Within both aboveground and underground activism there are several templates for basic organizational structures. These structures have been used by every resistance group in history, although not all groups have chosen the approach best suited for their situations and objectives. It is important to understand the pros, cons, and capabilities of the spectrum of different organizations that comprise effective resistance movements.
The simplest “unit” of resistance is the individual. Individuals are highly limited in their resistance activities. Aboveground individuals (Figure 8-1b) are usually limited to personal acts like alterations in diet, material consumption, or spirituality, which, as we’ve said, don’t match the scope of our problems. It’s true that individual aboveground activists can affect big changes at times, but they usually work by engaging other people or institutions. Underground individuals (Figure 8-1a) may have to worry about security less, in that they don’t have anyone who can betray their secrets under interrogation; but nor do they have anyone to watch their back. Underground individuals are also limited in their actions, although they can engage in sabotage (and even assassination, as all by himself Georg Elser almost assassinated Hitler).
Individual actions may not qualify as resistance. Julian Jackson wrote on this subject in his important history of the German Occupation of France: “The Resistance was increasingly sustained by hostility of the mass of the population towards the Occupation, but not all acts of individual hostility can be characterized as resistance, although they are the necessary precondition of it. A distinction needs to be drawn between dissidence and resistance.” This distinction is a crucial one for us to make as well. Jackson continues, “Workers who evaded [compulsory labor], or Jews who escaped the round-ups, or peasants who withheld their produce from the Germans, were transgressing the law, and their actions were subversive of authority. But they were not resisters in the same way as those who organized the escape of [forced laborers] and Jews. Contesting or disobeying a law on an individual basis is not the same as challenging the authority that makes those laws.”1
Of course, one’s options for resistance are greatly expanded in a group.
The most basic organizational unit is the affinity group. A group of fewer than a dozen people is a good compromise between groups too large to be socially functional, and too small to carry out important tasks. The activist’s affinity group has a mirror in the underground cell, and in the military squad. Groups this size are small enough for participatory decision making to take place, or in the case of a hierarchal group, for orders to be relayed quickly and easily.
The underground affinity group (Figure 8-2a, shown here with a distinct leader) has many benefits for the members. Members can specialize in different areas of expertise, pool their efforts, work together toward shared goals, and watch each others’ backs. The group can also offer social and emotional support that is much needed for people working underground. Because they do not have direct relationships with other movements or underground groups, they can be relatively secure. However, due to their close working relationships, if one member of the group is compromised, the entire affinity group is likely to be compromised. The more members are in the group, the more risk involved (and the more different relationships to deal with). Also because the affinity group is limited in size, it is limited in terms of the size of objectives it can go after, and their geographic range.
Aboveground affinity groups (Figure 8-2b) share many of the same clear benefits of a small-scale, deliberate community. However, they may rely more on outside relationships, both for friends and fellow activists. Members may also easily belong to more than one affinity group to follow their own interests and passions. This is not the case with underground groups—members must belong only to one affinity group or they are putting all groups at risk.
The obvious benefit of multiple overlapping aboveground groups is the formation of larger movements or “mesh” networks (Figure 8-3b). These larger, diverse groups are better able to get a lot done, although sometimes they can have coordination or unity problems if they grow beyond a certain size. In naturally forming social networks, each member of the group is likely to be only a few degrees of separation from any other person. This can be fantastic for sharing information or finding new contacts. However, for a group concerned about security issues, this type of organization is a disaster. If any individual were compromised, that person could easily compromise large numbers of people. Even if some members of the network can’t be compromised, the sheer number of connections between people makes it easy to just bypass the people who can’t be compromised. The kind of decentralized network that makes social networks so robust is a security nightmare.
Underground groups that want to bring larger numbers of people into the organization must take a different approach. A security-conscious underground network will largely consist of a number of different cells with limited connections to other cells (Figure 8-3a). One person in a cell would know all of the members in that cell, as well as a single member in another cell or two. This allows coordination and shared information between cells. This network is “compartmentalized.” Like all underground groups, it has a firewall between itself and the aboveground. But there are also different, internal firewalls between sections.
Such a network does have downsides. Having only a single link between cells is beneficial, in that if one cell is compromised, it is much more difficult to compromise other cells. However, the connection is also more brittle. If a “liaison” is removed from the network or loses communication for whatever reason, then the network may be broken up. A backup plan for regaining communication can reduce the damage from this, but increase the level of risk. Also, the nonhierarchal nature of this network means that choosing actions can be more difficult. The more cells are involved, the larger the number of people who must have critical information in order to make decisions. That said, these groups can be very effective and functional. The famous Underground Railroad was a decentralized underground network.
Some of these problems are addressed in both aboveground and underground groups through the use of a hierarchy. In underground hierarchies (Figure 8-4a), large numbers of cells can be connected and coordinated through branching, pyramidal structures. These types of groups have vastly greater potential than smaller networks. Their numbers make for increased risk, yes, but that increased risk can be reduced by the use of specialized counterintelligence cells within the network and wide-ranging coordinated attacks.
Aboveground hierarchies (Figure 8-4b) are quite familiar and common, in part because they are highly effective ways of coordinating large numbers of people to accomplish a specific objective. As shown, aboveground hierarchies facilitate many relationships between people in different parts of the hierarchy. This lack of compartmentalization might be good in terms of productivity, but not in terms of security.
There are very specific situations in which it may be acceptable to send information through an underground group’s firewall. The recruitment process necessarily involves communication with people outside the group. However, these people would not be active in aboveground movements, and, at least initially, they would only know one member of the organization in one cell. Of course, there are no direct relationships between people in the underground and aboveground groups.
In certain situations, one-way (and likely anonymous) communications may take place across the firewall. Informants who want to give information to the resistance network may pass on information to a member of an internal intelligence group. However, the intelligence group would not share information about identities or the network with those people. Information may also travel one-way in the opposite direction. The underground groups may want to send communiqués or other information to the media or press office. Of course, any communication across the firewall, even those thought anonymous, entails a certain small amount of risk. Therefore, the benefits must outweigh the risks.
All of the examples illustrated are simplified and generalized. Resistance groups in history have had a wide variety of internal structures based on these general templates. They often had to make a deliberate compromise between organizational security (which comes from loosely connected and decentralized cells) and organizational effectiveness (which comes from more densely connected and centralized cells).