PROLOGUE

Modern armed forces continue to be structured for large-scale war, but advanced societies whose small families lack expendable children have a very low tolerance for casualties. Even supposedly warlike Americans gravely count casualties in Iraq that in three years have yet to reach 3,000—fewer than were lost in many a single day of battle in past wars. Fortunately, this refusal to spill the blood needed to fuel battles diminishes the likelihood that advanced societies will deliberately set out to fight one another (pas des enfants, pas des Suisses, pas de guerre) unless they are somehow able to convince themselves that a war could be entirely or very largely aerial and naval. Such wars, however, are difficult to imagine, except when islands are involved, as in a China-Taiwan war, which is very improbable for its own reasons. Air and naval forces can certainly be employed advantageously against any less advanced enemy incautious enough to rely on a conventional defense, conducted by regular forces, but in that context as well there must be severe doubts about the continued usefulness of the ground forces of advanced countries that are intolerant of casualties. It is easy enough to blockade the enemy, to successfully bomb all the right nodal points and shut down electrical, transportation, and communications networks. Air strikes can disable runways and destroy both sheltered and unsheltered aircraft, ballistic missiles, and nuclear installations. Air power can also sink warships, or rout any mechanized forces deployed in the open, as the United States did with Iraq in 1991 and partly in 2003, and as it could do with Iran. No real role would remain for ground forces except to dislodge the enemy from any territory he had occupied, or to occupy his own territory. That, however, is bound to cost casualties that might not be tolerated; it is also bound to provoke an insurgency.

In that event, naval forces cannot do much, because insurgencies rarely have an important maritime dimension (the Sri Lanka case is an exception) and riverine operations are usually minor. Air forces can have surveillance and transport roles, but insurgents rarely present targets of sufficient stability and sufficient contrast to be identified, designated, and effectively attacked from the air. That leaves almost everything to the ground forces, and when the advanced attack the less advanced, the more advanced forces will have large advantages in firepower, mobility, and operational coherence. But they will also have no visible enemy to fight, so that the normal operational methods and tactics of conventional warfare cannot be applied. True, there are the alternative methods and tactics of counterinsurgency warfare, but do they actually work? Insurgents do not always win, but their defeats can rarely be attributed to counterinsurgency warfare, as we shall see.

THE THEORY OF COUNTERINSURGENCY WARFARE

Two distinguished American generals of exceptional intelligence, James N. Mattis of the Marine Corps and David H. Petraeus of the Army, each now responsible for the training and doctrine policy of his own service, have recently circulated the text of a new “counterinsurgency” field manual, FM 3-24 DRAFT, which they propose for official use. Its doctrines emerge from the chapter titles. After a first chapter of definitions (which any military manual must have, because the battlefield is no place for semantic debate) we come to the first substantive chapter, “Unity of Effort: Integrating Civil and Military Activities,” in which the authors duly recognize and strongly emphasize the essentially political nature of the struggle against insurgents. That is hardly an original discovery, as the two generals and their staffs would be the first to recognize, yet it is still necessary to affirm what should be obvious, because amid the frustrations of fighting a mostly invisible enemy, it is hard to resist the tempting delusion that some clever new tactics, or even some clever new technology, can defeat the insurgents.

Much more questionable is the proposition that follows, which is presented as self-evident, that a necessary if not sufficient condition of victory is to provide what the insurgents cannot: basic public services, physical reconstruction, the hope of economic development and social amelioration. The hidden assumption here is that there is only one kind of politics in this world, a politics in which popular support is important or even decisive, and that such support can be won by providing better government. Yet the extraordinary persistence of dictatorships as diverse in style as the regimes of

Cuba, Libya, North Korea, and Syria shows that in fact government needs no popular support as long as it can secure obedience. As for better government, that is certainly wanted in France, Norway, or the United States, but obviously not in Afghanistan or Iraq, where many people prefer indigenous and religious oppression to the freedoms offered by foreign invaders.

