"The drb sustains a level of commentary on Irish and international matters that no other journal in Ireland and few elsewhere can reach. It deserves all the support that can be given it." X
Space to Think, a new book celebrating ten years of the Dublin Review of Books More Information 

Smile, and turn up the power

Martin Tyrrell

Stanley Milgram’s research on obedience, carried out at Yale in the early 1960s, is well-known. Milgram, a social psychologist, found that hundreds of ordinary, inoffensive people were prepared to administer painful electric shocks to another person, similarly ordinary and inoffensive. A significant proportion did so even when the shocks they were administering appeared to be causing serious harm, possibly death. Milgram tested this effect across twenty-four separate variations (called “conditions” in the literature) on the basic “person shocking a stranger” design. For example, in his Condition 4, the stranger being shocked was visible to the person doing the shocking whereas in most of the other conditions he was in a separate room. The proportion willing to administer the most harmful levels of shock differed markedly by condition ‑ from a low of ten per cent under Condition 7 to a high of 92 per cent in Condition 13.

Milgram’s findings were first published in the Journal of Abnormal and Social Psychology in 1963 just when everyone was talking about Eichmann in Jerusalem. There, in his courtroom bubble, that prolific and ideological mass killer as bland as the rest of us, was evil in all its banality. But if Eichmann in the flesh was unremarkable might not, say, Middle America, or Middle Anywhere, turn genocidal too, given the right conditions? The obedience experiments suggested it was a strong possibility. Just as a biologist might identify a rogue gene or virus, Milgram has sometimes been seen as having identified some latent but universal “Nazi factor”.

“The results I observe in the laboratory are disturbing,” he said at the time. “If in this study an anonymous experimenter could successfully command adults to subdue a fifty-year-old man and force on him painful electric shocks against his protests, one can only wonder what government with its vastly greater authority and prestige can command of its subjects.”

Twelve years later, his book Obedience to Authority reported the project in full giving the findings a second wind. Reviewed extensively, and in the main favourably, the book did good business ‑ exceptionally so for an academic text. Its unsettling conclusions sat well with the naked apes and territorial imperatives of a pop ethology then in vogue (“Relatively few people have the resources to resist authority” blurbs the stark, faintly Soviet front cover of the Harper and Row hardback). Seldom has a social psychology project achieved so high a public profile, not just through the usual popularising channels such as newspaper serialisation and TV documentaries, but also film (The Tenth Level, with William Shatner in the Milgram role, and Michael Almereyda’s The Experimenter), literature (Berhard Schlink’s The Homecoming), and music (Peter Gabriel’s We Do What We’re Told). So widespread is interest in the Milgram effect that the original experiment has been replicated, not just in universities across the world, but on French and British TV. At a time when social psychology is being criticised for its inability to produce replicable effects, Milgram’s findings have proven worryingly robust ‑ consistent across cultures and generations.

These two substantial books evidence that continuing fascination with the obedience experiments, more than thirty years on from Milgram’s death at the early age of fifty-one. Also, they balance one another, Gina Perry’s The Shock Machine being decidedly revisionist and Thomas Blass’s The Man Who Shocked the World altogether more reverential. (Milgram, Blass says, is “one of the outstanding scientists of his generation”).

Social psychologists study social behaviour and also ‑ at least sometimes ‑ the thought processes that underlie it. Conformity, for example. Solomon Asch, in a famous 1950s experiment, found that people, faced with a group consensus that was clearly off the wall, still went along with it. More than a third of the time they denied their own better judgement and the evidence of their senses and said what everyone else was saying.

Milgram had been Asch’s student and his first solo project was to see if French and Norwegian participants would be as conforming as the Americans Asch had studied. (They were.) Milgram’s obedience experiments, as originally conceived, were to have been a further look at peer pressure and conformity.

Initially, he was going to have his participants perform either as individuals or as members of groups suitably packed out with shock-happy confederates (the experimenter’s plants, as in on the trick as a conjuror’s assistant) to assess which of these conditions ‑ group or individual ‑ would result in the greater punishment. Milgram fully expected that people in the group condition would be the most enthusiastic shockers. However, when he ran a pilot study, he found a similar level of willingness under either set-up. This shifted his interest from conformity (buckling under peer pressure) to obedience (complying with an authority).

