CHAPTER SEVEN
“When an individual wishes to stand in opposition to authority, he does best to find support for his position from others in his group.”1
STANLEY MILGRAM
IN HIS BOOK Obedience to Authority, Milgram notes:
The crux of the study is to vary the factors believed to alter the degree of obedience to the experimental commands and to learn under what conditions submission to authority is most probable and under what conditions defiance is brought to the fore.2
In the many years I have talked about the Milgram experiments with those professionals and managers who were familiar with them, I have never heard this emphasis. Milgram scholars know about the variations, but there has been a failure to effectively disseminate this knowledge broadly in the culture. We seem to be so stunned by the indictment of human behavior revealed by the core experiment (though Milgram was careful not to make moral judgments) that we have missed the potential gold to be found in the dark soil he mined. What helps individuals intelligently disobey?
Let’s examine a few of the variations Milgram did on the basic experiment to see what else we can learn about obedience or defiance in the face of questionable, dangerous, or outright inhumane orders.
Closeness of Subject
Milgram tested several degrees of closeness of the subject to the learner (victim). When the learner was remote, in another room, 65 percent of the teachers complied all the way to 450 volts when they could only hear voice feedback from the anguished learner. When the learner was placed in the same room as the subject, and his physical anguish could be seen as well as heard, full compliance dropped to 40 percent. When the experiment was altered so the subject had to force the learner’s hand into full contact with the shock plate, only 30 percent (less than half the original number) complied all the way to 450 volts. There are some sobering reflections we can make based on this data.
We live in an age of drone warfare and cyberattacks that can remotely assassinate targeted individuals and groups and disable whole cities’ power and water supplies. The individual who “pulls the lever” creates this effect without any direct experience of the devastation caused by these acts. Based on this variation of Milgram’s experiment, the conditions are near-perfect for compliance to authority. Is this acceptable to us? Is it our ideal to create human drones to launch mechanical drones? If not, how will we instill ethical responsibility in those charged with remotely executing these orders? What degree of information is required to enable them to make ethical choices regarding that execution? What feedback loops on the impact of their actions are appropriate in helping them determine whether to continue the action?
At the other end of the spectrum, we must contemplate that a significant minority of subjects (30 percent) will still execute harmful orders even if it means forcefully and physically holding the victim down in order to do so. Does this mean that there will always be some who can be relied on as henchmen of power run amok? Are there any social mechanisms to counteract this dark tendency? Will we find some counter measures in further variations of the experiment?
Closeness of Authority
Milgram postulated that if distance from the victim was a variable, so could distance from the authority affect compliance. He set up a variation in which the researcher, after giving initial instructions, left the laboratory and continued to give orders by telephone. Full compliance dropped to just 20 percent! Some of the subjects began to covertly disobey by administering only low-level shocks while falsely reporting that they were increasing the shock level as required. These instances were a different form of Intelligent Disobedience in which there is no overt resistance to authority but the subject nevertheless does what he deems is the right thing.
All Intelligent Disobedience is contextual—requiring evaluation of the degree of consequences of obeying and of disobeying, both to others and to self, and making a choice based on this evaluation as to what to do and how to do it. It could be deduced from this variation of the experiment that the more the physical presence of authority is required in order to exact compliance to potentially harmful orders, the less truly legitimate is that authority. Perhaps this is another criterion individuals can use or be trained to use in order to evaluate the context of a situation and whether it is appropriate to obey. Let’s return for a moment to the waterboarding scene to see how this might work.
Imagine again that you are part of a detail of enlisted soldiers who have been instructed to waterboard a high-value prisoner. After a couple of rounds of waterboarding have been completed, the senior officer leaves the area and tells you to continue while he is gone. Instead you and the group find yourselves hesitating, maybe only pouring water for ten seconds and leaving more space between doses for the prisoner to recover his breath and reduce his panic.
When the senior officer returns you resume the full treatment, placing accountability on the officer. You notice that you are behaving differently in the presence of this authority from how you are behaving in his absence. This should trigger an individual and group examination of the morality of the command. The evasion of the order is an indicator that the locus of accountability needs to shift back to you—a central aspect of Intelligent Disobedience—whether or not the senior officer returns to the scene. Is the order legitimate? Is it producing beneficial or harmful consequences? If called before an investigation of the matter, would you find yourself claiming you were just following orders? This is a bright red danger signal! If this signal appears, it should be a cue to find a way of creating the physical or mental distance needed from the authority figure to work out what is the right thing, and then do that—whether it is intelligent obedience or Intelligent Disobedience.
The essence, in either case, is what you DO. It is not what you feel or what you say. It is the choice you make and what you actually DO.