The very word “guerrilla,” which now refers only to a tactic, was first used to describe the ferocious insurgency of the illiterate Spanish poor against their would-be liberators, under the leadership of their traditional oppressors. On July 6, 1808, King Joseph of Spain presented a draft constitution that for the first time in Spain’s history offered an independent judiciary, freedom of the press, and the abolition of the remaining feudal privileges of the aristocracy and of the Church. At that time, abbeys, monasteries, and bishops still owned every building and every piece of land in 3,148 towns and villages, which were inhabited by some of Europe’s most wretched tenants. Despite the fact that the new constitution would have liberated them and let them keep their harvests for themselves, the Spanish peasantry failed to rise up in its support. Instead, they obeyed the priests, who summoned them to fight against the ungodly innovations of the foreign invader. For Joseph was the brother of Napoleon Bonaparte, placed on the Spanish throne by French troops. That was all that mattered to most Spaniards—not what was proposed but by whom it was proposed.

By then the French should have known better. In 1799 the same thing had happened in Naples, whose liberals, supported by the French, were slaughtered by the very peasants and plebeians they wished to emancipate. They were mustered into a militia of the “Holy Faith” by Cardinal Fabrizio Ruffo, coincidentally a member of Calabria’s largest land-owning family, who led his men forward on horseback. Ruffo easily persuaded his followers that all promises of material betterment were irrelevant, because the real aim of the French and the liberals was to destroy the Catholic religion in the service of Satan. Spain’s clergy did the same, and their illiterate followers could not know that the very first clause of Joseph’s draft constitution had not only recognized the Roman Apostolic Catholic Church but stated that it was the only one allowed in Spain.

The same kind of politics are now in evidence in Afghanistan and Iraq, including the ineffectual enshrinement of Islam in the new Iraqi constitution, and the emergence of clerical warlords who are as ready to use violence as Cardinal Ruffo was. Since the 2003 invasion, both Shiite and Sunni clerics have been repeating over and over again that the Americans and their “Christian” allies have come to Iraq to destroy Islam in its cultural heartland and to steal the country’s oil. The clerics dismiss all talk of democracy and human rights by the invaders as mere hypocrisy—with the exception of women’s rights, which the clerics say are only propagandized to persuade Iraqi daughters and wives to dishonor their families by imitating the shameless nakedness and impertinence of Western women.

The vast majority of Afghans and Iraqis naturally believe their religious leaders. The alternative would be to believe what for them is entirely unbelievable: that foreigners are unselfishly expending blood and treasure in order to help them. They themselves would never invade a foreign country except to plunder it, the way Iraq invaded Kuwait, thus having made Saddam Hussein genuinely popular for a time when troops brought back their loot. As many opinion polls and countless incidents demonstrate, the Americans and their allies are widely considered to be the worst of invaders, who came to rob Muslim Iraqis not only of their territory and oil but also of their religion and even their family honor. Many Muslims around the world believe as much, even in Turkey, whose most successful recent film depicted an American Jewish military doctor who was operating on Iraqis not to save their lives but to remove their kidneys, which of course he was sending back to the U.S. for transplantation and his personal profit (he was Jewish after all). It is the same in Afghanistan, where the American-imposed quota of women parliamentarians has caused widespread resentment, not least because most Afghans are scandalized by the spectacle of a woman contradicting a man in public—as in, for example, televised parliamentary debates.

In other words, “Integrating Civil and Military Activities” to improve local conditions need not gain public support. And even if it did, it does not automatically follow that such support would be decisive, or even important.

Next comes a very long section on “Intelligence in Counterinsurgency,” which reflects the crucial predicament of counterinsurgency warfare: the unseen enemy, who can choose when to emerge from civilian cover to launch his attacks, and increasingly can attack by remote control, reducing his exposure or avoiding exposure altogether. Everywhere outgunned, and in Iraq if not Afghanistan outnumbered as well, the insurgents would be easily defeated if their invisibility could be stripped away. That much is obvious. But the authors then automatically assume that it is simply an intelligence problem to identify the insurgents among the population. This is another very questionable proposition, as we shall see, because in fact it is a political problem, which always has a political solution, however unpalatable that may be.