All the same, Milgram did not completely set aside his interest in group conformity. Of the twenty-four separate conditions that comprised the eventual obedience project, four looked at group effects. It was a group condition that resulted in peak compliance, with 92 per cent of participants (thirty-seven out of forty) willing to deliver shocks to the maximum level available ‑ 450 volts.

The story of the obedience experiments, as both Blass and Perry tell it, is as follows. Milgram placed an advertisement in the New Haven Register seeking volunteers (mainly men) to take part in what was billed as a memory experiment. (Women participants were used in only one of the twenty-four conditions; the results suggested that gender had no impact on willingness to deliver punishment shocks.)

On arrival, participants were typically met by John Williams ‑ a dour, grey-coated scientist type ‑ who in turn introduced them to “Mr Wallace”, a round, affable man whose real name was Jim McDonough. McDonough, though he was presented as a volunteer like themselves, was actually a confederate of Milgram’s, chosen by him specifically because he looked passive and inoffensive. And Williams, although he was implied to be in charge of the experiment ‑ and is generally referred to as the “Experimenter” in the literature ‑ was also a confederate, a high school teacher recruited on account of his no-nonsense, authoritative manner.

After some preliminaries, Jim McDonough and the genuine participant drew lots to see which of them would be the Teacher and which the Learner. But the draw was rigged so that it was always the genuine participant who was assigned the role of Teacher, and always the confederate McDonough who got to be the Learner. McDonough was then sat in a separate but adjoining room, where he could be heard but not seen, and hooked up to the shock generator. This was a formidable looking console the front of which faced the Teacher and consisted of a row of lights under each of which was a push-down switch. Each switch purportedly delivered a progressively stronger electric shock, increasing in 15-volt increments from an initial 15 through to 435 (marked DANGER: EXTREME SHOCK! on the console) and, finally, 450 (marked XXX). As part of their initial briefing, all Teachers were given a sample shock of 45 volts to let them experience the type of pain they would be causing.

With that, the memory test of the advertisement began with the Teacher quizzing the Learner McDonough to see how well he had memorised a set of word pairs. When McDonough remembered correctly, fine. But when he got it wrong, his Teacher shocked him, beginning with a preliminary 15 volts then increasing the voltage with each subsequent wrong answer. Teachers who were prepared to deliver shocks right up to the maximum 450 volts were categorised as “obedient”. Those who dropped out at any stage before 450 volts were considered “defiant”, so that defiance was entirely relative. For example, according to Blass, no defiant participant in the first condition dropped out before 300 volts, that is before they had already administered twenty shocks. Defiance, however defined, was markedly less common than obedience. In most of the project’s twenty-four conditions, a clear majority of participants was willing to administer shocks right up to the maximum XXX, a voltage that appeared to leave McDonough silent, perhaps unconscious, or even dead. (And, throughout the project, across all its conditions, McDonough proved to be a poor student, warranting many punitive shocks.)

Milgram made a short, fly-on-the-wall documentary of the experiments entitled Obedience, intended to highlight the general effect. At the start of the film, during the initial chat between Experimenter, Learner and Teacher, McDonough happens to say that he has a heart condition that was diagnosed at the Veterans’ Hospital. “Nothing serious,” he says, as if to reassure his soon-to-be Teacher. And Williams settles him all the more ‑ the shocks they’ll be using, though painful, aren’t dangerous. (Not dangerous, even though the penultimate voltage in the series is unambiguously labelled DANGER: EXTREME SHOCK.)

In Obedience, when the first shock is administered, McDonough gives a slightly stagy “ouch” that causes some of the Teachers to smile or smirk (fourteen out of forty laughed “nervously” says Milgram in his voiceover). But when the level of shock reaches 150 volts, he says: “Get me out of here”, “My heart is starting to bother me”, “I refuse to go on, let me out”. And at the highest and most painful levels of shock he does not answer at all.

