December 21, 2008
Experimental Subjects Obey Killer Orders Just As In 1961

Think over the last 50 years that society has gone through huge changes that make people less willing to blindly follow authority? Think that the 1960s were a big turning point that caused people to question authority? Think they passed this down to their children and altered our relationship to authority figures? Nope. In a repeat of Stanley Milgram's classic experiment on obedience in experimental subjects the subjects of today are just as willing to deliver lethal electricity doses.

WASHINGTON Nearly 50 years after one of the most controversial behavioral experiments in history, a social psychologist has found that people are still just as willing to administer what they believe are painful electric shocks to others when urged on by an authority figure.

Jerry M. Burger, PhD, replicated one of the famous obedience experiments of the late Stanley Milgram, PhD, and found that compliance rates in the replication were only slightly lower than those found by Milgram. And, like Milgram, he found no difference in the rates of obedience between men and women.

Burger's findings are reported in the January issue of American Psychologist, the flagship journal of the American Psychological Association. The issue includes a special section reflecting on Milgram's work 24 years after his death on Dec. 20, 1984, and analyzing Burger's study.

Burger conducted his experiment at Santa Clara University. Milgram originally conducted his experiment at Yale. My guess is that a lot of Yale students would recognize this experiment and therefore refuse to play along. But it would have been a lot more interesting if a group of professors had gotten together and repeated this experiment at several universities before publishing their results.

Stanley Milgram was an assistant professor at Yale University in 1961 when he conducted the first in a series of experiments in which subjects thinking they were testing the effect of punishment on learning administered what they believed were increasingly powerful electric shocks to another person in a separate room. An authority figure conducting the experiment prodded the first person, who was assigned the role of "teacher" to continue shocking the other person, who was playing the role of "learner." In reality, both the authority figure and the learner were in on the real intent of the experiment, and the imposing-looking shock generator machine was a fake.

Milgram found that, after hearing the learner's first cries of pain at 150 volts, 82.5 percent of participants continued administering shocks; of those, 79 percent continued to the shock generator's end, at 450 volts. In Burger's replication, 70 percent of the participants had to be stopped as they continued past 150 volts a difference that was not statistically significant.

"Nearly four out of five of Milgram's participants who continued after 150 volts went all the way to the end of the shock generator," Burger said. "Because of this pattern, knowing how participants react at the 150-volt juncture allows us to make a reasonable guess about what they would have done if we had continued with the complete procedure."

This experiment did not exactly mirror Milgram's original experiment. So it is not a true comparison of attitudes now and almost 50 years ago. We really need experiments like this that compare different types of people in different settings. Will people all around the world do the same? Will smart and dumb people do the same as average people? How do young and old differ? Conservative, liberal, and libertarian, communist, socialist, and Episcopalian? Are the people who refuse to go all the way up in voltages wired up differently due to their genes? Are they otherwise better or worse citizens than those who followed orders up to a high voltage?

Also, are highly empathetic people more or less likely to turn up the voltage all the way? Are extroverts or introverts more likely to follow instructions even when they suspect someone is suffering as a result? We are eventually going to know much more about biological mechanisms as well as developmental influences and environmental factors that cause differences in behavior toward authority and differences in empathy, altruism, and many other traits. Once we know more we are going to be faced with a very difficult problem which we do not face today: what to choose once we can choose more traits for offspring?

Share |      Randall Parker, 2008 December 21 10:28 PM  Brain Obedience Authority


Comments
Shreela said at December 21, 2008 10:59 PM:

Perhaps I'm a bit more jaded than when I took psych 101 many years ago, but I wonder how many participants secretly enjoyed the delivering the "painful punishments". All their pent up resentments against whoever wronged them, redirected at someone they can't see, but they can always claim they were "made" to do it, giving them an out.

CM Collins said at December 22, 2008 3:51 AM:

"We really need experiments like this that compare different types of people in different settings."

No doubt. And not just psychologically. Given the recent oil price spikes we might do well to re-test subjects for tolerances for cold. Maybe we could start by seeing how long it takes to freeze to death some fat dumb creationists, 'cause, you know, they don't believe in science anyway so we won't have much use for them in our brave new world.


Assistant Village Idiot said at December 22, 2008 12:01 PM:

"...fat dumb creationists" don't believe in science anyway...