In any case, proceeding on their general premise, the detailed headings that follow point to different ways of overcoming the invisibility of insurgents by using all possible intelligence sources, methods, and assets. The chapter entitled “Intelligence Characteristics in Counterinsurgency” points to the need to have rather different skills and talents when the targets are extremely “low-contrast” insurgents, as opposed to high-contrast targets such as airfields or warships. Next comes “Predeployment Planning and Intelligence Preparation of the Battlefield,” a section that emphasizes the specificity of counterinsurgency as to time, place, population, culture, and more, as opposed to the general-purpose intelligence preparations for regular war. For example, it is useful to have trained Arabic-speaking interrogators if one plans to invade an Arab-speaking country, and, at a less elementary level, it is useful to have some cultural instruction before trying to analyze the behavior of the locals.

After that comes “Intelligence, Surveillance, and Reconnaissance Operations,” discussing, among other things, the different ways of using regular forces, with their regular platforms and sensors, to find the elusive insurgents. These methods may or may not work but certainly entail the use of ultrasophisticated and very expensive F-15s and F-18s, with the most advanced sensors to detect and track the man, the boy, and the donkey who may or may not be transporting an “improvised explosive device” to its intended emplacement. Then comes the delicate subject of “Counterintelligence and Counterreconnaissance.” The second of these is straightforward enough: before attacking a target, it is usually essential for insurgents to observe it in order to plan the action, and with a bit of luck such observers can be spotted by alert defenders. Often the best way of protecting potential targets is to anticipate attacks against them by countersurveillance (or “counterreconnaissance”), then to pre-empt or ambush the attackers. What is delicate is the “Counterintelligence” part, which points to the likelihood of insurgent penetrations of the local forces that are supposedly fighting the insurgency—the friends and allies in need that are being provided with training, weapons, and money. Penetrations always occur, even in the best of military forces and intelligence services, but there is a difference in scale between the consequences of a traitor or two and wholesale enlistments to serve the enemy cause (or the inadvertent recruitment of enemies into the ranks). For example, it must be universally recognized by now that in Iraq many if not most of the Shiites in the army and police force are actually under the orders of one or another of the Shiite militias, including the “Mahdi army,” that occasionally launch surprise attacks on American or British forces. Equally, many of the Kurds who are paid by the government or by the Americans directly are in the service of either Massoud Barazani or Jalal Talebani, the two chieftains who trade under the labels of the Kurdistan Democratic Party and the Patriotic Union of Kurdistan. (Talebani, moreover, is now Iraq’s president.) Far more dangerously, the Sunnis in the army and police force who have been recruited, trained, equipped, and salaried to fight the Sunni insurgents are just as likely to help the insurgency, or even to be insurgents themselves, on temporary detachment to the government forces, so to speak. That is the only way that Sunnis, whose families live among the Sunni population, can both receive a salary and also keep their families alive. But again, this is really a political problem, which has an unpalatable solution that is certainly more reliable than “Counterintelligence” could ever be.

Later comes “Intelligence Collaboration and Fusion,” which refers to the sharing and integration of intelligence coming in separate channels from different services, branches, and individual sources. Such cooperation is characteristic of regular war operations as well, but it is more critical when the targets are almost always unstable, elusive, and lowcontrast—if even identifiable at all, as they rarely are. What follows are more predictable chapters on “Designing Counterintelligence Operations” and executing them, and developing “Host Nation” security forces, the relative brevity of these chapters underlining the emphasis the authors place on intelligence operations. Recent scandals readily explain the need for an entire chapter on “Leadership and Ethics for Counterinsurgency,” and the chapter on “Building and Sustaining Capability and Capacity,” with a subsection on counterinsurgency logistics, also reflects the unhappy experience of Iraq.