All of this ‑ the shock generator, the shocks, McDonough the Learner’s cries of distress ‑ was, of course, a set-up or “elaborate deception” as Gina Perry calls it. McDonough was unharmed throughout. All his cries and protests were in fact pre-recorded with specific reactions – “ouch”, “get me out of here” ‑ paired with particular voltages so that every Teacher heard exactly the same reaction when they delivered that particular level of shock.

Such deception was the stuff of social psychology at the time. The obedience experiments are nothing if not definitive of the style and deploy all its tropes ‑ the false pretext (a test of memory, not obedience); the confederates; the shock generator, which was non-functional, a prop.

Participants were allegedly debriefed (“dehoaxed” as Milgram called it) after the experiment, though Perry has found that this was not always done, or done adequately. And from the outset, the research has been questioned on ethical grounds. Bruno Bettelheim likened it to the Nazis’ experiments on unwilling human guinea pigs.

Although Milgram tested obedience across twenty-four separate conditions, almost all popular reporting of his findings homes in either on Condition 13 (called Non Trigger Position), the group condition where compliance was a depressingly high 92 per cent; on Condition 1 (No Feedback – 65 per cent obedience) which was the first to be published (in the Journal of Abnormal and Clinical Psychology in 1963); or on the similar Condition 2 (Voice Feedback) ‑ with 62 per cent obedience.

Perry usefully gives some details on these three, as well as on the other conditions, and notes the various levels of obedience Milgram observed for each of them. Under Condition 1(No Feedback), McDonough remained silent until he was hit with a 300-volt shock, whereupon he banged on the connecting wall. And he banged on it again at 315 volts. But after that he was once more, and somewhat ominously, silent. Nonetheless, as noted, 65 per cent of Teachers still shocked him right up to the max. In contrast, under Condition 2 (Voice Feedback), he complained at 75 volts, demanded to be let out at 150, and fell silent at 315. All the same, 62 per cent still shocked him right up to 450 volts. Obedience, the film, seems to draw especially on another condition, Condition 5, Coronary Trouble. Although McDonough alerts his Teacher that he has a heart condition and complains of it during the test, 65 per cent still went on shocking him right up to 450 volts.

In contrast, when McDonough was dispensed with and participants shocked someone they knew ‑ a friend or a family member ‑ Teachers were much more likely to refuse to go on. Defiance also increased when a milder Experimenter than Williams was used, or when the Learner was surlier than McDonough, or when the experiment was moved away from Yale to the less salubrious location of a small upstairs office suite in Bridgeport. Milgram found too that the better Teachers could see and hear the Learner, the less likely they were to shock to the highest levels. But even where the Teacher sat beside the Learner and could put the Learner’s hand on the shock plate if the Learner would not do it himself, there was still 30 per cent compliance.

There is no consensus as to why Milgram found such high levels of obedience, or why his participants acted in the way they did. Blass sees Milgram as a “situationist … a strong believer in the power of the immediate situation in affecting a person’s behaviour”. In Obedience, Milgram says in his voiceover that it is “foolish” to consider his obedient participants sadistic. “The individual,” he advises, “becomes integrated into a situation that carries its own momentum.” But in his obedience experiments, it was a particular situation ‑ the “elaborate deception” of the obedience experiment ‑ that created the behaviour he observed across so large a number of participants. And it is surely risky to generalise from that particular situation to the wider world, or to say that that specific set-up has yielded some lasting truth about the human condition.

It was a commonplace at the time that Milgram had isolated some kind of “Nazi factor”, a claim that psychologist Diana Baumrind, for one, rejected. Nazi Germany, she said, was the outcome of a specific social and historical context. Canadian psychologist Hank Stam, likewise, disputes that Milgram found some raw form of obedience, or even obedience at all. The behaviour demonstrated in the Milgram experiments, he argues, might be obedience. But it might just as easily be trust, or rule-following, or all of these. Or none of them. François Rochat, a Swiss psychologist, reckons Milgram might simply have tapped into “the inexorable subordination of the less powerful to the powerful”.