That is an unwarranted generalization from limited data. To believe that a supernatural being can overrule science may not be persuasive to you or me, but it does not tell us anything about a person's general knowledge and regard for science. You are confusing a cultural difference with a logical one.

Your comment does tell us a bit about you, however.

Shannon Love said at December 22, 2008 2:18 PM:

I've always found the Milgram experiments dubious at best. The circumstances of the experiments are far divorced from anything one might find in the real world. For example, since the participants known they're in a an experiment, they don't take everything they see at face value. Are people actually being "obedient" or are we just slapping a label on their behavior based on the experimenter's preconceptions of what the experiment is supposed to measure?

I am reminded of the findings in economic psychology. In experiments with highly abstracted circumstances, participants seldom reach the "rational" optimum outcome. However, as the circumstances grow more and more realistic, the participants choices become progressively more optimal. In short, I don't think the Milgram experiments measure people's obedience to authority but rather just their behavior in the artificial world of a psychology experiment. If the experiment was more realistic, for example, if the delivery of the shock and the authority figure didn't seem to be related to experiment in the first place, then the results might have more general application.

It certainly doesn't help that Milgram's experiment seems to always be held up as an example of how stupid and amoral ordinary people are. People who make such implications usually follow up with the idea that ordinary people need to be over seen by some wise and benevolent elite.

th said at December 22, 2008 2:26 PM:

how awful, you'd think with all the govt school liberal indoctrination this would've improved at least a bit, oh well, alls well that ends well, keep tryin'.

homunculus said at December 22, 2008 2:36 PM:

We are eventually going to know much more about biological mechanisms as well as developmental influences and environmental factors that cause differences in behavior toward authority and differences in empathy, altruism, and many other traits.

You know, it's takes quite a leap to assume that the subjects' behavior arises deterministically in these cases. This wasn't a simple preference or scalar response test. It involved multiple stages of conscious decision making. Which biological mechanism, developmental influence, or environmental factor is it that would be monitoring this process to control the subject while also causing them to believe that they are reasoning about the process as it unfolds?

tim maguire said at December 22, 2008 2:43 PM:

While I don't think human nature has changed a bit since we came down out of the trees (really, the 60's? Hippies are supposed to have made us all better people?), I think there's a lot of truth to Shannon's observations.

We are not talking about the real world here, we are talking about a study under controlled conditions. It's perfectly reasonable for most or even all the participants to assume that there is no real danger because the university would not take the liability risk and live subject studies are subjected to layers of review and approval. Despite the experimentor's assumptions, it may never have occurred to any of these people that they were really hurting anyone.

Byron said at December 22, 2008 2:50 PM:

If you watch films of the original Milgram experiments, you will be less likely to conclude that the behavior observed was some artifact of
"being in an experiment." The setting was very realistic and, most important, the subjects were ordinary people from the community, not college students who had heard about psych experiments. The most important thing for the level of obedience shown was probably the fact that the experimenter assured the subject that he, the experimenter, was responsible for anything that happened in the experiment. The subject acted merely the agent of the experimenter, not as the responsible party. It's not so surprising that people will do questionable things when they see their own role as merely ancillary; it's the way a lot of policies end up being carried out.

Byron said at December 22, 2008 2:59 PM:

One comment in response to Tim Maguire. It's well to remember that in those days there were no such things as institutional review boards or human subjects committees. Those did not become common until after about 1970. Many, many experiments involved procedures (such as giving undergraduate subjects, without their knowledge, epinepherine injections to create emotional agitation) that could not be used today. If you were a subject in an experiment, you had no reason to expect any guarantees about your, or anyone else's, safety. That seems somewhat incredible today, but that's the way it was.

Byron said at December 22, 2008 3:16 PM:

Milgram also replicated his procedures without any apparent connection to Yale. A rather questionable looking office calling itself, if I remember correctly, "Research Associates of Bridgeport" was opened in a run-down industrial area. Obedience levels observed were not significantly different from those found when the experiments were run in the Yale labs. The Milgram results are not going to be explained away as experimental artifact. I understand there was another version run later (not by Milgram, and not in the US), in which obedience consisted of delivering painful electric shocks to puppies; there may be a reference to that in Milgram's later book, "Obedience to Authority."

