Hacker News Reader

Comment by conjecTech | Hacker News Reader

I'd just like to reemphasize what conclusions you should and shouldn't draw from an experiment being invalidated. If an experiment is run that says the sky is blue, then its invalidation does not mean that the sky is not blue. It means we no longer have evidence for that conclusion. The correct answer then isn't the negation of the original study's conclusion. The correct conclusion is that we don't know or aren't sure.

Likewise, if the SPE is invalid, it doesn't imply that average people aren't capable of horrific behavior - it simply means we have less evidence to determine that than we thought we did.

View on HN

>Likewise, if the SPE is invalid, it doesn't imply that average people aren't capable of horrific behavior - it simply means we have less evidence to determine that than we thought we did.

Agreed; SPE just doesn't constitute evidence for Zimbardo's particular narrative.

>According to Alex Haslam and Stephen Reicher, psychologists who co-directed an attempted replication of the Stanford prison experiment in Great Britain in 2001, a critical factor in making people commit atrocities is a leader assuring them that they are acting in the service of a higher moral cause with which they identify — for instance, scientific progress or prison reform.

This quote, and everything actually observed during the SPE, is consistent with Milgram's work, which has been replicated a zillion times.

> This quote, and everything actually observed during the SPE, is consistent with Milgram's work, which has been replicated a zillion times.

I think you may be cherry-picking quotes here, here is another from much later in the article:

> In another blow to the experiment’s scientific credibility, Haslam and Reicher’s attempted replication, in which guards received no coaching and prisoners were free to quit at any time, failed to reproduce Zimbardo’s findings. Far from breaking down under escalating abuse, prisoners banded together and won extra privileges from guards, who became increasingly passive and cowed. According to Reicher, Zimbardo did not take it well when they attempted to publish their findings in the British Journal of Social Psychology.

Milgram's work is on the power of authority, so your quote showing that behavior did not follow the desired results without the professor present is also consistent with Milgram.

> At the end of each of the Bridgeport experiments [Condition 23], before he revealed the hoax, Williams asked each man what he thought the Research Associates of Bridgeport was and why he had volunteered. Milgram had presumably told Williams to ask this to demonstrate that subjects had no idea that Yale was behind the research. Most said they didn't know, but many assumed that it was an organization conducting scientific research. Williams probed this issue with several subjects, and it was obvious that even though many didn't have the slightest clue about the organization and what it ostensibly did, they were still prepared to trust it because of the scientific imprimatur. One man who had gone to 450 volts said that he was motivated by his six-year-old daughter.

> Man: I can only say that I was- look, I'm willing to do anything that's ah, to help humanity, let's put it that way.

> Williams: Right, that's what we're doing.

> Man: I've got- I've got a child that's a cerebral palsy child.

> Williams: Have you really?

> Man: And you know they're experimenting steadily on trying to find a cure for it. It's a sad thing.

From his correspondence we know these experiments were very personal to Milgram, motivated in no small part by his desire to understand the Holocaust. Yet these experiments were not testing obedience to generic authority, it wasn't testing obedience to military officers or gestapo thugs. The above man was compliant because he thought he was helping to advance medical science. The whole thing is farcical.

First, the results have largely been replicated by other scientists, though the ethics limits the amount of replication able to be done.

Second, presumably the people in the control group without any apparent authority figure present would also have assumed it was a scientific experiment. So I'm not seeing how you think this ruins the experiment.

Finally, one of the Nazi justifications for the Holocaust was an attempt to advance medical science. So this test would still fit his motivation, though it does muddy any conclusions you can draw.

What control group? Condition 23 was meant to isolate the role the influence of Yale played, yet it was still a setting that the participants considered scientific as evidenced by that quotation of one of the Condition 23 participants.

I don't doubt that if you replicate the experiment you'll replicate the results. But the experiment isn't proving what it's said to prove. On the contrary, much of the experimental results contradict the mainstream narrative told about the experiments: http://journals.plos.org/plosbiology/article?id=10.1371/jour...

Military officers and gestapo thugs were compliant because they thought they are helping their race in struggle for survival in competition against hostile ennemy race. They also believe that it is the highest calling and expectation on manly man. Some of them thought also to advance medical science.

Gestapo were top police and believed that they are enforcing the law as they are supposed to.

>"Military officers and gestapo thugs were compliant because they thought they are helping their race in struggle for survival in competition against hostile ennemy race."

Yes, this seems likely. Which flies completely in the face with the tradition narrative of the Milgram Experiments: that people follow the orders of authority figures blindly.

What the Milgram experiments actually showed is that people are willing to do something they normally wouldn't if they believe in the cause. The teachers in the Milgram experiments believed in science and were willing to step outside their comfort zone to advance science. The Gestapo believed in their cause too, and were similarly willing to do things a man otherwise wouldn't.

In both of these cases they weren't simply doing what they were told, they were not blindly following orders as Adolf Eichmann had unpersuasively pleaded.

Adolf Eichmann was a bit too high and had bit too successful career for someone who "just followed orders". You did not even had to join SS, much less take leadership position.