Here, in short, is a series of experimental findings open to a number of theoretical explanations, and to challenges. Milgram himself showed some inconsistency in how he accounted for what he had found, shifting from situationism, to diffusion of responsibility, to that old standby of attributing the results to the peculiarities of the people that he studied. In a letter to Henry Recken of the National Science Foundation, he comments: “I once wondered whether in all of the United States a vicious government could find enough moral imbeciles to meet the personnel requirements of a national system of death camps of the sort that were maintained in Germany. I am now beginning to think that the full complement could be recruited in New Haven.”

Gina Perry suggests that some of the effect might be down to the participants having sussed that it was all a set-up. Candid Camera was an established feature of American television by the time of the Milgram experiments and had just entered its period of peak popularity, one that would last most of the decade. Its format ‑ much imitated ‑ was to stage practical jokes (a car that appeared to run without an engine, a man walking an invisible dog) that would entrap hapless members of the general public, drawing them in while they were covertly being filmed. One, from the early 1960s, for example, has people waiting for a flight being told their pilot will be … a woman. Their reactions range from bewilderment to anxiety. One man looks as though he will never move again. Another smiles so wide it looks as though it’s going to stay that way. (“Why us?” his wife asks.) A female flight attendant says she is shaking at the prospect and suggests that it would be better if the passengers were not informed ‑ perhaps the first officer could do all the talking and the woman pilot say nothing. “Look over there,” someone says to her. “Smile, you’re on Candid Camera!” She laughs with relief.

“Like Candid Camera,” Perry writes, “Milgram’s [obedience] experiment involved trickery and secret surveillance aimed at capturing people’s behaviour.” The show was, in fact, a Milgram favourite. He considered its creator, Allen Funt, a natural social psychologist (and Funt had, in fact, worked with Gestaltist Kurt Lewin). Might the cultural ubiquity of Candid Camera at the time of the obedience experiments explain at least something of what Milgram found? The following exchange between a participant and Milgram and Williams in the course of a debriefing is revealing.

“Isn’t this awful?” the participant says. “I feel like I’m on Candid Camera.”
“You feel like you’re on Candid Camera?” says Milgram.
Most people do,” says Williams [my emphasis].
And then Milgram: “I know it is very much like, er, um, the difference of course is that … Yes, it’s not unlike it …”

That people might have seen through the “elaborate deception” of the obedience experiment and acted as though it were some kind of practical joke was a line of argument Milgram always scorned. His follow-up survey of the obedience participants found that a good 80 per cent of them (527 people) had little doubt that what they were doing was for real (compared with fewer than a hundred who were sceptical). And only 13 per cent were sceptical. What is more, many of his participants were stressed and anxious. Who gets stressed and anxious over something they think is fake, he argued? The anxiety is clear enough in Obedience where the participants are evidently perturbed at the harm they are causing. One of them asks Williams to check in on McDonough. But Williams declines, saying that he cannot do this once the process has begun. Nonetheless, the Teacher continues to beat himself up. “Are you OK?” he asks McDonough through the wall and now, when he calls out the prompt word, he says “Answer please”. Another Teacher starts hinting the correct answers to McDonough, and even offers to change places with him. And when he administers a shock, he presses and releases each switch almost immediately rather than hold it down as though he is trying not to make it too continuous and damaging.

Perry also notes that obedient participants (those who administered shocks right up to 450 volts) were more likely to be sceptical (“I expected someone to say, “You have been on Candid Camera …”’) and defiants more likely to believe it was real.

In other words, Perry suggests, if you thought the experiment was some kind of Candid Camera stunt, you were more likely to administer the full range of available shocks. But if you thought you were really hitting poor Jim McDonough with an increasing level of voltage you were more likely to drop out. But the differences are small. As are the numbers. Of more than six hundred participants, ninety-one doubted ‑ or said they doubted ‑ that the situation was real. And of that ninety-one, compliance was more common than defiance. But more people ‑ more than five hundred compliant and defiant participants ‑ believed the situation was real.