Milgram offers the example of a subject who expresses repeated objections to the experiment but continues to administer the increasingly strong shocks. He observes there is a disassociation between the subject’s words and actions. There is also an out-of-place politeness toward the experimenter. Milgram notes, “He thinks he is killing someone, yet he uses the language of the tea table.”3 We may think we can let ourselves off the moral hook by saying “I didn’t feel what we were doing was right. I told them we should stop.” No, this doesn’t absolve us to the world and shouldn’t absolve us to ourselves. We still obeyed. We were polite. We followed orders we should not have followed. It is always our actions for which we will be held accountable.
Conflicting Authorities
There is a standard in many cultures that two parents ought not to disagree in front of their children. By extension, in many groups it is considered bad form for two authority figures to disagree in front of subordinates. It is true that such disagreement may generate discomfort on the part of those who witness it and uncertainty as to who is in charge and whose lead to follow. Is this always a bad thing?
Milgram structured one variation on the experiment by introducing two researchers, both in lab coats. They took turns giving instructions. While they were in agreement, this posed no problem. When the experiment reached the 150 volt level, which is the level when the victim first protests vehemently, one of the researchers says the experiment needs to be stopped. The other expresses the usual response: “The experiment requires that we go on.” They each repeat their contrary position several times. In this variation, not a single subject followed the orders to continue.
There is obviously an important lesson in this data. Life is often not so simple that there is only one authority figure. Often there are multiple authority figures, whether at our airport screening, the hospital emergency room, the WorldCom accounting department, or the school where you work or study. We are discovering in this data that obedience occurs within a social context. It is not just the individual receiving and acting on the order who is culpable (though he is), or even just the individual giving the harmful order (though he is). The rest of the system is also responsible. We will get to the role of other nonauthority participants shortly. First, let’s examine the role of third-party authority figures in the dynamic.
We see from Milgram’s data that one authority figure countermanding another is sufficient to completely disrupt compliance to the harmful order. The second authority figure did not hold rank above the other. He simply held equal legitimacy by reason of his role and its symbols. His taking a moral stand vis-à-vis the researcher who insisted on continuing the experiment was almost completely effective in reducing obedience to harmful orders. It seems that this places even greater responsibility on peers of authority figures to stop destructive acts than on the recipients of the orders. Yet, social dynamics make it no easier for peers of authority to do this than for subordinates. The same ability to step back and rise above the pressures of the moment are needed for peers to act.
Let me share a recent story of a young friend to illustrate this, a new school teacher, highly committed to teaching and her students’ welfare. I will call her Marcy. Marcy had a colleague who taught another class at the same level. I will call her Michelle.
Michelle had a somewhat difficult personality and relations had been strained between her and Marcy. One day Marcy needed to borrow something from the other classroom. When she walked in, she found a student duct-taped to a chair! He was immobilized in the chair in plain sight of Michelle and the class. Obviously, Michelle had either done the duct-taping, or ordered it or, even less plausibly, condoned it being done by other students. Marcy was reflexively shocked but stifled her natural response out of concern for worsening her already delicate relationship with Michelle. Seeing Marcy, Michelle asked if she would watch the class for five minutes while she changed clothes for the next activity. Marcy agreed. When Michelle left, Marcy asked the student who was bound to the chair if he was okay? He seemed to indicate that he was, so Marcy did nothing further.
The next day, the parents of the student came to the school principal outraged at what had been done to their son. The principal confirmed their story with Michelle who had indeed taken the action as a disciplinary measure. The principal immediately dismissed her from her employment for grossly unacceptable behavior. The following day, Marcy came to work as usual. Through further investigation, the principal had discovered that Marcy had also been in the classroom for a few minutes and had done nothing to counter Michelle’s actions. She, too, was dismissed on the spot.
We may be inclined to discount this as an outrageous example that isn’t reflective of the everyday world in which we live and work. I thought so, until I read a report written for the US Senate, Health, Education, Labor and Pensions committee titled the Dangerous Use of Seclusion and Restraint Remains Widespread and Difficult to Remedy, February 12, 2014. The report starts with the provocative sentence:
“This past August an Arizona teacher used duct tape to restrain a second grader to a chair because she was getting up to sharpen her pencil too frequently.”
The executive summary of the report cites at least 66,000 incidents of seclusion or restraint used in the first year of reporting in 2009–2010. Other sources cite the incidence of these practices as high as a quarter of a million in subsequent years! Federal law on the use of restraints and seclusion did not cover school instances of this, and the use of these practices varied among school systems. Thus it fell to the discretion of the local authority figures until recently, when the practice was banned in all but extreme circumstances such as when there is a need to prevent physical harm to the student or others.
