Contesting the “Nature” Of Conformity: What Milgram and Zimbardo's Studies Really Show

December 1, 2012 3:23 AM

Contesting the “Nature” Of Conformity: What Milgram and Zimbardo's Studies Really Show [FULL TEXT]

Understanding of the psychology of tyranny is dominated by classic studies from the 1960s and 1970s: Milgram's research (video of a replication) on obedience to authority and Zimbardo's Stanford Prison Experiment (documentary). Supporting popular notions of the banality of evil, this research has been taken to show that people conform passively and unthinkingly to both the instructions and the roles that authorities provide, however malevolent these may be. Recently, though, this consensus has been challenged by empirical work informed by social identity theorizing. This suggests that individuals' willingness to follow authorities is conditional on identification with the authority in question and an associated belief that the authority is right.

Standford Prison Experiment Previously

Migram Experiment Previously

posted by Blasdelb (24 comments total)

The author begs the question. It is simple tautology that someone who commits what we call evil is demonstrating that they hold something else as more important than what we call virtue.

This "insight" carries no information whatsoever.

True, massive scale horrors can only be carried out with the full complicit participation of large social groups. In order for those groups to have any cohesion they must self-identify as normal. Thus, in its own context, large scale evil will always be banal.

And most importantly, by knowing that evil will look banal from the inside we can have the introspection to identify the evils in which we are currently complicit. I mean literally, us, right now.
posted by idiopath at 3:41 AM on December 1, 2012 [16 favorites]

"Success requires leaders and followers who do not adhere rigidly to a pre-determined script. Rigidity cannot steel them for the challenges of their task "

Can someone please tell George Osborne et al about this?

In order for those groups to have any cohesion they must self-identify as normal.
Posted by idiopath.

Surely that should be "right" not "normal" - in the prison experiment, when he told them it was for the good of science, i.e. - right - they carried on, but when he said you have no choice, they stopped.

"Moreover, on the basis of this shared identification, the hallmark of the tyrannical regime was not conformity but creative leadership and engaged followership within a group of true believers...

This is very scary when I see some of the fanatics engaged in the political discourse these days.

Fantastic article, really interesting stuff.
posted by marienbad at 5:01 AM on December 1, 2012 [2 favorites]

"for science" is an exhortation that fit the group's self-identity. "you have no choice" conflicts with it. The issue is still normalcy (you just have to contend with inherited / culturally specific ideas of normal).
posted by idiopath at 5:23 AM on December 1, 2012

I have always thought that Milgram's and Zimbardo's experiments show two equally awful but very different aspects of human nature.

Milgram's subjects commit evil because their acts have been redefined as good along a different axis -- "for the good of science." There are of course very close parallels here with [GODWIN EXEMPTION INVOKED] and the nuclear arms race, as well as other lesser horrors. I would disagree with idiopath that this has anything to do with group identification; Milgram's setup was very intimate and personal. It is about specific actions and consequences. It is about agency; subjects who are told they have no choice get ornery and demonstrate that they do by quitting. Subjects who believe there is a higher good at work carry the torture to its limit.

Zimbardo's prison experiment is more about tribal identification and otherness. The guards aren't told there is a greater good at work; they are simply made a part of a different group. Here idiopath is right, and the abuse escalates mainly because the prisoners become ever less human to the guards.

IRL it is of course possible for both of these effects to manifest at the same time, with the prison guards being told that the extremes they are encouraged to enforce are serving a higher good. There is a bay in Cuba where one might likely observe this, for example.
posted by localroger at 5:33 AM on December 1, 2012 [2 favorites]

I don't much like studies that do not include IQ as a variable. IO, the branch of Psychology which studies leadership among other qualities as applied to the work environment, shows that intelligence is the predominant variant in the explanation of different human behaviors in most leadership studies and this is a leadership study.

On the other end, I'm still puzzled by why some women voted for Romney. Do I now have a choice between morons, evils, and sheep?
posted by francesca too at 5:48 AM on December 1, 2012

Do I now have a choice between morons, evils, and sheep?

Yes, you can choose to either shock them into proper behavior or cordon them off from the rest of the society and keep them from misbehaving.
posted by (Arsenio) Hall and (Warren) Oates at 6:19 AM on December 1, 2012

IQ as a variable

My wife's IQ ranged from 67 to 150 over the course of her childhood. Anyone who tells you IQ is a useful variable for anything is selling snake oil.

