Group Effects: Conformity, Obedience, and the Power of Collective Action

Standing up to authority alone is difficult. However, a group has power. Freud described this archetypal event back in 1921: oppressed sons unite to rebel against a despotic father. Delacroix depicted mass uprisings against unjust rulers; Gandhi successfully united people in nonviolent resistance to British rule. Prisoners at Attica conspired and temporarily overthrew the prison administration. Relationships among equals can rival, and sometimes overcome, the bonds that tie individuals to authority.

The Difference Between Conformity and Obedience

It’s important to distinguish between obedience and conformity. Conformity can be broadly defined, but here it refers to a subject agreeing with peers—people of equal status who have no special right to direct their behavior. Obedience, on the other hand, is when a subject follows the lead of an authority figure.

Take, for example, a new army recruit. He carefully follows all orders from superiors (obedience), but also adopts the habits, customs, and language of other recruits (conformity). In the first case, he’s obeying; in the second, he’s conforming.

Solomon Asch conducted a series of brilliant experiments on conformity (1951). In a group of six, where all but one were confederates, participants were shown a line of a certain length and asked which of three other lines matched it. The confederates were instructed to give incorrect answers. The naive subject heard the majority’s answers before giving his own. Asch found that under this social pressure, a large percentage of subjects preferred to agree with the group rather than trust their own eyes.

Asch’s subjects displayed conformity to the group. In our own research, subjects displayed obedience to the experimenter. In both cases, individuals surrendered initiative to an external source, but there are key differences:

1. Hierarchy. Obedience occurs within a hierarchical structure, where the subject feels the superior has the right to direct them. Conformity regulates behavior among equals; obedience links different statuses.
2. Imitation. Unlike obedience, conformity is imitation. Conformity makes behavior uniform: the influenced person adopts the group’s behavior. In obedience, the individual follows orders without imitating the authority. A soldier doesn’t just repeat an order; he carries it out.
3. Explicitness. In obedience, the directive is explicit—a command or order. In conformity, the demand for solidarity is often implicit. In Asch’s experiment, there was no open demand for the subject to agree; he chose his behavior spontaneously. Many people resist explicit calls for obedience from peers, since equals have no right to command each other.
4. Voluntariness. The clearest difference appears after the fact, in how people explain their behavior. Subjects deny conformity and admit obedience as explanations. In Asch’s experiment, subjects usually underestimated the extent to which their behavior was influenced by the group, minimizing the group effect and emphasizing their independence—even if they conformed every time. They often insisted they simply made a mistake (e.g., poor eyesight). In obedience experiments, subjects explained their willingness to administer electric shocks by saying they wouldn’t have done it on their own, but followed the authority’s demands. Thus, with conformity, people insist the group didn’t make them less autonomous; with obedience, they say they had no autonomy and everything depended on authority.

Why is this? Conformity is a response to implicit pressure: the subject sees their behavior as voluntary and can’t name a good reason to follow the group, so they deny conformity even to themselves. With obedience, the situation is publicly defined as involuntary: it’s explicitly stated that obedience is expected, and subjects refer to this public definition when explaining their actions.

Thus, obedience and conformity have different psychological consequences. Both, however, are powerful forms of social influence, and we can now examine their roles in the experiment.

Experiment 17: Two Equals Rebel

As mentioned, it’s easier to rebel against hostile authority collectively than alone. Every revolutionary group knows this, and it’s easy to demonstrate experimentally. We’ve seen that there’s often a gap between subjects’ moral principles and their behavior in the lab. Despite the victim’s protests and obvious internal conflict, many follow the experimenter’s instructions, administering shocks up to the maximum level.

Let’s examine to what extent group influence can free a subject from authoritarian control and allow them to follow their own values. To do this, we modified the basic experiment: the subject is seated between two peers (confederates) who resist the experimenter and refuse to punish the victim against her will. How much will their actions influence the naive subject?

Method

Four people enter the lab to participate in a study on “the effects of group learning and punishment on memory and learning.” Three are confederates; one is the naive subject. The experimenter explains that there will be three “teachers” and one “learner.” Through a rigged draw, the naive subject becomes “Teacher 3.” The confederates take the roles of “Teacher 1,” “Teacher 2,” and “Learner.” The “learner” is strapped to the “electric chair,” and the three “teachers” sit at the shock generator. “Teacher 1” reads word pairs, “Teacher 2” judges the answers, and “Teacher 3” (the naive subject) administers punishment. As in the basic experiment, the teachers are told to increase the shock level by one step each time the learner makes a mistake.