Marcy is a very dedicated teacher and, happily, was able to have the mark expunged from her record. She is entering a new school system, having learned a very important lesson. Her silence in an attempt to avoid adverse social consequences for speaking up made her complicit in the act of her colleague. The principal’s reaction may seem unfair since it treated Marcy with the same level of culpability as the teacher who had actually bound the student to the chair and kept him there for the whole period. Perhaps the reaction was harsher than needed, yet it correctly sent a message of the accountability of all authority figures who become aware of harmful behavior to act to interrupt the behavior, not just leave doing so to those in the chain of command.
Marcy could have asked Michelle to step outside for a moment and confronted her in the hallway out of earshot of her students. Avoiding publicly embarrassing the other authority figure, if possible, is a good relationship principle. If that were insufficient to dissuade Michelle to cease the behavior, speaking up publicly may have been the course required to disrupt the harmful behavior, despite the social discomfort that would create.
The Power of the Bystander
There is a vast amount of literature these days on bullying in schools, in virtual electronic spaces, and in the workplace. The preponderance of the literature focuses on not tolerating bullies and on better supporting their victims and equipping them to withstand bullying behavior. Only a few authors give the attention due to the third actor in the bullying dynamic—the bystander, though fortunately this seems to be changing. The bystander is present in the vicinity of the bullying but is not the direct target of the traumatizing behavior. Therefore, the bystander is better equipped psychologically to respond to the situation. These authors correctly identify the bystander as the actor with the greatest potential power to interrupt the destructive behavior.4
One of these authors, Barbara Coloroso, correctly observes that where state power turns genocidal, whether in Nazi Germany or any of the other dismal instances of genocide, the same structure applies. It is the silence of the majority who are bystanders that permits the crime to occur.
Although workplace bullies may be positional authority figures, many are not; school yard and virtual bullies are certainly not. Nevertheless, through their campaigns of terror, they develop a pseudo-legitimacy that makes others comply with their instructions to participate in the bullying. This has become a serious concern for school administrators, teachers, and parents in the education and online social systems and for human resource directors in the workplace. What can be done to break the power of the bully-authority?
Milgram devised a variation of his experiment that shows us a path for colleagues or bystanders to trigger Intelligent Disobedience. Later research on bullying itself corroborates the principles that surfaced in this variation.
Milgram placed the “naïve” subject between two other individuals who appear to be fellow teachers in the experiment. In fact, they are “confederates” who have been given exact roles to play. Teacher 1 (confederate 1) is assigned the task of reading the paired words to the learner. Teacher 2 (confederate 2) reports whether the learner’s answers are correct. Teacher 3 (the naïve subject) pulls the levers to administer shocks.
At 150 volts, teacher 1 refuses to go on, gets up from the table, and seats himself elsewhere in the room despite the experimenter’s demands that the experiment requires continuing. The naïve subject is told to take over the role of reading the questions while continuing the role of administering shocks. By 195 volts, 32 percent of the participants refused to continue. At 210 volts, the second confederate, teacher 2, also refuses. He relocates himself to another seat, saying he is willing to answer the experimenter’s questions afterward but not to further participate in the experiment. An additional 30 percent of subjects immediately also refuse to continue. Only 10 percent continue to the end of the experiment at 450 volts. Ninety percent have resisted the urgings of authority to complete the experiment!
Milgram considers the lessons learned from this variation of the experiment as the most effective at reducing obedience to an authority that is issuing harmful orders. He identifies three principal reasons:
1. Most of us have internalized standards of behavior; when one or more others resist orders that violate those standards, it normalizes resistance to the orders.
2. If we see others take a stand without being unduly penalized by authority, we recalculate the risks involved of doing so ourselves.
3. If others refuse to participate in the harmful acts, we become aware that if we continue to comply, the group’s sanctions may turn on us.
Some of this is obviously problematic. What if authority does penalize individuals who first disobey? That is a tactic often applied by those seeking to hold onto power through force. That is precisely why the most important act of disobedience is often that of the second or third resister. Their support for the first resister tips the calculations of others. Instead of placing greater weight on the power of formal authority or pseudo-authority, others begin placing greater weight on the power of social norms, of doing what we intuitively know is the right thing. The authority loses standing and his orders lose force.
We are assuming, of course, that the social norms that replace the force of authority uphold human values of decency; if they don’t, and the history of groups repressing other groups tells us they may not, Intelligent Disobedience will need to find its own path, independent of either authority or convention. The internal compass pointing us to the right thing to do will need to be our primary orientation. How to give it greater weight than the social pressures around us will not come easily for many. We will examine the preparation needed to do so in succeeding chapters.
One Final Variation
There are other variations of the basic experiment that hold lessons on obedience and disobedience with which researchers and teachers of Milgram’s works are familiar. For our purposes, I want to shine a light on one more variation of the basic experiment, the one that is the most frightening. Why? Because in this variation, 90 percent of the subjects comply with the experiment to the end! It is also frightening because it is the experiment that tells us why most individuals in large bureaucracies participate in destructive acts when they do: they themselves are not directly and immediately causing harm.
In this variation of the experiment, the subject does not administer shocks to the victim. The subject is given one of the ancillary roles like reading the question or documenting the answers. Because the subject is not directly causing pain, there is less psychological strain. The norms of the experiment itself prevail over greater moral norms. The experiment norms include the agreement to participate, the perception of legitimate authority and legitimate purpose, the social evidence of peer participation, and the absence of personal culpability in administering the pain.
Transfer this from the experiment to the case examples we have seen. This is not the soldier pouring water onto the wet towel over the prisoner’s nose and mouth as he gasps violently for air. This is the analyst who identifies the prisoner as a possible high-value target for enhanced questioning. The analyst is doing her job, fulfilling her duty, implementing her training. She does not have to grapple with the immediate consequences of her analysis and the recommendations that proceed from it.
In the complex society in which we live, most of us occupy the equivalent of the analyst’s role. We are not knowingly installing life-threatening equipment in automobiles—we work in a contracting office using accepted standards to purchase ignition parts at the lowest cost. We are not administering ineffective medicine to a dangerously sick patient—we are part of a research team cleaning up messy statistical data in a drug trial so efficacy can be more readily evaluated. We are not intentionally contributing to increased juvenile diabetes—we are providing affordable school lunches that can be predictably delivered to the cafeteria and appeal to the students’ taste. We are not advocating inhumane factory working conditions on the other side of the globe—we are upgrading our phones to the newest user-friendly technology.
In these cases and a thousand like them, we are not directly harming a fellow human being. We are not hearing or seeing suffering caused earlier or later in the chain of events. We can screen out the problematic information on the periphery of our awareness and focus on doing the jobs to which we are assigned.
My analysis of this example slips from the narrower domain of Intelligent Disobedience to a direct order, into the more difficult domain of principled refusal to participate in a system producing harm. This is a hybrid somewhere between Intelligent Disobedience and civil disobedience. It is a standard that can be very difficult to hold ourselves to. I shine light on it because it must not be kept in the dark. Though the subjects in this variation of the experiment have a role in the process, they are effectively bystanders who are remaining silent. Yet, even in this variation of the Milgram experiment, 10 percent hold themselves sufficiently morally accountable to refuse to continue in their ancillary roles—refuse to be silent, complicit bystanders.
In our age of social media, sometimes one of these rare individuals feels called to raise public awareness of the abuses. Movements have been generated to label the origin of products, to avoid purchasing products from war-torn areas or those that are produced without fair and sustainable practices. Due to the amplifying powers of social media, tens of thousands pick up the cause and carry it forward. The producers of the products find it in their commercial interest to be responsive, and change does occur. This is another example in which you may not be the first to intelligently disobey, but you can support those who do when you recognize that they are standing up for doing the right thing.
Let’s review some of the lessons we can draw from the variations to the basic experiment:
1. If you are uncomfortable about what you are told to do, speak up early and do not let your discomfort be dissipated by answers that are technical rather than moral.
2. If the order seems morally wrong or potentially illegal, unless there is immediate danger, require the order be given to you in writing; if the authority will not put the order in writing, do not implement it.
3. If the order conflicts with a higher set of values, rely on the authority of those values to make the decision for which you will be accountable.
4. If you are responsible for the ethical development of others, design opportunities to practice questioning technical orders that conflict with higher level moral orders.
5. If the action you are ordered to take will make its impact felt at a remote location or point in time, visualize that impact and whether it will conform to the morals and laws you are committed to upholding.
6. If you find yourself avoiding implementation of an order when the authority giving the order is not present, recognize you are not fully convinced the order is a correct one and question it more closely.
7. You are accountable if you implement a wrong order even if you feel bad about doing so and express disagreement with the order; only refusing to implement the order absolves you from complicity.
8. If you are an authority and disagree with the correctness or morality of the order of another authority, you are responsible for speaking up against the wrong orders and, thereby, helping others to do so.
9. Do not allow social politeness to keep you from speaking up clearly against morally incorrect orders from others, regardless of their position in the hierarchy and despite the discomfort of doing so.
10. If someone else resists a morally or operationally wrong order or systemic abuse before you do, support that person in refusing to comply with the order as given or in examining alternatives to the current practice.
There is more we can learn from Milgram, but let’s leave him for a while and return to the inspiration for this book, the guide dog. What lessons will we find in the benign efforts to teach Intelligent Disobedience?