Also, the idea that general intelligence, however you try to measure it, is linked in any way with leadership is directly contradicted by all kinds of studies and anecdata. Some people are naturally dominant and persuasive. Many of those people are also stupid. Lots of the people who are both go into politics, and they succeed because we are naturally susceptible to their charms..
posted by localroger at 6:21 AM on December 1, 2012 [14 favorites]

I'm glad this article exists, I never really understood how Milgram/Zimbardo's experiments were supposed to be an empirical validation of our inherent propensity for conformity or evil (no control group, little to no repetition of experimental design, pretty shoddy simple random sampling, lots of confounding variables, etc etc); on the other hand, I think the argument that a reinterpretation/reexamination of these experiments counts as "empirical scrutiny" demonstrates a fundamental misunderstanding of...uh...empiricism in general. But maybe I'm too entrenched in a modern research mindset and am looking at this from an overly limited viewpoint.
posted by pugh at 6:36 AM on December 1, 2012

I have a feeling that the authors did not read Milgram's Obedience to Authority, or, at least, not too carefully. It is actually a great masterpiece of scientific writing, walking the reader through an experimental design that gets ever more damning, with one control after another. In a sense, his work, while about the banality of evil, was also about the importance of empiricism in establishing that said phenomenon is real.
posted by Blazecock Pileon at 7:46 AM on December 1, 2012 [6 favorites]

You don't hear of self-righteousness-as-evil discussed at all, but that's exactly what it is. Hell is the application of heavenly decrees and destinies on earth. Pure evil is just a majority of such idiots.
posted by Brian B. at 8:02 AM on December 1, 2012 [1 favorite]

Hell is the application of heavenly decrees and destinies on earth

A YA book recommendation related to the broader discussion and this sentence in particular.
posted by a snickering nuthatch at 8:16 AM on December 1, 2012

I have a feeling that the authors did not read Milgram's Obedience to Authority, or, at least, not too carefully. It is actually a great masterpiece of scientific writing, walking the reader through an experimental design that gets ever more damning, with one control after another. In a sense, his work, while about the banality of evil, was also about the importance of empiricism in establishing that said phenomenon is real.

Yeah, great point.

It's a little misleading that Milgram's experiments and the Stanford Prison Experiment are lumped together like this. The research has broadly similar subject matters, but the methodologies are worlds apart. Milgram's experiment was actually this small, focused design that could be repeated over and over again in different permutations, which lends itself to discriminating hypotheses and confirming findings.

The SPE by contrast was just this big, wild mess. That's not to say that the results are meaningless, but it's not exactly the "gold standard" of science.
posted by grobstein at 8:53 AM on December 1, 2012 [7 favorites]

Sorry, my comment was definitely focused towards SPE. The two are so often lumped together that I frequently confuse the stringency of their methodology when they are definitely worlds apart
posted by pugh at 8:57 AM on December 1, 2012

francesca too: "I don't much like studies that do not include IQ as a variable. IO, the branch of Psychology which studies leadership among other qualities as applied to the work environment, shows that intelligence is the predominant variant in the explanation of different human behaviors in most leadership studies and this is a leadership study."

No everyone would agree with this. The behavioral perspective has found many ways to conceptualize leadership, as well as the miligram/zimbardo stuff, without having to use the concept of "intelligence" in any way, instead focusing on how altering environmental factors have a statistical change on the behavior of random subjects.

I know that most IO programs aren't taught from an experimental science perspective, so the behavioral perspective is kinda unique in that way. So most people would probably disagree with me. But that's fine, I'm just saying that there are other perspectives out there.
posted by rebent at 9:00 AM on December 1, 2012

I also have to add, I don't actually see any experimental evidence conducted in the main article. In fact, the article lost all credibility to me when it stated

However, some of the most compelling evidence that participants' administration of shocks results from their identification with Milgram's scientific goals comes from what happened after the study had ended. In his debriefing, Milgram praised participants for their commitment to the advancement of science, especially as it had come at the cost of personal discomfort. This inoculated them against doubts concerning their own punitive actions, but it also it led them to support more of such actions in the future. “I am happy to have been of service,” one typical participant responded, “Continue your experiments by all means as long as good can come of them. In this crazy mixed up world of ours, every bit of goodness is needed” (S. Haslam, SD Reicher, K Millward, R MacDonald, unpublished data).

Here the author confuses A) The "scientific goals" that they knew about when they activated the shock lever, and B) the scientific goals of understanding tyranny. The author is saying that the reason the participants followed through with A was because they agreed with B. Which they didn't know about when they did A.

And the whole "inoculated them against doubts" doesn't make any difference - the experiment was over. They didn't go back and try it again. Their results were already observed and calculated.

So I guess what I'm saying is, this author couldn't tell the difference between an IV and a DV.
posted by rebent at 9:30 AM on December 1, 2012 [4 favorites]

This suggests that individuals' willingness to follow authorities is conditional on identification with the authority in question and an associated belief that the authority is right.

This is where might enters the equation. If the authority has enough of it, he becomes right ... or you're off to the gulag, the concentration camp, the torture chamber, the retirement home ...

The issue is still normalcy (you just have to contend with inherited / culturally specific ideas of normal).

the trouble with normal is only gets worse.

I should now read the article.
posted by philip-random at 9:57 AM on December 1, 2012

Milgram's project indicated that people would furnish their own rationalizations for specific acts which they presumably wouldn't perform in a different context. Then, afterward, they further buttressed their belief that the deed they performed was necessary for the greater good. Not stated was that post-experimental fervor may have been proportional to the dramatic profile of the experiment. Point here is that the individual brought his own vision of "the good" being performed, and required only the vaguest hook to hang it on. When he was compelled to administer the shocks, he refused, indicating that he'd refocused his reality back to include his personal values. Not explored was the link between the personal paradigm and the institutional paradigm. Was it just the prospect of adventure?--maybe it was the small fee they were paid, and once they decided to accept the fee, the process of rationalization led them over to the dark side.

The SPE was a clusterfuck that nevertheless revealed something useful. The two groups made an identity adjustment--ritualized by the changing into "uniforms"--and they performed according to their individual visions of how members of that group ought to perform. Interesting here is that they engaged in a bit of extemporaneous briefing among themselves: prisoners as prisoners and guards as guards, and came up with rules that more or less fit the circumstances--all this in just a few days. The disturbing part of this for me was how easily the two groups slipped into roles. Are we who we think we are? Is it that simple? But what else this implies is more disturbing, and relates to, let's say, putting lipstick on a pig--which is bad enough--but then going on to kiss the pig.

Common to both experiments was the idea that none of the participates merely went along with the program blindly. They all knew what they were doing, but justified it. Nowadays we tend to forget that, back in the 1950's, "brainwashing" was a popular theme. Commies were robots under some sort of propagandistic narcosis, mindlessly doing the bidding of their leaders. (Except for the few who huddled in thier basements and listened to The Voice of America). The War Crimes trials were quite adept at dismember the "we only followed orders" defense promoted by trial lawyers. The Nazi upper echelons found it didn't work, but the average German soldier sort of slid by the wayside, legally speaking. This doesn't speak well for the legal system, but it does reflect the problem of putting a million soldiers on trial. Also, brainwashing was a topic relating to the Asian shores, famously illustrated by Laurence Harvey in the movie "The Manchurian Candidate." The plot pot boiled right along in the movie, but it did touch on the facts, namely that many Americans were subjected to "brainwashing" by the North Koreans. In fact, but not explored in the movie, quite a few remained in North Korea when they were release from the POW camps after the truce had been struck. You have only to remember Harvey turning the cards over to get the point. His handlers were well aware of the big picture, however.

This whole mindset amounts to a paradigm: Why we do what we do--robots or what?

The SPE group was the most disturbing on one level, at least for me. I asked myself what I would do. It's easy to think I'd just drop the pose and abandon the experiment, but the broader implications are that I would not. I would rationalize, and try to contain certain acts, but I would try to keep the wheels on the cart, and move the project along.

The Milgram experiment didn't bother me as much, because I have been there and done that, so to speak. Yet it disturbed me on another level because of all the other people who get sucked into stuff like this without bothering to test the theories with a little common sense. You can see that our personal blind spots let us off the hook to a large extent. Test this by the statements: Everyone thinks he's a good driver. My breath doesn't stink. My politics are correct.

...and so on.
posted by mule98J at 10:06 AM on December 1, 2012 [1 favorite]

FINAL PARAGRAPH OF THE CONCLUSION:

At root, the fundamental point is that tyranny does not flourish because perpetrators are helpless and ignorant of their actions. It flourishes because they actively identify with those who promote vicious acts as virtuous [49]. It is this conviction that steels participants to do their dirty work and that makes them work energetically and creatively to ensure its success. Moreover, this work is something for which they actively wish to be held accountable—so long as it secures the approbation of those in power.

GODWIN'S been invoked already ... so I guess I'm allowed to acknowledge the salad days of the Third Reich here.

I read this paragraph and two things come to mind:

1. it absolutely speaks to likes of Eichmann, SS types and any number of ambitious middlemen who did their duty, served the fuhrer, furthered the higher ideals of blood + soil. And no doubt earned promotion etc as a result.

2. it doesn't speak to the mass of everyday citizens and soldiers (millions of them) who were required to further the absurdly ambitious (ie: batshitinsane) goals of the Reich. Here, I think you have to acknowledge that routinely human complexity that has allowed every conquering army (and the nation/tribe behind it) since pre-history to go forth and conquer. On some level, the troops really are just following orders and, given their druthers, would prefer to be hanging out on a beach somewhere, drinking wine, chasing girls (and boys), singing songs, NOT massacring entire towns, cities, bloodlines ...

but the average German soldier sort of slid by the wayside, legally speaking. This doesn't speak well for the legal system, but it does reflect the problem of putting a million soldiers on trial.

Which, I guess, is why the Nuremberg Trials were taken so seriously by the Americans et al. It was the closest they could come to putting the entire German nation on trial.

... and so on.
posted by philip-random at 10:22 AM on December 1, 2012

My wife's IQ ranged from 67 to 150 over the course of her childhood. Anyone who tells you IQ is a useful variable for anything is selling snake oil.

I agree that it is totally useless and biased if given to children. The IQ tests given to adults, while varying depending on the circumstances, do not usually vary more than 10-12 %. The IO professionals, for whom I have a lot of respect, are aware of the pitfalls of the bias and vagaries of tests and do try to compensate.
posted by francesca too at 11:22 AM on December 1, 2012

I should also have specified intelligence a bit more, indicating its problem solving component, which is not the total aspect of intelligence.
posted by francesca too at 11:27 AM on December 1, 2012

My IQ ranged from 150 to 67 as I read this article.
posted by found missing at 11:41 AM on December 1, 2012 [2 favorites]

totally useless and biased if given to children

Odd that, since when I was in college (admittedly 30 years ago) I was taught that it was mostly useless applied to adults being a comparison of developmental age to physical age, a comparison that flattens when applied to adults.

I suppose you could just redefine it as Z-score with median 100 and standard deviation 15, in which case I think you'd be better off just using Z-scores which have a lot less baggage attached.

I consider it a given that nobody has solved the probably unsolvable problem of testing for innate talent without an experience bias. That's obviously what caused my wife's IQ to blossom as she was tested and tested and re-re-tested according to the whims of various school systems.

In any case I'd prefer to believe that it's bullshit mainly because if it's not, then I will have an awfully hard time getting my head through doors because I've never scored less than 4 standard deviations out on a standardized test in my life, and almost always more than 5. And I do not really think I am that much better than everyone else. I prefer to suspect that I'm just really good at taking tests.
posted by localroger at 3:33 PM on December 1, 2012 [4 favorites]

localroger, it means that you are really good at solving a very specific set of puzzles, the measurement of which has a bunch of important implications for a bunch of things and very little to do with what most people would consider to be generalized intelligence with a moments thought.

This is also a particularly bizarre derail as the results of the Migram Experiment were widely replicated at the time and found no meaningful correlation with IQ scores. All of the subjects of the Standford Prison Experiment were also matched equally according to IQ as one of the various measured parameters and divided into guards and prisoners by random allotment.

IQ has absolutely nothing to do with any of this.
posted by Blasdelb at 3:56 PM on December 1, 2012 [5 favorites]

IQ has absolutely nothing to do with any of this.

Yeah, this was kind of my point.
posted by localroger at 6:29 PM on December 1, 2012 [1 favorite]