Ammunition supply—along with fuel, usually the biggest item in regular operations budgets—is relatively unimportant, since very little ammunition is expended fighting an insurgency, yet logistics still present difficulties in Iraq because supplies cannot simply be trucked from A to B without a high risk of destructive attacks by insurgents, hijacking by militias in need of supplies, simple highway robbery, or opportunistic looting by ordinary civilians. (There are many natural predators in the Iraqi population, probably because of the high proportion of ex-nomads and their direct descendants, for whom the razzia is still an honorable and manly tradition.) Truck convoys can be more secure if rather less efficient, but their safety depends on the quality of their escorts, which are necessarily very scarce and expensive if they are U.S. troops (the British, near the seaport, as is their wont, are more easily supplied), less scarce but even more expensive if they are security contractors, and very cheap but extremely unreliable if they are Iraqi soldiers or police. The end result is that hyper-expensive helicopter hours are very often used to carry even low-value, non-urgent supplies, which is one reason why the occupation costs so much even though very little ordnance is being expended.

Incidentally, in spite of all the advances in “jointness,” when the U.S. Air Force tried to help out the Army (and dissuade it from acquiring its own intra-theater fixed-wing aircraft) by providing a regular shuttle service of C-130 turboprop transports between Kuwait and different bases in Iraq, the aircraft ended up carrying mostly “sailboat fuel”—that is, they flew empty because it was easier for Army formations to use their own helicopters and smaller fixed-wing aircraft than to “interface” with the Air Force.

Two more things can be noted about the new field manual before turning to the peculiar politics of insurgency and counterinsurgency. There is an Appendix to the chapter on “Predeployment Planning and Intelligence Preparation of the Battlefield” that further emphasizes the subject’s importance and includes a section on “Linguistic Support.” It is a sad story indeed when the astonishing linguistic incapacity of U.S. military forces and intelligence organizations is contrasted with the abundance of American civilians who speak all known foreign languages, and the brilliant record of foreign-language education in the U.S. Army and Navy, which used to produce as many good Chinese and Japanese speakers as they wanted by selecting for natural aptitude in the recruit pool, giving them a year of intensive courses (eight hours a day, six days a week), and quickly sending away those who failed to keep up with their classes. Nothing prevents the military from doing the same for Arabic, Persian and, say, Azeri now, except for an unwillingness to invest in the future, and probably a lack of disciplined volunteers willing to learn a language eight hours a day, six days a week, for a whole year or more.

FM 3-24 DRAFT ends with a list of suggested readings, and one of the first books on the list is Small Wars: A Tactical Handbook for Imperial Soldiers (1890), by Charles E. Calwell. The previous counterinsurgency manual, FM 3-07.22, also had such a list, and its first suggested reading was The Battle of the Casbah: Terrorism and Counterterrorism in Algeria, 1955‒57, by Paul Aussaresses. Is it therefore the case that counterinsurgency doctrine has been evolving backward, from the doubts of the 1950s to the certitudes of 1890? That is no accusation, alas, because one needs to go back even further to find convincing models of success in defeating insurgents by military means.

COUNTERINSURGENCY IN PRACTICE: IRAQ

We begin with some elementary observations. The armed forces of the most advanced countries, and certainly of the United States, all formidable against enemies assembled in conveniently targetable massed formations, are least effective in fighting insurgents. That was demonstrated in Vietnam in many different ways over many years, even as the occasional North Vietnamese regular unit that ventured to fight conventionally was efficiently destroyed. The same two-part proposition is unnecessarily being proven all over again in Iraq, damaging the reputation of the United States for wisdom and strength, misusing fine soldiers, wasting vast amounts of money on skillful but ineffectual air and ground operations, inflicting added suffering on Iraqis at large, and taking the lives of young Americans whose sacrifice, one fears, will be deemed futile.

There is no mystery about the first part of the proposition. Because of their abundant resources and all-round competence, morale, discipline, and skills, the armed forces of the United States on the largest scale, and of other advanced countries on their scale, can usually generate much more firepower than their antagonists. These days, moreover, they can do so with routine precision because of sensors that reveal targets even in poor visibility; platforms and weapons that can reach targets at any planetary range; accurate guidance and homing devices; and command and communication networks that combine all those abilities. Up to a point, the second part of the proposition is merely the logical consequence of the first: faced with especially superior firepower, insurgents strive to be especially elusive—more so than if they were facing less formidable regular forces—and as targets diminish, so does the value of firepower.

