that their own homes would somehow be spared. Hence it was sometimes thought humane to dispense with choice altogether and forcibly to deport villagers from areas that were considered under enemy control. Then the third rule of engagement went into effect. (3) Once the civilian population had been moved out, the village and surrounding country might be declared a “free fire zone” that could be bombed and shelled at will. It was assumed that anyone still living in the area was a guerrilla or a “hardened” guerrilla supporter. Deportation had stripped away civilian cover as defoliation stripped away natural cover, and left the enemy exposed.18
In considering these rules, the first thing to note is that they were radically ineffective. “My investigation disclosed,” writes Schell, “that the procedures for applying these restraints were modified or twisted or ignored to such an extent that in practice the restraints evaporated entirely. . . .”19 Often, in fact, no warning was given, or the leaflets were of little help to villagers who could not read, or the forcible evacuation left large numbers of civilians behind, or no adequate provision was made for the deported families and they drifted back to their homes and farms. None of this, of course, would reflect on the value of the rules themselves, unless the ineffectiveness were somehow intrinsic to them or to the situation in which they were applied. This was clearly the case in Vietnam. For where the guerrillas have significant popular support and have established a political apparatus in the villages, it is unrealistic to think that the villagers will or can expel them. This has nothing to do with the virtues of guerrilla rule: it would have been equally unrealistic to think that German workers, though their homes were bombed and their families killed, would overthrow the Nazis. Hence the only protection the rules provide is in advising or enforcing the departure not of guerrillas from peaceful villages but of civilians from what is likely to become a battlefield.
Now, in a conventional war, removing civilians from a battlefield is clearly a good thing to do; positive international law requires it wherever possible. Similarly in the case of a besieged city: civilians must be allowed to leave; and if they refuse (so I have argued), they can be attacked along with the defending soldiers. But a battlefield and a city are determinate areas, and a battle and a siege are, usually, of limited duration. Civilians move out; then they move back. Guerrilla war is likely to be very different. The battlefield extends over much of the country and the struggle is, as Mao has written, “protracted.” Here the proper analogy is not to the siege of a city but to the blockade or strategic devastation of a much wider area. The policy underlying the American rules of engagement actually envisaged the uprooting and resettlement of a very substantial part of the rural population of Vietnam: millions of men, women, and children. But that is an incredible task, and, leaving aside for the moment the likely criminality of the project, there was never more than a pretense that sufficient resources would be made available to accomplish it. It was inevitable then, and it was known to be inevitable, that civilians would be living in the villages that were shelled and bombed.
What happened is quickly described:20
In August 1967, during Operation Benton, the “pacification” camps became so full that Army units were ordered not to “generate” any more refugees. The Army complied. But search and destroy operations continued. Only now the peasants were not warned before an air-strike was called on their village. They were killed in their villages because there was no room for them in the swamped pacification camps.
I should add that this sort of thing doesn’t always happen, even in anti-guerrilla war—though the policy of forced resettlement or “concentration,” from its origins in the Cuban Insurgency and the Boer War, has rarely been carried out in a humane manner or with adequate resources.21 But one can find counter-examples. In Malaya, in the early 1950s, where the guerrillas had the support of only a relatively small part of the rural population, a limited resettlement (to new villages, not concentration camps) seems to have worked. At any rate, it has been said that after the fighting was over, few of the resettled villagers wanted to return to their former homes.22 That is not a sufficient criterion of moral success, but it is one sign of a permissible program. Since governments are generally thought to be entitled to resettle (relatively small numbers of) their own citizens for the sake of some commonly accepted social purpose, the policy cannot be ruled out altogether in time of guerrilla war. But unless the numbers are restricted, it will be difficult to make the case for common acceptance. And here, as in peacetime, there is some requirement to provide adequate economic support and comparable living space. In Vietnam, that was never possible. The scope of the war was too wide; new villages could not be built; the camps were dismal; and hundreds of thousands of displaced peasants crowded into the cities, forming there a new lumpen proletariat, miserable, sick, jobless, or quickly exploited in ill-paid and menial jobs or as servants, prostitutes, and so on.