Definitely agree on that one.

Well, the message people got was that students who were knowingly participating in a college psych study became spontaneously sadistic within just a day or two.

That feels intuitively ridiculous, and has no evidence I know of to support it, except this study.

The idea behind the study was to try to understand how an entire nation - Germany - could go berserk with hatred and sadism. Scientific or not, the burden of proof lies with those who dispute Zimbardo’s (and Milgram’s) findings, because they have been borne out so many times in so many different contexts. Their work has never been more relevant than it is now, as we witness the rise of Trumpism.

What alternative explanations would you propose?

1) Milgram's and Zimbardo's findings are very different. Milgram's have been backed up by many replications and are often mis-understood.

2) We have both experimental and factual evidence that shows that Zimbardo's findings were lies.

3) The entire nation of Germany did not go berserk with hatred and sadism.

4) Name one instance where Zimbardo's findings have been born out is "so many different contexts"?

The rise and success of Nazism (like most social movements, including 'Trumpism') was the result of a confluence of a number of factors. To pretend that it was just "how people are" is do both people and the truth a disservice.

Name one instance where Zimbardo's findings have been born out is "so many different contexts"?

Probably the best examples would be found in the in-group/out-group dynamics in religious cults and closed-state regimes. The Nazis exploited these principles in the 1930s to impressive effect. But what happened in Zimbardo's experiment is more closely analogous to what you see in places like North Korea today, along with abuses carried out in insular groups ranging from the Branch Davidians to the Mormons. The same hierarchies form and the same roles are played.

It is foolish in the extreme to dismiss Zimbardo's work because of a few ethical hangups. What he did is highly reproducible, but only outside the bounds of regulated academia.

Milgram has indeed come under similar criticism from people who are desperate to find another explanation -- any other explanation -- to distract themselves from the truth about what human beings actually are.

Did you seriously just compare Mormons to Branch Davidians? And North Korea? Mormonism has its problems (excellent article on that at https://longreads.com/2018/06/07/meet-the-new-mormons) but that is totally over the top. If you want to leave North Korea, you risk your life and your family's lives in a terrifying journey over the Chinese border. If you want to leave Mormonism, you write a letter saying "please remove me from your list."

I agree with you that GP's comparison was a bit drastic, to say the least.

However, you have gone too far in the other direction. You make it sound like leaving a religion is just as easy as unsubscribing from a newsletter.

Visit one of the ex-religion subreddits sometime. I know Mormonism and Christianity both have one. To be fair, those online groups will be selection biased towards the most difficult situations. But I think many of those experiences are not terribly uncommon.

Are we forgetting about the actual founder of the faith.

Either he was a crazy old coot and we can safely say that present company has moved past his behavior in which case its hard to credit a religion solely based on his revelations or he was a true prophet and we ought to analyze their nature of the organization based on his words and deeds rather than just present lukewarm implementations.

Look up blood atonement whereby killing sinners presumably including heretics was the proscribed solution.

They are taught that this results in infinite torture from birth.

At least in that regard, mormons are not so different from regular christians.

Many Christians do not do that, FYI. I know many still do, and I can't speak for all of them.

I would wager the majority of Christians believe in hell and teach that belief. Perhaps not in San Francisco... but across the country as a whole.

> Probably the best examples would be found in the in-group/out-group dynamics in religious cults and closed-state regimes. The Nazis exploited these principles in the 1930s to impressive effect. But what happened in Zimbardo's experiment is more closely analogous to what you see in places like North Korea today, along with abuses carried out in insular groups ranging from the Branch Davidians to the Mormons. The same hierarchies form and the same roles are played.

You means groups that use heavy levels of indoctrination to induce this behavior? hat is the exact opposite of Zimbardo's claim that all you have to do is randomly assign someone a role and they will exhibit this behavior without prompting.

> It is foolish in the extreme to dismiss Zimbardo's work because of a few ethical hangups.

His work is not dismissed because of "ethical hangups" about how he treated experimental subjects, but because he has repeatedly lied about and misrepresented the experimental protocol he followed. Lying about experimental protocols are some of the BEST reasons to dismiss a scientists work.

> What he did is highly reproducible, but only outside the bounds of regulated academia.

Except the study he claims to have run has been reproduced and did not have similar results.

> What he did is highly reproducible, but only outside the bounds of regulated academia.

We have evidence that he didn't actually do what he claimed to do.

We also have evidence that if you do what he claimed to do, you get different results.

> Milgram has indeed come under similar criticism from people who are desperate to find another explanation

Miligram did excellent science. Unlike Zimbardo, he did careful follow studies to try to understand and elucidate the results he found and was generally careful about characterizing the conclusions that could be drawn from his work. Milgram's studies and the conclusions he drew from them are widely misrepresented in popular culture by people eager to provide an excuse we can use to absolve ourselves of responsibility for "just doing our jobs".

> "Milgram's have been backed up by many replications"

From Behind the Shock Machine by Gina Perry:

>On the recording, a much younger Hannah sounded perky and confident as she talked to Williams at the beginning of the experiment. But you could hear her confidence getting shakier once the learner started to give incorrect answers, and it became clear that his memory was not reliable. By the third shock, 45 volts, she was stumbling over the words. At the fifth shock, 75 volts, when the learner made his first sound of discomfort, there was a pause. Then I heard the following exchange.

>Hannah: [to Williams] Is he all right? [into the microphone] The correct one was "hair". [to Williams] Is he all right?

>Williams: Yes, please continue.

>Hannah: All right.

To put it lightly, the Milgram experiments were trash. It was measuring trust in scientists not compliance to authorities. What undergrads are taught about Milgram is that the experimenter would ask the teacher to "Please continue" but what actually happened here is the experimenter assured the teacher (Hannah) that the learner was all right. The experimenters played fast and loose with the procedure, saying whatever they thought was necessary to get the teacher to continue. And sometimes it wasn't even measuring trust in scientists; there is at least one learner who recalls the experimenter, in response to their protests, asking if they'd like to swap places with the learner (essentially threatening the teacher to get them to comply.) The experimenters and particularly Milgram knew what results they wanted before the experimentation even began, and made sure they got those results.

I had not heard Behind the Shock Machine, thank you.

It's been a while, but my understanding was that the results of the Milgram studies have been fairly widely replicated with consistent results.

Now, what those results mean / what is being measured by them is another issue.

It's not unheard of though. Just likely not ad hoc repeatable, reproducible.

That said, these group experiments are very hard to control (as in control for confounding variables/factors), and very easy to screw up methodologically (maybe you don't pre-register one interesting hypothesis). Very hard to design properly (forget to control lighting), and even harder to run it well (the building where you want to do it doesn't have adjustable lighting, or you want cold water showers only, but it's not that simple to achieve, so you don't and just mention it in the paper - was that the cause of a negative/positive result or not?).

Except we do have evidence that the specific claims made by Zimbardo were fraudulent and we also have experimental evidence that shows that these results are not accurate if you actually implement his claimed methodology.

SPE now needs to be taken as a cautionary tale about ethics and science, nothing more.

For the very interested but so far unlucky: where are these specific claims and the documentation of their fraudulent nature?

Thanks in advanace!

The article goes over several. Off the top of my head: the supposed mental breakdown of Korpi (he was faking it), Zimbardo claiming he wasn't politically motivated, Zimbardo claiming the students were free to leave at any time, Zimbardo lying or misleading as to what degree the 'guards' had been coached (specifically with regard to the role of Jaffe.) Details about each of these are available in the article.

The study wasn't just invalidated, it was misleadingly characterized by the people who ran it and promoted its findings.

This does more than just withdraw some evidence in favor of the idea that ordinary people can easily commit horrific behavior. It provides new, positive evidence that people have an incentive to spread this idea fraudulently. This should properly cause us to re-examine all evidence for this thesis very closely.

Anyone who was accepted to Stanford in the early 1970s was, by definition, not an average person. That's the whole point of being accepted at Stanford.

I'm not saying it means Ivy League students are more likely to be psychopaths, but I am saying it's not scientifically valid to take a study of a population from a hyper-elite American college during a draft war and apply the results of that study to the average person, especially if the study is supposed to remark on factors regarding social pressure and social expectations.

"We knew [the guards] couldn’t hurt us, they couldn’t hit us. They were white college kids just like us, so it was a very safe situation. "

College Students Are WIERD. As the paper put it.

I think you have presented a fallacy:

> If an experiment is run that says the sky is blue, then its invalidation does not mean that the sky is not blue.

If an experiment's hypothesis is that the sky is blue and the sky is indeed blue, then there is nothing wrong and there is nothing to be invalidated.

If an experiment's hypothesis is that the sky is blue because there is a bunch of leprechauns throwing blue Skittles at it, and there are evidence that this not the case; then the experiment is to be invalidated and, in your own words, "[i]t means we no longer have evidence for that conclusion."

If you don't have the evidence to support your hypothesis, then you can't draw any conclusions except for, and at best, that there is no evidence to support your hypothesis. To put it into the context of the running example, this translates to the sky is blue but it's not because leprechauns are throwing blue Skittles at it. You can't say just because my hypothesis about leprechauns is invalidated, it doesn't mean that there are not leprechauns throwing skittles at it—because that's simply just what you believe and wish to be the case; and a good scientist should be impartial to beliefs.

More importantly, the whole point of the article is beyond invalidating the SPE. It's, in my opinion, a great piece about much deeper problems in scientific or, well, "scientific" methods.

It still could be leprechauns, though. A null hypothesis is not the same as the inverse of the H1.

The premise of the experiment isn't just that people are capable of horrific behavior. It's that the thing that brings this about is largely circumstantial and a function of our role in society.

A problem with that is that people will remain tainted by the urban legend type persistence of the conclusions. So if one could void this study that’d be a better outcome.

Reply on news.ycombinator.com