We are, after all, a social species whose primordial mode of existence was strongly hierarchical. Little about social life is possible without something like authority relations, and our species does not survive outside of social life. That doesn't mean that people can't learn to resist authority, but it does suggest that doing that is not our natural proclivity. The default setting toward which we tend is likely in the other direction, which is what troubled Milgram and the research tradition out of which he came. That included studies of people's tendency to conform by Solomon Asch, as well as work on what was termed the Authoritarian Personality.

Shannon Love said at December 22, 2008 3:26 PM:

Bryon,

If you watch films of the original Milgram experiments, you will be less likely to conclude that the behavior observed was some artifact of
"being in an experiment."

I really don't think you can remove someone from the context of being in an experiment. After all, any experiment is a strange and unfamiliar experience. In college I spent half and hour counting spoons, something I would normally never do. Does that indicate I am dangerously obedient to authority because I followed the researchers directions to carry out an inane act? People in these experiments have no context for their actions as they would in the real world. Given the complexities of human moral reasoning, I simply don't think that the way that people behave in this artificial and alien environment is indicative of their normal behavior.

For that matter, I can't really think of any real world circumstances in which people just suddenly begin doing horrible things to strangers just because some authority figure told them to. If you look at the behavior of people in authoritarian regimes for example, people don't just blindly follow orders. Instead, they follow orders because they buy into an elaborate and highly defined world model advanced by the authority figures. Germans didn't murder Jews and other "untermensch" just because Hitler told them to. They murdered Jews because they believed the Nazi arguments that they constituted an active threat to the well being of the German people. Stalin, Mao, Pol Pot etc all likewise created convincing explanations of why people had to die.

In short, I think the mindless obedience is just an artifact of the experiment itself and doesn't really tell us how people will react in a real world circumstance.

Thomas Casey said at December 22, 2008 3:41 PM:

The Milgram and Zimbardo experiments only told us what we already knew but didn't want to look at squarely. A book on the German extermination units' activities on the Russian Front in 1941 was titled "Ordinary Men". Those "ordinary men" were quite capable of extraordinary brutality.

And it certainly didn't start with them. LTC Dave Grossman wrote major works, "On Killing" and "On Combat", in which he expressed his belief that not only are we capable of these things (see Abu Gharaib) but now encourage them in numerous ways; not least of which are gratuitous violence in movies, TV, music,and video games.

With over 26 years' experience in the military and law enforcement--I believe it.

Shannon Love said at December 22, 2008 3:52 PM:

The best way to understand the weakness of Milgram's experiment is to reflect that the subject's belief that the authority figures had the circumstances under control and would not allow the people being shocked to come to any real harm, was in fact correct. In other words, the subjects accurately judged that the experimenters would not actually allow the recipients of the shocks to come to harm.

The Milgram experiment only seems "shocking" if you buy into the idea that the subject's sole source of information was the feedback they received from the feigned cries of the experiment's "victims". In reality, the subjects also had a vast store of information such as their knowledge that anyone who actual did what the experimenters claimed they were doing would face criminal and civil penalties. Based on this store of knowledge, the subjects correctly predicted that the subjects were not being harmed, even if the evidence immediately in front of them said otherwise.

Like the economics experiments I mentioned above, the subjects actions appear more reasonable the more the considered parameters approach that of the real world.

Byron said at December 22, 2008 4:00 PM:

Shannon: You need to watch the films. These subjects did not obey "mindlessly." They resisted to varying degrees, often quite strenuously. But with enough assurances from the experimenter, most (not all) finally obeyed his instructions, often with repeated protests that they were not responsible, and with much evident emotional upset. They did not see themselves as independent actors, but rather they became the experimenter's agent. This is not different from a nurse or hospital orderly carrying out orders that cause great pain and distress to the patient. Think about the sorts of procedures that have been carried out routinely in mental hospitals. Nor was the experimental context inane; it was presented to subjects as a study of the effects of punishment on learning, a completely plausible story. Most important, again, was that these were ordinary people from the community, not university types who had learned to be suspicious.