Even had all this worked, in the limited sense that civilian deaths had been avoided, the rules of engagement and the policy they embodied could hardly be defended. It seems to violate even the principle of proportionality—which is by no means easy to do, as we have seen again and again, since the values against which destruction and suffering are to be measured are so readily inflated. But in this case, the argument is clear, for the defense of resettlement comes down finally to a claim something like that made by an American officer with reference to the town of Ben Tre: we had to destroy the town in order to save it.23 In order to save Vietnam, we had to destroy the rural culture and the village society of the Vietnamese. Surely the equation does not work and the policy cannot be approved, at least in the context of the Vietnamese struggle itself. (One can always shift, I suppose, to the higher mathematics of international statecraft.)
But the rules of engagement raise a more interesting question. Suppose that civilians, duly warned, not only refuse to expel the guerrillas but also refuse to leave themselves. Can they be attacked and killed, as the rules imply? What are their rights? They can certainly be exposed to risks, for battles are likely to be fought in their villages. And the risks they must live with will be considerably greater than those of conventional combat. The increased risk results from the intimacies I have already described; I would suggest now that it is the only result of those intimacies, at least in the moral realm. It is serious enough. Anti-guerrilla war is a terrible strain on conventional troops, and even if they are both disciplined and careful, as they should be, civilians are certain to die at their hands. A soldier who, once he is engaged, simply fires at every male villager between the ages of fifteen and fifty (say) is probably justified in doing so, as he would not be in an ordinary firefight. The innocent deaths that result from this kind of fighting are the responsibility of the guerrillas and their civilian supporters; the soldiers are cleared by the doctrine of double effect. It has to be stressed, however, that the supporters themselves, so long as they give only political support, are not legitimate targets, either as a group or as distinguishable individuals. Conceivably, some of them can be charged with complicity (not in guerrilla war generally but) in particular acts of assassination and sabotage. But charges of that sort must be proved before some sort of judicial tribunal. So far as combat goes, these people cannot be shot on sight, when no firefight is in progress; nor can their villages be attacked merely because they might be used as firebases or because it is expected that they will be used; nor can they be randomly bombed and shelled, even after warning has been given.
The American rules have only the appearance of recognizing and attending to the combatant/noncombatant distinction. In fact, they set up a new distinction: between loyal and disloyal, or friendly and hostile noncombatants. The same dichotomy can be seen at work in the claims American soldiers made about the villages they attacked: “This place is almost entire V.C. controlled, or pro-V.C.” “We consider just about everyone here to be a hard-core V.C., or at least some kind of supporter.”24 It is not the military activities of the villagers that are being stressed in statements of this sort, but their political allegiance. Even with reference to that, the statements are palpably false, since at least some of the villagers are children who cannot be said to have any allegiance at all. In any case, as I have already argued in the example of the villagers of occupied France, political hostility does not make people enemies in the sense of the war convention. (If it did, there would be no civilian immunity at all, except when wars were fought in neutral countries.) They have done nothing to forfeit their right to life, and that right must be respected as best it can be in the course of attacks against the irregular fighters the villagers both resemble and harbor.
