Book Excerpt: Underground Tactics

Featured image: power line sabotaged by the African National Congress

Editor’s note: The following is from the chapter “Tactics and Targets” of the book Deep Green Resistance: A Strategy to Save the  Planet. This book is now available for free online.

     by Aric McBay

Some tactics can be carried out underground—like general liberation organizing and propaganda—but are more effective aboveground. Where open speech is dangerous, these types of tactics may move underground to adapt to circumstances. The African National Congress, in its struggle for basic human rights, should have been allowed to work aboveground, but that simply wasn’t possible in repressive apartheid South Africa.

And then there are tactics that are only appropriate for the underground, obligate underground operations that depend on secrecy and security. Escape lines and safehouses for persecuted persons and resistance fugitives are example of those operations. There’s a reason it’s called the “underground” railroad—it’s not transferable to the aboveground, because the entire operation is completely dependent on secrecy. Clandestine intelligence gathering is another case; the French Resistance didn’t gather enemy secrets by walking up to the nearest SS office and asking for a list of their troop deployments.

Some tactics are almost always limited to the underground:

  • Clandestine intelligence
  • Escape
  • Sabotage and attacks on materiel
  • Attacks on troops
  • Intimidation
  • Assassination

As operational categories, intelligence and escape are pretty clear, and few people looking at historical struggles will deny the importance of gathering information or aiding people to escape persecution. Of course, some abolitionists in the antebellum US didn’t support the Underground Railroad. And many Jewish authorities tried to make German Jews cooperate with registration and population control measures. In hindsight, it’s clear to us that these were huge strategic and moral mistakes, but at the time it may only have been clear to the particularly perceptive and farsighted.

Sabotage and attacks on materiel are overlapping tactics. Oftentimes, sabotage is more subtle; for example, machinery may be disabled without being recognized as sabotage. Attacks on materiel are often more overt efforts to destroy and disable the adversary’s equipment and supplies. In any case, they form an inclusive continuum, with sabotage on the more clandestine end of the scale.

It’s true that harm can be caused through sabotage, and that sabotage can be a form of violence. But allowing a machine to operate can also be more violent than sabotaging it. Think of a drift net. How many living creatures does a drift net kill as it passes through the ocean, regardless of whether it’s being used for fishing or not? Destroying a drift net—or sabotaging a boat so that a drift net cannot be deployed—would save countless lives. Sabotaging a drift net is clearly a nonviolent act. However, you could argue conversely that not sabotaging a drift net (provided you had the means and opportunity) is a profoundly violent act—indeed, violent not just for individual creatures, but violent on a massive, ecological scale. The drift net is an obvious example, but we could make a similar (if longer and more roundabout) argument for most any industrial machinery.

You’re opposed to violence? So where’s your monkey wrench?

Sabotage is not categorically violent, but the next few underground categories may involve violence on the part of resisters. Attacks on troops, intimidation, assassination, and the like have been used to great effect by a great many resistance movements in history. From the assassination of SS officers by escaping concentration camp inmates to the killing of slave owners by revolting slaves to the assassination of British torturers by Michael Collins’s Twelve Apostles, the selective use of violence has been essential for victory in a great many resistance and liberation struggles.

Attacks on troops are common where a politically conscious population lives under overt military occupation. In these situations, there is often little distinction between uniformed militaries, police, and government paramilitaries (like the Black and Tans or the miliciens). The violence may be secondary. Sometimes the resistance members are trying to capture equipment, documents, or intelligence; how many guerrillas have gotten started by killing occupying soldiers to get guns? Sometimes the attack is intended to force the enemy to increase its defensive garrisons or pull back to more defended positions and abandon remote or outlying areas. Sometimes the point is to demonstrate the strength or capabilities of the resistance to the population and the occupier. Sometimes the point is actually to kill enemy soldiers and deplete the occupying force. Sometimes the troops are just sentries or guards, and the primary target is an enemy building or facility.

Of course, for these attacks to happen successfully, they must follow the basic rules of asymmetric conflict and general good strategy. When raiding police stations for guns, the IRA chose remote, poorly guarded sites. Guerrillas like to go after locations with only one or two sentries, and any attack on those small sites forces the occupier to make tough choices: abandon an outpost because it can’t be adequately defended or increase security by doubling the number of guards. Either benefits the resistance and saps the resources of the occupier.

And although in industrial conflicts it’s often true that destroying materiel and disrupting logistics can be very effective, that’s sometimes not enough. Take American involvement in the Vietnam War. The American cost in terms of materiel was enormous—in modern dollars, the war cost close to $600 billion. But it wasn’t the cost of replacing helicopters or fueling convoys that turned US sentiment against the war. It was the growing stream of American bodies being flown home in coffins.

There’s a world of difference—socially, organizationally, psychologically—between fighting the occupation of a foreign government and the occupation of a domestic one. There’s something about the psychology of resistance that makes it easier for people to unite against a foreign enemy. Most people make no distinction between the people living in their country and the government of that country, which is why the news will say “America pulls out of climate talks” when they are talking about the US government. This psychology is why millions of Vietnamese people took up arms against the American invasion, but only a handful of Americans took up arms against that invasion (some of them being soldiers who fragged their officers, and some of them being groups like the Weather Underground who went out of their way not to injure the people who were burning Vietnamese peasants alive by the tens of thousands). This psychology explains why some of the patriots who fought in the French Resistance went on to torture people to repress the Algerian Resistance. And it explains why most Germans didn’t even support theoretical resistance against Hitler a decade after the war.

This doesn’t bode well for resistance in the minority world, where the rich and powerful minority live. People in poorer countries may be able to rally against foreign corporations and colonial dictatorships, but those in the center of empire contend with power structures that most people consider natural, familiar, even friendly. But these domestic institutions of power—be they corporate or governmental—are just as foreign, and just as destructive, as an invading army. They may be based in the same geographic region as we are, but they are just as alien as if they were run by robots or little green men.

Intimidation is another tactic related to violence that is usually conducted underground. This tactic is used by the “Gulabi Gang” (also called the Pink Sari Gang) of Uttar Pradesh, a state in India.4 Leader Sampat Pal Devi calls it “a gang for justice.” The Gulabi Gang formed as a response to deeply entrenched and violent patriarchy (especially domestic and sexual violence) and caste-based discrimination. The members use a variety of tactics to fight for women’s rights, but their “vigilante violence” has gained global attention. With over 500 members, they can exert considerable force. They’ve stopped child marriages. They’ve beaten up men who perpetrate domestic violence. The gang forced the police to register crimes against Untouchables by slapping police officers until they complied. They’ve hijacked trucks full of food that were going to be sold for a profit by corrupt officials. Their hundreds of members practice self-defense with the lathi (a traditional Indian stick or staff weapon). It’s no surprise their ranks are growing.

Gulabi Gang

Many of these examples tread the boundary of our aboveground-underground distinction. When struggling against systems of patriarchy that have closely allied themselves with governments and police (which is to say, virtually all systems of patriarchy), women’s groups that have been forced to use violence or the threat of violence may have to operate in a clandestine fashion at least some of the time. At the same time, the effects of their self-defense must be prominent and publicized. Killing a rapist or abuser has the obvious benefit of stopping any future abuses by that individual. But the larger beneficial effect is to intimidate other would-be abusers—to turn the tables and prevent other incidents of rape or abuse by making the consequences for perpetrators known. The Gulabi Gang is so popular and effective in part because they openly defy abuses of male power, so the effect on both men and women is very large. Their aboveground defiance rallies more support than they could by causing abusive men to die in a series of mysterious accidents. The Black Panthers were similarly popular because they publically defied the violent oppression meted out by police on a daily basis. And by openly bearing arms, they were able to intimidate the police (and other people, like drug dealers) into reducing their abuses.

There are limits to the use of intimidation on those in power. The most powerful people are the most physically isolated—they might have bodyguards or live in gated houses. They have far more coercive force at their fingertips than any resistance movement. For that reason, resistance groups have historically used intimidation primarily on low-level functionaries and collaborators who give information to those in power when asked or who cooperate with them in a more limited way.

It’s important to acknowledge the distinction between intimidation and terrorism. Terrorism consists of violent attacks on civilians. Resistance intimidation directly targets those responsible for oppressive and exploitative acts and power structures, and lets those people know that there are consequences for their actions. The reason it gets people so riled up is because it involves violence (or the threat of violence) going up the hierarchy. But resistance intimidation is ultimately, of course, an attempt to reduce violence. Groups like the Gulabi Gang beat abusive men instead of just killing them. There’s a reasonable escalation that gives men a chance to stop their wrongdoing and also makes the consequences for further wrongdoing clear. Rape and domestic abuse are terrorism; they’re senseless and unprovoked acts of violence against unarmed civilians, designed to threaten and terrorize women (and men) into compliance. The intimidation of rapists or domestic abusers is one tactic that can be used to stop their violence while employing the minimum amount of violence possible.

No resistance movement wants to engage in needless cycles of violence and retribution with those in power. But a refusal to employ violent tactics when they are appropriate will very likely lead to more violence. Many abolitionists did not support John Brown because they considered his plan for a defensive liberation struggle to be too violent—but Brown’s failure led inevitably to a lengthy and gruesome Civil War (as well as continued years of bloody slavery), a consequence that was orders of magnitude more violent than Brown’s intended plan.

This leads us to the last major underground tactic: assassination.

In talking about assassination (or any attack on humans) in the context of resistance, two key questions must be asked. First, is the act strategically beneficial, that is, would assassination further the strategy of the group? Second, is the act morally just, given the person in question? (The issue of justice is necessarily particular to the target; it’s assumed that the broader strategy incorporates aims to increase justice.)

As is shown on my two-by-two grid of all combinations, an assassination may be strategic and just, it may be strategic and unjust, it may be unstrategic but just, or it may be both unstrategic and unjust. Obviously, any action in the last category would be out of the question. Any action in the strategic and just category could be a good bet for an armed resistance movement. The other two categories are where things get complex.

Figure 13-3

Hitler exemplified a number of different strategy vs. justice combinations at different points in time. It’s a common moral quandary to ask whether it would be a good idea to go back in time and kill Hitler as a child, provided time travel were possible. There’s a good bet that this would have averted World War II and the Holocaust, which would have been a good thing, so put a check mark in the “strategic” column. The problem is that most people would consider it unjust to murder an innocent child who had yet to commit any crimes, so it would be difficult to call that action just in the immediate sense.

Once Hitler had risen to power in the late 1930s, though, his aim was clear, as he had already been whipping up hate and expanding his control of Nazi Germany. At that point, it would have been both strategic and just to assassinate him. Indeed, elements in the Wehrmacht (army) and the Abwehr (intelligence) considered it, because they knew what Hitler was planning to do. Unfortunately, they were indecisive, and did not commit to the plan. Hitler soon began invading Germany’s neighbors, and as his popularity soared, the assassination plan was shelved. It was years before inside elements would actually stage an assassination attempt.

Figure 13-4

That famous attempt took place—and failed—on July 20, 1944.5What’s interesting is that the Allies were also considering an attempt on Hitler’s life, which they called Operation Foxley. They knew that Hitler routinely went on walks alone in a remote area, and devised a plan to parachute in two operatives dressed as German officers, one of them a sniper, who would lay in wait and assassinate Hitler when he walked by. The plan was never enacted because of internal controversy. Many in the SOE and British government believed that Hitler was a poor strategist, a maniac whose overreach would be his downfall. If he were assassinated, they believed, his replacement (likely Himmler) would be a more competent leader, and this would draw out the war and increase Allied losses. In the opinion of the Allies it was unquestionably just to kill Hitler, but no longer strategically beneficial.

There is no shortage of situations where assassination would have been just, but of questionable strategic value. Resistance groups pondering assassination have many questions to ask themselves in deciding whether they are being strategic or not. What is the value of this potential target to the enemy? Is this an exceptional person or does his or her influence come from his or her role in the organization? Who would replace this person, and would that person be better or worse for the struggle? Will it make any difference on an organizational scale or is the potential target simply an interchangeable cog? Uniquely valuable individuals make uniquely valuable targets for assassination by resistance groups.

Of course, in a military context (and this overlaps with attacks on troops), snipers routinely target officers over enlisted soldiers. In theory, officers or enlisted soldiers are standardized and replaceable, but, in practice, officers constitute more valuable targets. There’s a difference between theoretical and practical equivalence; there might be other officers to replace an assassinated one, but the replacement might not arrive in a timely manner nor would he have the experience of his predecessor (experience being a key reason that Michael Collins assassinated intelligence officers). That said, snipers don’t just target officers. Snipers target any enemy soldiers available, because war is essentially about destroying the other side’s ability to wage war.

The benefits must also outweigh costs or side effects. Resistance members may be captured or killed in the attempt. Assassination also provokes a major response—and major reprisals—because it is a direct attack on those in power. When SS boss Reinhard Heydrich (“the butcher of Prague”) was assassinated in 1942, the Nazis massacred more than 1,000 Czech people in response. In Canada, martial law (via the War Measures Act) has only ever been declared three times—during WWI and WWII, and again after the assassination of the Quebec Vice Premier of Quebec by the Front de Libération du Québec. Remember, aboveground allies may bear the brunt of reprisals for assassinations, and those reprisals can range from martial law and police crackdowns to mass arrests or even executions.

There’s an important distinction to be made between assassination as an ideological tactic versus as a military tactic. As a military tactic, employed by countless snipers in the history of war, assassination decisively weakens the adversary by killing people with important experience or talents, weakening the entire organization. Assassination as an ideological tactic—attacking or killing prominent figures because of ideological disagreements—almost always goes sour, and quickly. There are few more effective ways to create martyrs and trigger cycles of violence without actually accomplishing anything decisive. The assassination of Michael Collins, for example, by his former allies led only to bloody civil war.


Individuals working underground focus mostly on small-scale acts of sabotage and subversion that make the most of their skill and opportunity. Because they lack escape networks, and because they must be opportunistic, it’s ideal for their actions to be what French resisters called insaisissable–untraceable or appearing like an accident—unless the nature of the action requires otherwise.

Individual saboteurs are more effective with some informal coordination—if, for example, a general day of action has been called. It also helps if the individuals seize an opportunity by springing into action when those in power are already off balance or under attack, like the two teenaged French girls who sabotaged trains carrying German tanks after D-Day, thus hampering the German ability to respond to the Allied landing.

One individual resister who attempted truly decisive action was Georg Elser, a German-born carpenter who opposed Hitler from the beginning. When Hitler started the World War II in 1939, Elser resolved to assassinate Hitler. He spent hours every night secretly hollowing out a hidden cavity in the beer hall where Hitler spoke each year on the anniversary of his failed coup. Elser used knowledge he learned from working at a watch factory to build a timer, and planted a bomb in the hidden cavity. The bomb went off on time, but by chance Hilter left early and survived. When Elser was captured, the Gestapo tortured him for information, refusing to believe that a single tradesperson with a grade-school education could come so close to killing Hitler without help. But Elser, indeed, worked entirely alone.

Underground networks can accomplish decisive operations that require greater coordination, numbers, and geographic scope. This is crucial. Large-scale coordination can turn even minor tactics—like simple sabotage—into dramatically decisive events. Underground saboteurs from the French Resistance to the ANC relied on simple techniques, homemade tools, and “appropriate technology.” With synchronization between even a handful of groups, these underground networks can make an entire economy grind to a halt.

The change is more than quantitative, it’s qualitative. A massively coordinated set of actions is fundamentally different from an uncoordinated set of the same actions. Complex systems respond in a nonlinear fashion. They can adapt and maintain equilibrium in the face of small insults, minor disruptions. But beyond a certain point, increasing attacks undermine the entire system, causing widespread failure or collapse.

Because of this, coordination is perhaps the most compelling argument for underground networks over mere isolated cells. I’ll discuss coordinated actions in more detail in the next chapter: Decisive Ecological Warfare.


Since individuals working underground are pretty much alone, they have very few options for sustaining operations. They may potentially recruit or train others to form an underground cell. Or they may try to make contact with other people or groups (either underground or aboveground) to work as an auxiliary of some kind, such as an intelligence source, especially if they are able to pass on information from inside a government or corporate bureaucracy. But making this connection is often very challenging.

Individual escape and evasion may also be a decisive or sustaining action, at least on a small scale. Antebellum American slavery offers some examples. In a discussion of slave revolts, historian Deborah Gray White explains, “[I]ndividual resistance did not overthrow slavery, but it might have encouraged masters to make perpetual servitude more tolerable and lasting. Still, for many African Americans, individual rebellions against the authority of slaveholders fulfilled much the same function as did the slave family, Christianity, and folk religion: it created the psychic space that enabled Black people to survive.”6

Historian John Michael Vlach observes: “Southern plantations actually served as the training grounds for those most inclined to seek their freedom.” Slaves would often escape for short periods of time as a temporary respite from compelled labor before returning to plantations, a practice often tolerated by owners. These escapes provided opportunities to build a camp or even steal and stock up on provisions for another escape. Sometimes slaves would use temporary escapes as attempts to compel better behavior from plantation owners.7 In any case, these escapes and minor thefts helped to build a culture of resistance by challenging the omnipotence of slave owners and reclaiming some small measure of autonomy and freedom.

Individuals have some ability to assert power, but recruitment is key in underground sustaining operations. A single cell can gather or steal equipment and supplies for itself, but it can’t participate in wider sustaining operations unless it forms a network by recruiting organizationally, training new members and auxiliaries, and extending into new cells. One underground cell is all you need to create an entire network. Creating the first cell—finding those first few trusted comrades, developing communications and signals—is the hardest part, because other cells can be founded on the same template, and the members of the existing cell can be used to recruit, screen, and train new members.

Even though it’s inherently difficult for an underground group to coordinate with other distinct underground groups, it is possible for an underground cell to offer supporting operations to aboveground campaigns. It was an underground group—the Citizen’s Commission to Investigate the FBI—that exposed COINTELPRO, and allowed many aboveground groups to understand and counteract the FBI’s covert attacks on them. And the judicious use of sabotage could buy valuable time for aboveground groups to mobilize in a given campaign.

There are clearly campaigns in which aboveground groups have no desire for help from the underground, in which case it’s best for the underground to focus on other projects. But the two can work together on the same strategy without direct coordination. If a popular aboveground campaign against a big-box store or unwanted new industrial site fails because of corrupt politicians, an underground group can always pick up the slack and damage or destroy the facility under construction. Sometimes people argue that there’s no point in sabotaging anything, because those in power will just build it again. But there may come a day when those in power start to say “there’s no point in building it—they’ll just burn it down again.”

Underground cells may also run a safehouse or safehouses for themselves and allies. Single cells can’t run true underground railroads, but even single safehouses are valuable in dealing with repression or persecution. A key challenge in underground railroads and escape lines is that the escapees have to make contact with underground helpers without exposing themselves to those in power. Larger, more “formalized” underground networks have specialized methods and personnel for this, but a single cell running a safehouse may not. If an underground cell is conscientious, its members will be the only ones aware that the safehouse exists at all, which puts the burden on them to contact someone who requires refuge.

Mass persecution and repression has happened enough times in history to provide a wealth of examples where this would be appropriate. The internment of Japanese Canadians during World War II is quite well-known. Less well-known is the internment of hundreds of leftist radicals and labor activists starting in 1940. Leading activists associated with certain other ethnic organizations (especially Ukrainian), the labor movement, and the Communist party were arrested and sent to isolated work camps in various locations around Canada. A few managed to go into hiding, at least temporarily, but the vast majority were captured and sent to the camps, where a number of them died.8 In a situation like that, an underground cell could offer shelter to a persecuted aboveground activist or activists on an invitational basis without having to expose themselves openly.

Many of these operations work in tandem. Resistance networks from the SOE to the ANC have used their escape lines and underground railroads to sneak recruits to training sites in friendly areas and then infiltrated those people back into occupied territory to take up the fight.

Underground networks may be large enough to create “areas of persistence” where they exert a sizeable influence and have developed an underground infrastructure rooted in a culture of resistance. If an underground network reaches a critical mass in a certain area, it may be able to significantly disrupt the command and control systems of those in power, allowing resisters both aboveground and underground a greater amount of latitude in their work.

There are a number of examples of resistance movements successfully creating areas of persistence. The Zapatistas in Mexico exert considerable influence in Chiapas, so much so that they can post signs to that effect. “You are in Zapatista rebel territory” proclaims one typical sign (translated from Spanish). “Here the people give the orders and the government obeys.” The posting also warns against drug and alcohol trafficking or use and against the illegal sale of wood. “No to the destruction of nature.”9 Other Latin American resistance movements, such as the FMLN in El Salvador and the Sandinistas in Nicaragua, created areas of persistence in Latin America in the late twentieth century. Hamas in Palestine and Hezbollah in Lebanon have similarly established large areas of persistence in the Middle East.


Because working underground is dangerous and difficult, effective resisters mostly focus on decisive and sustaining operations that will be worth their while. That said, there are still some shaping operations for the underground.

This includes general counterintelligence and security work. Ferreting out and removing informers and infiltrators is a key step in allowing resistance organizations of every type to grow and resistance strategies to succeed. Neither the ANC nor the IRA were able to win until they could deal effectively with such people.

Underground cells can also carry out some specialized propaganda operations. For reasons already discussed, propaganda in general is best carried out by aboveground groups, but there are exceptions. In particularly repressive regimes, basic propaganda and education projects must move underground to continue to function and protect identities. Underground newspapers and forms of pirate radio are two examples. Entire, vast underground networks have been built on this principle. In Soviet Russia, samizdat was the secret copying of and distribution of illegal or censored texts. A person who received a piece of illegal literature—say, Vaclav Havel’s Power of the Powerless—was expected to make more copies and pass them on. In a pre–personal computer age, in a country where copy machines and printing presses were under state control, this often meant laboriously copying books by hand or typewriter.

Underground groups may also want to carry out certain high-profile or spectacular “demonstration” actions to demonstrate that underground resistance is possible and that it is happening, and to offer a model for a particular tactic or target to be emulated by others. Of course, demonstrative actions may be valuable, but they can also degrade into symbolism for the sake of symbolism. Plenty of underground groups, the Weather Underground included, hoped to use their actions to “ignite a revolution.” But, in general—and especially when “the masses” can’t be reasonably expected to join in the fight—underground groups must get their job done by being as decisive as possible.

Editor’s note: continue reading at Target Selection.

Leave a Reply

Your email address will not be published. Required fields are marked *