It would be comforting, indeed, if what Milgram found could be discounted as experimental artifact, but no. Later research suggests that resistance to authority is enhanced by independent knowledge about what is going on. The more that nurses know about a particular drug and its standard dosages, the less likely they are to accept a doctor's order to administer what would be a massive overdose. Milgram's subjects knew nothing about the psychology of learning, and that no doubt increased their vulnerability to becoming obedient agents of the experimenter.


sceptical but not cynical said at December 22, 2008 4:06 PM:

Milgram's methodology conflates moral authority with technical authority; there is no way to know from this whether respondents were reacting more to the one than to the other.

In a world where coffee shops regularly post "warning, coffee is hot!" disclaimers, people are accustomed to discounting posted dangers. When a technical expert - e.g. an electrician - tells you, or refrains from admonishing you, that a given dose of electricity is dangerous, who you gonna believe: the living, breathing expert or yet another warning label? I suspect most people'd go with the expert, cuz that's what we usually do. Signage has the lesser credibility.

Milgram's experiment has one person playing the role of authority for both the technical and the societal positions. Subjects might well believe that the electrical dose is not really as dangerous as purported. And as for the screams etc, I suspect that most people don't relish it, but they are loathe to ruin an experiment over what they have been told is merely transient physical pain. Previous commenters' points about this being an experiment in a lab setting are also valid here.

Perhaps the experiment could be redone with two authority figures: one would be the societal authority, the other would be a 'mere' technical expert. Have the tech expert acknowledge to the subject that the next dose of electricity has a high probability of being fatal, have the societal authority then push for the subject to continue, and then compare the results to those obtained by Milgram. I suspect the numbers would show quite a few less subjects who would push beyond the safety limits, once they were technically convinced that those limits were real and not just formal legal disclaimers.


sceptical but not cynical said at December 22, 2008 4:17 PM:

Further to my previous comment: IIRC, firing rates, even in battle in the Civil War, were very low, with at times three quarters of soldiers unable to fire their weapon on other human beings, even as they themselves were being shot at and admonished by their own commanders to return fire.

Here we have a real-world experiment, not done in a lab, that gives the opposite results to Milgram's, and it's hardly a one-off. Milgram's lab methodology is missing something when compared to this and to other real-world situations. Some good people can indeed be hardened and made cold-blooded, as happened with the Nazis, but this takes more than just a few minutes in a psych lab with an insistent researcher.

Shannon Love said at December 22, 2008 5:37 PM:

Bryon,

It would be comforting, indeed, if what Milgram found could be discounted as experimental artifact, but no.

Well, again, I would point out that the subjects did correctly determine that the experimenter was in fact in control and that the supposed victims were not actually hurt. A person in that circumstance brings with them an entire body knowledge external to the artificial world of the experiment. Just because the experimenter arbitrarily decides that the subject is only operating on the information that the experimenter provides does mean that the subjects are actually doing so. The subjects brought with a wealth of knowledge about the limits of actions that people in our society can take and that information told them, accurately, that the experimenters would not hurt the subjects.

Byron said at December 22, 2008 5:40 PM:

Sceptical but not cynical: You miss the point of the Milgram experiments. Milgram did not claim that his procedures made people hardened and cold-blooded; in fact, their emotional and verbal responses were very much contrary to any such notion. All Milgram felt that he had demonstrated was that many people will act without limitations of conscience when ordered to do so by a legitimate authority. That by itself was hardly news. What was news was that you could demonstrate that kind of thing in a relatively low-potency laboratory setting involving a mere scientific authority. Those results suggested that the phenomenon was pretty robust, and that it was not restricted to settings like military commands, where authority figures have enormous power over the whole range of the subordinate's existence. In other words, obedience does not depend on turning someone hardened and cold-blooded, and it does not depend on his being subordinated within a massively powerful, controlling network. This is what made the Milgram experiments so troubling.

The example of soldiers not firing at the enemy is not analogous to Milgram's situation, because commands are in that case spread over many individuals whose performance cannot be observed in detail. It is not possible for a battlefield commander to know who is aiming to hit the enemy and who is purposely aiming to miss. In the Milgram situation, it was one-on-one, with the experimenter monitoring every response the subject made or refused to make. In fact, one form of disobedience consisted of the subject misreporting the shock level he administered, announcing a higher level than the one actually delivered.