Confederate behavior: The confederates follow the experimenter’s orders until the victim’s first strong protests (after a 150-volt shock). At this point, “Teacher 1” tells the experimenter he doesn’t want to continue because the victim is complaining. The experimenter insists, but “Teacher 1” refuses, gets up, and moves to another part of the room. The experimenter’s attempts to bring him back fail, so he orders the remaining two to continue. “Teacher 3” (the naive subject) must now both administer shocks and read the words.

After level 14 (210 volts), “Teacher 2” expresses concern for the learner and also refuses to continue. The experimenter orders him to go on, but he too gets up and moves to the opposite corner, saying, “I’ll answer any questions you have, but I won’t shock someone against their will. I’m not participating.”

The naive subject is left alone at the generator, having seen both peers defy the experimenter. The experimenter tells him to continue, saying it’s important to finish the experiment.

The results are shown in Table 5. In this setup, 36 out of 40 subjects said “no” to the experimenter (compared to 14 when there was no group pressure). Thus, group resistance is highly effective at undermining the experimenter’s authority. In fact, in all the variations we studied, the experimenter never faced such decisive resistance as in this scenario.

Reactions to Disobedient Peers

Naive subjects’ reactions to their disobedient peers varied, partly depending on when they themselves refused to continue. One subject, who stopped right after “Teacher 1,” said, “I was already thinking of quitting when that guy said ‘no.’” Most disobedient subjects spoke positively about their peers. For example: “I thought they were good people. The victim said ‘enough,’ and they stopped.” (This subject quit after level 11.) “I thought they were compassionate… and they didn’t even know what would happen to them.” (Quit after level 14.) A subject who resisted after level 21 was more reserved: “I thought they could have waited a bit longer. But I don’t blame them for refusing to continue.”

Four disobedient subjects openly admitted being influenced by their peers: “It never occurred to me to stop until those two spoke up.” (Quit after level 14.) “I stopped because I didn’t want to look heartless and cruel in front of two people who had already quit.” (Quit after level 14.) However, most who refused denied being influenced by their peers.

Analyzing this experimental situation, several factors emerge that increase the group’s effectiveness:

1. Peers introduce the idea of defying the experimenter—some subjects might not have even considered it otherwise.
2. In previous experiments, a lone subject didn’t know how their refusal would be perceived: as a scandal or as normal. Here, they see two people refuse, making it seem like a natural reaction.
3. The peers’ actions confirm that shocking the victim is wrong, reinforcing the subject’s suspicion that it’s unacceptable to punish someone against their will, even in a psychological experiment.
4. Disobedient peers remain in the lab after refusing to participate (since they agreed to answer questions afterward). Every additional shock administered by the naive subject is met with social disapproval from the peers.
5. While the peers participate, responsibility for the shocks is shared. After they refuse, all responsibility falls on the naive subject.
6. The naive subject witnesses two acts of defiance and sees that the consequences are minimal.
7. The experimenter’s authority is weakened by his inability to enforce obedience: as a general rule, every failure of authority to command obedience reduces the perceived power of that authority (Homans, 1961).

In summary, groups can successfully challenge the experimenter. This reminds us that people act as they do for three main reasons: they have their own principles; they are sensitive to potential sanctions from authority; and they are sensitive to potential sanctions from the group. When an individual wants to oppose authority, they seek support from other group members. Mutual support is the strongest bulwark against abuse of power. (Of course, the group isn’t always right—group influence can be negative, as in lynch mobs or criminal gangs.)

Experiment 18: An Equal Administers the Shock

Those in authority understand the power of groups and often use them to establish obedience. This is demonstrated by a simple modification of our experiment. Consider this: whenever something or someone comes between the subject and the consequences of the shocks, any factor that increases the distance between the subject and the victim reduces the pressure on the participant, and thus reduces disobedience. In modern society, there are often other people between us and the harmful acts we contribute to.

This is typical of modern bureaucracy, even when it serves destructive ends: most people involved in the chain don’t directly harm anyone. They shuffle papers, load ammunition, or do other tasks that contribute to later destruction, but are removed from the acts themselves, both physically and mentally.

To study this phenomenon in the lab, we created a variation where the shocks are administered not by the naive subject, but by his peer (a confederate). The naive subject performs auxiliary tasks that, while supporting the experiment, don’t involve directly operating the shock generator.

This new role was easy for subjects. Table 5 shows the results for 40 subjects: only three refused to participate until the end. The rest played a decorative role in administering shocks and were not psychologically involved enough for internal conflict to lead to disobedience.

In a destructive bureaucratic system, a skilled manager can select personnel so that only the most callous and insensitive people commit acts of violence. The majority of the staff can be men and women who, thanks to their distance from the acts of cruelty, feel little internal conflict as they perform their supporting roles. They are freed from a sense of responsibility for two reasons: first, they are following the orders of legitimate authority; second, they themselves are not inflicting any physical harm.