Also, when these participants told Milgram that they thought the whole thing was a fake, or that they thought it was real, they were doing so some time after the event and in the knowledge that they had been, not simply deceived, but deceived into behaving in a way that was shaming and negative. For obedients, it legitimised their behaviour to say that they knew all along it was a hoax ‑ they were not New Haven Nazis, they had rumbled the hoax. And for defiants it was similarly legitimising to say that they were completely taken in ‑ they had taken a stand against causing harm. There is also the problem that besets virtually all surveys ‑ that people often make themselves out to be better than they are. What interests me is that, in this survey of 658 ex-participants in an experiment where obedience was markedly more common than defiance, more than half of all those surveyed ‑ 56 per cent  ‑ claim to have been defiant. Either that’s a seriously unrepresentative sample, or some people have, retrospectively, changed their vote.

Perry is on stronger ground, I think, when she suggests that Milgram might have, in effect, nudged his participants in the direction of a particular kind of behaviour. “[W]e have become aware,” she writes, “that science is as much a process of construction as of discovery; that scientists are storytellers too …” She notes Milgram’s love of showmanship and performance and tells us how, periodically, he had threatened to give up psychology and start writing fiction.

The obedience experiments were carefully stage-managed, says Perry. Milgram, she says, rehearsed with both his “actors”, Williams and McDonagh, whose performances he tightly scripted. Also, he was attentive to the smallest detail, insisting, say, that Williams wore a grey rather than a white lab coat so as not to convey the authority of a physician (and, instead, merely that of a scientist, an expert, at an Ivy League university). Even the shock generator itself was professionally designed in the university’s electronic and mechanical workshop to look the part.

But that, she contends, is not the limit of Milgram’s showmanship. Not only did he prepare his props and actors, “ … he refined, tightened and scripted a scenario that would deliver the results he wanted … he knew before his first subject arrived on 7 August 1961 what sort of results he wanted to achieve, and he’d used pilot studies and pre-tests to hone the design to achieve just that”.

Look again at Obedience. Those Teachers are never quite spontaneous in their actions. They are egged on by Wallace, who uses the same, unvarying and scripted prompts should any of them prove reluctant. “Please continue” is the weakest of these and, if that fails, “Continue please. Go on.” Then the prompts are more insistent: “The experiment requires you to go on Teacher. Go on, please” and “It’s absolutely essential that you continue, Teacher. Go on.” Finally, there is a touch of emotional blackmail: “If you don’t continue, we’ll have to discontinue the entire experiment.” Only if that one fell flat was the Teacher left free to drop out. This might be one reason why, if participants were defiant at all, they tended to be so at a relatively late stage.

So Perry is right, I think, to look at that defiance, however relative it might have been, and to give it more prominence than Milgram generally did. However much obedience was found, there was always at least some defiance. In Condition 1 (No Feedback) written up in the Journal of Abnormal and Social Psychology, a third of participants were defiant. And even in that group condition (Condition 13, No Trigger) in which thirty-seven participants were willing to progress to the highest level of shock, still three did not.

It is the group conditions that most interest me. Two of them in particular ‑ Condition 7 (Group Pressure to Disobey) and Condition 9 (Group Pressure to Obey). In both of these, there were three Teachers ‑ two plants and one genuine. In the pressure to obey condition, any time the genuine Teacher, the one recruited from the advertisement in the New Haven Register, dithered over whether to continue administering shocks, the two confederates muttered and complained, thereby exerting a strong social pressure for him to go on. The result was a very high level of compliance – 72 per cent. In contrast, in the pressure to disobey condition, the two confederates dropped out of the experiment at, respectively, 150 and 210 volts. They refused to go on, rejected the Experimenter’s pleadings, and sat to one side, watching the genuine participant. As a result, in thirty-six out of forty cases, the genuine participant also dropped out—a mere 10 per cent compliance.

Milgram surmised at length as to what might have happened. For example, that the defiant peers somehow instilled the idea of defiance in their fellow participant, or that their defiance made the business of administering electric shocks suddenly immoral. Maybe, too, the continued physical presence of this pair of defiants made the genuine participant feel guilty, or made them aware of that responsibility. And perhaps also, when two people have defied the Experimenter and got away with it, the authority of that experimenter is weakened. Milgram concluded: “The mutual support provided by men (sic) for each other is the strongest bulwark we have against the excess of authority.”

Solomon Asch had found the same back in the 1950s. If he briefed one of his confederates to dissent from the crazy consensus, the genuine participant was less likely to conform to it. Even that little bit of support, it seemed, makes a person more willing to stand up to and stand out from the crowd.

It is odd, says Perry, that Milgram did not make as much of the defiance he found as he did of the obedience, particularly since he was living through a time of significant social change, upheaval even. The experiments were carried out in 1961 and Obedience to Authority published in 1974. The years between had been years of significant dissent and constructive, purposeful disobedience ‑ feminism, the protests against the Vietnam war, the mass campaign for civil rights. And yet it is obedience and conformity that interested Milgram.

People conform. And they obey. Milgram showed this, and Asch showed it before him. And Irving Janis would demonstrate it a few years later with his “groupthink” analysis of the Bay of Pigs fiasco. But people cannot just conform. If the world was all conformity, there would be no change, no resistance. No hope.

It was left to other psychologists ‑ Serge Moscovici, Gabriel Mugny, and Charlan Nemeth ‑ to study dissent, resistance to conformity, and minority influence. Their work has not had the high public recognition that Milgram’s has enjoyed. But it deserves to be better known.  

Milgram seems to have been an eccentric and difficult academic with a line in sarcasm and a tendency to overwrite. I get the sense he was an acquired taste and that not everyone acquired it. He taught at CUNY because neither Yale nor Harvard had warmed to him. And at CUNY, he was “Professor Milgram” to his students and they were Mr or Miss Whatever to him. Once, Blass says, he chose to sing his seminar, insisting that his students likewise participate in song. And once he had them grade one another, insensitive to the distress this might cause.

Blass struggles, I think, to find instances of kindness on Milgram’s part, even towards his own family. “An equal opportunities insulter”, one student described him. But brilliant with it. After a summer in Paris his French was so fluent, so polished, people took him for French. The child of immigrants, short of money, he got himself to where he was. From The Man Who Shocked the World, he comes across as someone of massive but unfocused ambition, who maybe did not take to specialisation ‑ psychology one day, fiction another. If, in some places, Obedience to Authority reads like The Mighty Thor, circa 1965, in other places it is tight and to the point:

In modern society others often stand between us and the final destructive act to which we contribute. Indeed, it is typical of modern bureaucracy, even when it is designed for destructive purposes, that most people involved in its organisation do not directly carry out any destructive actions. They shuffle papers or load ammunition or perform some other act which, though it contributes to the final destructive effect, is remote from it in the eyes and mind of the functionary.

The impression I get from Blass is that Milgram was interesting but maybe did not do enough to stretch to a biography. There was the obedience work, then nothing much after. His later research on six degrees of separation, mental maps, and cyranoids is intriguing, but it is more like notes for projects rather than fully finished research.

In the 1970s, social psychology began to look at social phenomenon such as culture and collective identity to explain behaviour. In parallel, there was a move from experimentation towards reflection, and a shift in the locus of control from North America to Europe. That might have suited Milgram. There is a Europeanness to his work, a desire to do more than just experiment ‑ to explain, engage, be relevant, a public intellectual, like Serge Moscovici, with whom I learn from Blass he was in regular contact. And there, at least, I think, his obedience work succeeded. It touched on real life issues. It got people talking. And they are talking still.


Martin Tyrrell studied psychology at Queen’s University, Belfast, and at Yale. His writings on group psychology and collective identity have been published in the journal Critical Review and the anthology Political Knowledge (Routledge, 2013). Behind the Shock Machine: the untold story of the notorious Milgram psychology experiments, by Gina Perry, is published by Scribe Publications. The Man Who Shocked the World: the life and legacy of Stanley Milgram, by Thomas Blass, is published by Basic Books.