A later experiment (not Milgram) was done in a field setting, not a laboratory. A nurse on duty on a hospital ward received a telephone call from doctor she had never heard of, instructing her to administer to a patient a drug she had never heard of. The label showed that the instructed dose was many times the maximum recommended. Nearly all nurses administered the drug (placebo, in fact) without question and without seeking advice from any third party. The point is not that these nurses were hardened and cold-blooded, but only that they were willing to substitute the judgment of a legitimate authority for their own, and to act as the agent of that authority. That's all the Milgram experiments sought to demonstrate, and they did. Again, everybody already knew that, but the new ground broken was to demonstrate how little it takes to get someone into the role of agent. That is, after all, how organized evil gets done. Once personal responsibility is off-loaded, many actions become possible; the essence of the authority relation is that substitution of responsibility. What makes it all more difficult is that most of the good, productive things that get accomplished in society depend on authority relations, also.

Byron said at December 22, 2008 5:55 PM:

Shannon wrote: "Well, again, I would point out that the subjects did correctly determine that the experimenter was in fact in control and that the supposed victims were not actually hurt."

Really? Where are you finding the evidence for this claim? The post-experimental interviews and de-briefings suggest nothing of the kind, either among subjects who obeyed or among those who refused to obey.

"Andrew Meyer" said at December 22, 2008 5:56 PM:

Don't tase me, bro!

Gabriel Hanna said at December 22, 2008 6:37 PM:

I don't know much about Milgram but I have read about the Einstazgruppen teams who killed Jews in Occupied Europe. They were normal guys, but quite a few did refuse outright, with many more "calling in sick". Of those who didn't, most of them had to be drunk, and they usually didn't last long. See Michael Burleigh, "The Third Reich: A New History". This book emphasizes the Aktion T4 euthanasia programs and how they evolved into the Einsatzgruppen.

One reason for the Einsatzgruppen was that the regular military, with some notable exceptions, objected strongly to carrying out murders of civilians and insisted the SS do it. This was because it was so destructive to morale, not because the military cared what happened to Jews.

Shannon Love said at December 22, 2008 9:15 PM:

Bryon,

Really? Where are you finding the evidence for this claim? The post-experimental interviews and de-briefings suggest nothing of the kind, either among subjects who obeyed or among those who refused to obey.

My point is that Milgram wasn't actually shocking people and I think it likely that people unconsciously understood this was probably the case given their contextual understanding of how our culture and society works. There subconscious model was in fact the correct one. In other words, their behavior was largely indistinguishable from someone who believed that the experimenters would not let the victims come to any real harm.

Milgram et al could not control for the influence of this kind of broad contextual information and mere subjective reports by subjects would not reveal it because the subjects themselves were largely unaware of it. He would have to create some kind of circumstance in which the broad social and cultural information would not play a role.

I am dubious in large part because we simply don't see this kind of behavior in the real world. If people were really so quick to defer to authority I think that con artist and others would make use of it. (I know of only once such scam, the Treasury agent scam) Moreover, I think we would see people accidentally obeying by carrying out orders they mistakenly believed where given by an authority figure.

Randall Parker said at December 22, 2008 9:21 PM:

Shannon Love says:

The circumstances of the experiments are far divorced from anything one might find in the real world.

Just because some set of circumstances hasn't happened up to some point doesn't mean it won't happen. If a set of circumstances will cause people to behave in a way that is desired by some group with the resources to create those circumstances then suddenly the odds of those circumstances occurring can go way up.

George S. said at December 22, 2008 9:36 PM:

Shannon, are you familiar with the Milgram experiments at all?!?! While the subject knew he was part of an experiment, the experiment was explained & carried out as memorization of word pairs, NOT as an obedience test. The subject was misled, so the true nature of the experiment was unknown to the subject. The subjects really believed they were administering pain to another person, even to the point of pressing switches labeled "Danger: Severe Shock" and "XXX". The fact that the test administrator absolved the test subject from responsibility made the experiment even more real-world. Released from all responsibility, "off the hook" legally, would the subject respond to authority or rely on his own moral compass & defy the authority? Milgram and others assumed nearly all people would refuse to obey for long, and were stunned by the results which showed otherwise.

Of the films Milgram took that I have seen, I can say they are amazing (as others have pointed out here). I also heartily recommend "The Man Who Shocked The World" by Thomas Blass, which is a great read. Blass included some partial transcripts involving the test subjects, which DID confirm the subjects believed they were causing harm to another person in the experiment. Do the "Look Inside" feature at Amazon & search on "XXX" or "pain" and see some of the pages. To say that most of the test subjects saw through the aims of the experiment is not at all true.