But there is much more to it than that. Specifically, there is the matter of politics, on both sides. Unless insurgents confine their operations to thoroughly deserted areas where there is no one to observe them, they must have at least the passive cooperation of local inhabitants. Whether they fail to report the insurgents to the authorities out of sympathy for their cause or in terror of their vengeance is entirely irrelevant. In either case, the insurgents are in control of the population around them, and not the authorities. That essentially political advantage is enough to allow motivated insurgents to overcome all manner of tactical weaknesses in combat skills and weapons.

As in so many previous cases, in a manner abundantly familiar from previous insurgencies, that political situation is now playing out in Iraq, where insurgents live very safely in Sunni neighborhoods, towns, and villages, emerging to place bombs or launch attacks when and where it suits them before resuming innocuous civilian identities once again. Local insurgents may indeed pass unobserved by their neighbors when inactive, but not when they take up weapons and gather for operations, while the foreign volunteers among them necessarily attract attention even when they carry no weapons because of their distinct speech and manner. Many of the local inhabitants certainly know who the insurgents are and where they keep their stores of explosives and weapons, but they are not telling. That is why U.S. Army and Marine patrols cannot find insurgents unless they choose to reveal themselves by engaging in direct combat, which of course they rarely do, and only when they think that they have a great advantage. The mostly futile American patrols therefore expose soldiers to the mines, remote-controlled explosives, snipers, and mortar bombs that inflict daily casualties.

Naturally, every form of technical intelligence and every possible sensor is being employed to supplant the lack of very elementary but indispensable human intelligence, including synthetic-aperture radars aboard big four-engine aircraft and the infrared and video sensors of the latest targeting pods on two-seat heavyweight jet fighters. The expense of these flights alone is huge, amounting to hundreds of millions of dollars a month, but the results are very meager. The aim, of course, is to gather immediately actionable imagery, especially at night, showing such things as insurgents placing side bombs alongside U.S. patrol routes or approaching oil pipelines bearing explosives. Failing that, it is at least hoped that possible insurgent activities could be detected for further investigation; for example, people furtively bringing things to isolated buildings at night. But in practice, unless insurgents carry recognizable weapons, it is simply impossible to differentiate between them and innocent people going about their peaceful business. In the meantime, very elaborate equipment that is very costly to operate, and very effective in identifying armored vehicles, bunkers, missile launchers, and any other readily recognizable target of classic form, is still being employed every day in futile attempts to detect deliveries of a few dollars of food, or the emplacement of readily improvised explosive devices. This too is an aspect of the structural unsuitability of modern armed forces to fight elusive enemies that present no stable targets.

The essentially political advantage of the insurgents in commanding at least the silence of the local population cannot be overcome by technical means no matter how advanced. Nor can the better operational methods and tactics advocated in FM 3-24 DRAFT be of much help. So few of the insurgents ever engage in direct combat, so much of the insurgency takes covert forms, ranging from the infiltration of the government to bombings, sabotage, and assassinations, that

the tactical defeats inflicted on the insurgents—including the killing of their top leaders and heroes—have no perceptible impact on the volume of the violence, and of its political consequences.

In Iraq, as noted, there is supposed to be a far better way of finding insurgents than patrols driving about or sensors, howsoever sophisticated: the Iraqi police and army. Their recruitment, training, equipment, and upkeep, a very costly enterprise in both money and blood, has also yielded meager results because the politics of the situation are again central, and again unfavorable. It is easy to recruit local auxiliaries in any poor country, and any number of Shiites, Kurds, and Sunni Arabs can readily be recruited—in recent decades that is how many of them made a living, by exploiting their privileged access as Sunnis and Arabs to prized military and police salaries. Other jobs were much less desirable, because they required work, and now, in any case, they are very scarce. But while they are willing to wear the uniforms and accept training up to a point, Sunni Arabs are naturally disinclined to help capture or kill insurgents who are fighting to restore the Sunni Arab ascendancy over Iraq. Besides, their families would be in deadly peril if they were suspected of loyalty to their government, and by extension to the Americans. Some of those policemen and soldiers know much about the insurgents and where exactly they might be found, but are still of no help in finding them, precisely because they are insurgents themselves. Even if specifically ordered into action on those rare occasions in which there is overt combat, most Sunni Arab policemen and soldiers will not fight the insurgents; if they cannot simply stand back quietly, they are apt to desert, usually with their weapons. As for army and police units manned mostly by Shiite Arabs or Kurds, they are not actively disloyal, but they cannot gather information on the insurgents either. Sunni Arab civilians will not confide in them any more than in Americans, and perhaps less, because sooner or later the Americans will leave Iraq, but the Kurds and Shiites will not, and they are therefore the greater enemy.

The adverse political terrain of counterinsurgency is simply a given in Iraq, as it is everywhere else, for if insurgents do not receive, or cannot forcibly exact, at least the passive collaboration of the population at large, they normally cannot survive at all.

THE EASY AND RELIABLE WAY OF DEFEATING ALL INSURGENCIES EVERYWHERE

Perfectly ordinary regular armed forces, with no counterinsurgency doctrine or training whatever, have in the past regularly defeated insurgents, by using a number of well-proven methods. It is enough to consider these methods to see why the armed forces of the United States or of any other democratic country cannot possibly use them.

The simple starting point is that insurgents are not the only ones who can intimidate or terrorize civilians. For instance, whenever insurgents are believed to be present in a village, small town, or distinct city district—a very common occurrence in Iraq at present, as in other insurgency situations—the local notables can be compelled to surrender them to the authorities, under the threat of escalating punishments, all the way to mass executions. That is how the Ottoman Empire could control entire provinces with a few feared janissaries and a squadron or two of cavalry. The Turks were simply too few to hunt down hidden rebels, but they did not have to: they went to the village chiefs and town notables instead, to demand their surrender, or else. A massacre once in a while remained an effective warning for decades. So it was mostly by social pressure rather than brute force that the Ottomans preserved their rule: it was the leaders of each ethnic or religious group inclined to rebellion that did their best to keep things quiet, and if they failed, they were quite likely to tell the Turks where to find the rebels before more harm was done.

Long before the Ottoman Empire, the Romans knew how to combine sticks and carrots to obtain obedience and suppress insurgencies. Conquered peoples too proud to accept the benefits of their rule, from public baths and free circus shows to reliable law courts, were “de-bellicized” (a very Roman idea). It was done by killing all who dared to resist in arms—it made good combat practice for the legions—by selling into slavery any who were captured in battle, by leveling towns that held out under siege instead of promptly surrendering, and by readily accepting as peaceful subjects and future citizens all who submitted to Roman rule. In the first two and most successful centuries of imperial Rome, some 300,000 soldiers in all, only half of them highly trained legionary troops, were enough to secure a vast empire that stretched well beyond the Mediterranean basin that formed its core, today the territory of some thirty European, Middle Eastern, and North African states. The Romans could not disperse their soldiers in hundreds of cities, thousands of towns, and countless hamlets to repress riot or rebellion; the troops were needed to guard the frontiers. Instead, they relied on deterrence, which was periodically reinforced by exemplary punishments. Most inhabitants of the empire never rebelled after their initial conquest. A few tribes and nations had to be reconquered after trying and failing to overthrow Roman rule. A few simply refused to become obedient, and so they were killed off: “They make a wasteland and call it peace” was the bitter complaint of a Scottish chieftain (as reported by Tacitus).

Terrible reprisals to deter any form of resistance were standard operating procedure for the German armed forces in the Second World War, and very effective they were in containing resistance with very few troops. As against all the dramatic films and books that describe the heroic achievements of the resistance all over occupied Europe, military historians have documented the tranquillity that the German occupiers mostly enjoyed, and the normality of collaboration, not merely by notorious traitors such as the incautious French poet or the failed Norwegian politician but by vast numbers of ordinary people. Polish railwaymen, for example, secured the entire sustenance of the German eastern front. As for the daring resistance attacks that feature in films, they did happen occasionally, but not often, and not because of any lack of bravery in fighting the routinely formidable Germans but because of the terrible punishments they inflicted on the population.

Occupiers can thus be successful without need of any specialized counterinsurgency methods or tactics if they are willing to out-terrorize the insurgents, so that the fear of reprisals outweighs the desire to help the insurgents or their threats. The Germans also established secure and economical forms of occupation by exploiting isolated resistance attacks to achieve much broader demonstration effects. Lone German dispatch riders were easily toppled by tensed wires or otherwise intercepted and killed, but then troops would arrive on the scene to burn or demolish the surrounding buildings or farms or the nearest village, seizing and killing anyone who aroused suspicion or just happened to be there. After word of the terrible deeds spread and was duly exaggerated, German dispatch riders could safely continue on their way, until reaching some other uninstructed part of the world, where the sequence would have to be repeated.

Likewise in the Vietnam War, the Viet Cong and North Vietnamese were skilled in using terror to secure their pervasive territorial control and very ready to use any amount of violence against civilians, from countless individual assassinations to mass executions, as in Hue in 1968. The Communist cause had its enthusiasts, “fellow travelers,” and opportunistic followers, but Vietnamese who were none of the above, and not outright enemies, were compelled to collaborate actively or passively by the threat of the violence so liberally used. That is exactly what the insurgents in Iraq are now doing, and this is no coincidence. All insurgencies follow the same pattern. Locals who are not sympathetic to begin with, who cannot be recruited to the cause, are compelled to collaborate by the fear of violence, readily reinforced by the demonstrative killing of those who insist on refusing to help the resistance. Neutrality is not an option.

By contrast, the capacity of American armed forces to inflict collective punishments does not extend much beyond curfews and other such restrictions, inconvenient to be sure and perhaps sufficient to impose real hardship, but obviously insufficient to out-terrorize insurgents. Needless to say, this is not a political limitation that Americans would ever want their armed forces to overcome, but it does leave the insurgents in control of the population, the real “terrain” of any insurgency. Of course, the ordinary administrative functions of government can also be employed against the insurgents, less compellingly perhaps but without need of violence. Insurgents everywhere seek to prohibit any form of collaboration or contact with the authorities, but they cannot normally prevent civilians from entering government offices to apply for obligatory licenses, permits, travel documents, and such. That provides venues for intelligence officers on site to ask applicants to provide information on the insurgents, in exchange for the approval of their requests and perhaps other rewards. This effective and straightforward method has been widely used, and there is no ethical or legal reason why it should not be used by the armed forces of the United States as well. But it does require the apparatus of military government, complete with administrative services for civilians. During and after the Second World War, after very detailed preparations, the U.S. Army and Navy governed the American zone of Germany, all of Japan, and parts of Italy. Initially, U.S. officers were themselves the administrators, with such assistance from local officials they chose to re-employ. Since then, however, the United States has preferred both in Vietnam long ago and now in Iraq to leave government to the locals.

That decision reflects another kind of politics, manifest in the ambivalence of a United States government that is willing to fight wars, that is willing to start wars because of future threats, that is willing to conquer territory or even entire countries, and yet is unwilling to govern what it conquers, even for a few years. Consequently, for all the real talent manifest in the writing of FM 3-24 DRAFT, its prescriptions are in the end of little or no use and amount to a kind of malpractice. All its best methods, all its clever tactics, all the treasure and blood that the United States has been willing to expend, cannot overcome the crippling ambivalence of occupiers who refuse to govern, and their principled and inevitable refusal to out-terrorize the insurgents, the necessary and sufficient condition of a tranquil occupation.