It is important to say something now about the possible shape of those attacks, though I cannot talk about them like a military strategist; I can only report on some of the things that strategists say. Bombing and shelling from a distance have undoubtedly been defended in terms of military necessity. But that is as bad an argument strategically as it is morally. For there are other and more effective ways of fighting. Thus a British expert on counter-insurgency writes that the use of “heavily armed helicopters” against peasant villages “can only be justified if the campaign has deteriorated to the extent where it is virtually indistinguishable from conventional war.”25 I doubt that it can be justified even then, but I want to stress again what this expert has grasped: that counter-insurgency requires a strategy and tactics of discrimination. Guerrillas can be defeated (and, similarly, they can win) only at close quarters. With regard to peasant villages, this suggests two different sorts of campaigns, both of which have been extensively discussed in the literature. In areas of “low intensity operations,” the villages must be occupied by small units specially trained for the political and police work necessary to seek out guerrilla supporters and informants. In areas where the guerrillas are effectively in control and the fighting intense, the villages must be encircled and entered in force. Bernard Fall has reported in some detail on a French attack of this sort in Vietnam in the 1950s.26 What is involved here is an effort to bring numbers, expertise, and technology directly to bear, forcing the guerrillas to give battle in a situation where fire can be relatively precise, or driving them into a surrounding net of soldiers. If the soldiers are properly prepared and equipped, they need not accept unbearable risks in fighting of this sort, and they need not inflict indiscriminate destruction. As Fall points out, a very considerable number of men are required for this strategy: “No sealing off of an enemy force could be successful unless the proportion of attackers to defenders was 15 to 1 or even 20 to 1, for the enemy had in its favor an intimate knowledge of the terrain, the advantages of defensive organization, and the sympathy of the population.” But these proportions are frequently achieved in guerrilla war, and the “surround and storm” strategy would be eminently feasible were it not for a second and more serious difficulty.
Since the villages are not (or should not be) destroyed when they are stormed, and since the villagers are not resettled, it is always possible for the guerrillas to return once the specially assembled task force has moved on. Success requires that the military operation be followed by a political campaign—and this neither the French in Vietnam nor the Americans who followed them were able to mount in any serious fashion. The decision to destroy villages from a distance was a consequence of this failure, which is not at all the same thing as the “deterioration” of guerrilla into conventional war.
At some point in the military progress of the rebellion, or in the decline of the political capacity of the government that opposes it, it may well become impossible to fight the guerrillas at close quarters. There aren’t enough men or, more likely, the government, though it can win particular battles, has no staying power. As soon as the fighting is over, the villagers welcome back the insurgent forces. Now the government (and its foreign allies) face what is in effect, or rather what has become, a people’s war. This honorific name can be applied, however, only after the guerrilla movement has won very substantial popular support. It is by no means true all the time. One need only study Che Guevara’s abortive campaign in the jungles of Bolivia to realize how easy it is to destroy a guerrilla band that has no popular support at all.27 From there, one might trace a continuum of increasing difficulty: at some point along that continuum, guerrilla fighters acquire war rights, and at some further point, the right of the government to continue the struggle must be called into question.
This last is not a point which soldiers are likely to recognize or acknowledge. For it is an axiom of the war convention (and a qualification on the rules of war) that if attack is morally possible, counter-attack cannot be ruled out. It cannot be the case that guerrillas can hug the civilian population and make themselves invulnerable. But if it is always morally possible to fight, it is not always possible to do whatever is required to win. In any struggle, conventional or unconventional, the rules of war may at some point become a hindrance to the victory of one side or another. If they could then be set aside, however, they would have no value at all. It is precisely then that the restraints they impose are most important. We can see this clearly in the Vietnam case. The alternative strategies I have briefly outlined were conceivably a way of winning (as the British won in Malaya) until the guerrillas consolidated their political base in the villages. That victory effectively ended the war. It is not, I suppose, a victory that can be distinguished in any definitive fashion from the political and military struggle that preceded it. But one can say with some assurance that it has occurred whenever ordinary soldiers (who are not moral monsters and would fight by the rules if they could) become convinced that old men and women and children are their enemies. For after that, it is unlikely that the war can be fought except by setting out systematically to kill civilians or to destroy their society and culture.
I am inclined to say more than this. In the theory of war, as we have seen, considerations of jus ad bellum and jus in bello are logically independent, and the judgments we make in terms of one and the other are not necessarily the same. But here they come together. The war cannot be won, and it should not be won. It cannot be won, because the only available strategy involves a war against civilians; and it should not be won, because the degree of civilian support that rules out alternative strategies also makes the guerrillas the legitimate rulers of the country. The struggle against them is an unjust struggle as well as one that can only be carried on unjustly. Fought by foreigners, it is a war of aggression; if by a local regime alone, it is an act of tyranny. The position of the anti-guerrilla forces has become doubly untenable.