Anne said at December 22, 2008 10:23 PM:

I have to agree with Shannon that somewhere in the subjects' minds must have been the fact that if people were being brought in body bags out of the Yale psych department, it would have made the newspapers. On some level they knew "this can't really be happening", and they were right.

I've only seen a few snippets from the movies, but what I remember is consistent with the subjects being confused. They were shifting in their chairs, sweating, kept asking "are you sure"? I haven't seen the de-briefings, but I really doubt they said, "Oh, yes, I was 100% sure that guy was going to die, but I figured when the police called I'd just say the head experimenter took responsibility and I'd be totally in the clear."

Plumb Bob said at December 23, 2008 4:03 AM:

Sceptical-but-not-cynical suggests that Milgram conflated moral and technical authority. Unless I'm mistaken, that was part of Milgram's point, that we as a culture have surrendered our moral sense to the authority of science. It was not Milgram, but rather the subjects, who invested the man in the lab coat with both moral and technical authority, and who really failed to distinguish between them.

The reason they did this was probably their sense that scientists know things they don't, and this is why the firing rates in the civil war do not test the same conditions. Most people know that an electric shock is unpleasant but don't know what levels of voltage and amperage produces what effect, and ultimately trusted that the scientist knew the things they did not and surrendered their responsibility to him. No soldier on the battlefield had the slightest doubt that his weapon was lethal, and no officer was standing there telling him implicitly, "It's ok, it won't really hurt them, it's only going to sting a little."

TBlakely said at December 23, 2008 4:19 AM:

I watched a documentary a while back about Heinrich Himmler who was in charge of the program to exterminate the Jews. The thing that really suprised me was his concern about his underlings welfare. He fully understood that the actions required by his extermination programs severely affected the morale of the personel executing his orders and ensured they had even more leave time than full combat troops. One assumes that someone who would plan and execute such a monsterous crime would be a total monster himself utterly unconcerned about the welfare of anyone else.

Randall Parker said at December 23, 2008 12:44 PM:

"sceptical but not cynical" says:

In a world where coffee shops regularly post "warning, coffee is hot!" disclaimers, people are accustomed to discounting posted dangers.

But America in 1961 was not such a world. The lawyers hadn't done their damage yet. The people of the time were not yet accustomed to excessive lawsuit-driven warnings about danger.

I think Byron gets it right here:

We are, after all, a social species whose primordial mode of existence was strongly hierarchical. Little about social life is possible without something like authority relations, and our species does not survive outside of social life. That doesn't mean that people can't learn to resist authority, but it does suggest that doing that is not our natural proclivity. The default setting toward which we tend is likely in the other direction, which is what troubled Milgram and the research tradition out of which he came. That included studies of people's tendency to conform by Solomon Asch, as well as work on what was termed the Authoritarian Personality.

What we are seeing in these experiments is a product of humanity's evolutionary genetic inheritance.

S said at December 27, 2008 11:36 PM:

A later experiment (not Milgram) was done in a field setting, not a laboratory. A nurse on duty on a hospital ward received a telephone call from doctor she had never heard of, instructing her to administer to a patient a drug she had never heard of. The label showed that the instructed dose was many times the maximum recommended. Nearly all nurses administered the drug (placebo, in fact) without question and without seeking advice from any third party.

@Byron: About what year was this experiment done? I verified drugs and dosages quite a few times when I worked at a hospital. If we didn't know the drug, we looked it up, and if the dosage was out of range for that patient, we called for clarification. Occasionally there was a reason to give a drug out of normal dosing amounts, and we'd chart that we verified this with the doctor before administering the drug.

But it's my understanding that it hasn't always been this way; sometime before my nursing school in '95, we were taught that a nurse gave Milk of Magnesia (IIRC) to a patient via sub-clavian, and that was supposedly the case that caused nurses to be held accountable for 'blindly' following bad orders.

W.Palmer said at December 30, 2008 3:53 PM:

Face it, God or no god there is a good and evil, now what?

Post a comment
Comments:
Name (not anon or anonymous):
Email Address:
URL:
Remember info?

                       
Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright