The list of experiments well-known as the Such-And-Such Experiment is pretty short. Off the top of my head, I can think of the Michelson-Morley Experiment, the Miller-Urey Experiment, and Millikan’s Oil-Drop Experiment. A few others come to mind, but either don’t have an apt name or aren’t really experiments, per se: Eddington measuring the stars moving during an eclipse, Rutherford and his gold foil, Pavlov’s dogs, a number of things Galileo did, Mendel’s peas, and the Manhattan Project.
The Stanford Prison Experiment definitely qualifies. It is a famous experiment conducted by Philip Zimbardo in 1971. It put twenty-four randomly-assigned college students in a simulated prison setting for two weeks. Zimbardo got the actual local police to “arrest” the students selected to be prisoners, and then had them turned over to the students selected to be guards in a fake prison constructed in the basement of the Stanford University Psychology building. Zimbardo played the prison superintendant, and one of his research assistants played the warden.
What Went Down
The guards were explicitly told that they could not hurt the prisoners. Despite this instruction, the situation quickly got out of hand. The prisoners rioted on the second day, due to terrible treatment from the guards. The guards voluntarily spent extra time in the prison to suppress the riots, attacking the prisoners with fire extinguishers without provocation from Zimbardo or his assistants. Sanitary conditions deteriorated, exacerbated by guards not letting prisoners go to the bathroom. Some guards punished prisoners by forcing them to walk around naked.
Rumors of a breakout led to the building of a second prison and an attempt to move the experiment to an actual prison, which the real police would not permit. By the sixth day, an outside observer finally questioned the morality of the situation (due to the filthy conditions) and the experiment was halted early. The researchers reported that one-third of the guards had behaved in a genuinely sadistic manner, and most were upset to see the experiment end early. Zimbardo himself cited his own absorption into his role as prison superintendant.
Many of the prisoners exhibited signs of emotional trauma. Some did leave the experiment early, but others who requested but were refused “parole” decided to stay rather than quit. Zimbardo cited this as internalization of their role as prisoner; if they wanted to leave (since they asked for parole), why would they not just quit the experiment (which would have gotten them out of there immediately)? Similarly, the guards were said to have internalized their roles, per their shocking behavior.
What Was Concluded
The participants (both guards and prisoners) in the Stanford Prison Experiment were the male applicants considered most stable and healthy by the researchers. As such, the stark and horrifying results are often cited as evidence that perfectly normal people will do awful things when put in an awful situation. It seems that the situation caused the participants’ behaviors, as opposed to the participants acting in accordance with their own predispositions.
These results have huge implications for how society punishes people for their actions during extreme situations. The most obvious example would be the holocaust: were the guards in the concentration camps just victims of circumstance? Would you or I do the same thing, if put in a similar situation? What about the torture and abuse at Abu Ghraib? Professor Zimbardo himself famously testified on the behalf of the guards, citing the similarities between the experimental and real-world situations. Indeed, some of the similarities are rather striking; I will leave you to investigate the details of either on your own. Zimbardo dismissed the army’s blaming the scandal on “a few bad apples,” eventually penning the book The Lucifer Effect: Understanding How Good People Turn Evil about Abu Ghraib and his experiment.
What Should Give Us Pause
If you’re surprised that something like the Stanford Prison Experiment was even permitted to occur at a university, you aren’t alone. The experiment was wildly unethical, even by the standards of the time. The experiment has never been and never will be repeated, for fairly obvious reasons. This is a problem for the experimental results, as they can never be replicated. We’re ultimately dealing with a single data point. But, what an amazing point it is!
Well, we need to dial back that amazement once again. The study has some serious scientific problems, as well. Most egregiously, Professor Zimbardo acted as the leader of the prison during the experiment. A scientist should strive for impartial observation as much as possible, so that the act of measurement does not impact the results. Zimbardo, by directly instructing the guards prior to and acting as their leader during the experiment, could hardly have influenced the results more.
This fact alone casts serious doubt upon the reliability of the results, but some digging makes it even worse. Zimbardo went out of his way to dehumanize the prisoners, and encouraged the guards to: scare the prisoners, make their lives feel arbitrary, eliminate all privacy, and emphasize that the guards control their entire life. On his website dedicated to the study, he cites historical examples where such things have occurred. This is a ludicrously circular approach. If you’re trying to prove that regular person A will do awful thing B in situation C, why would you explicitly encourage B in your simulation of C?
It gets worse. A variety of the experimental results were anecdotal and subjective. The selection of the “most normal” of the applicants to participate in the study was not well-described in the study, and so cannot be analyzed by other researchers. Even according to the subjective anecdotes, most of the guards did not participate in the awful treatment of the prisoners. Zimbardo made no effort to explain these guards’ behavior, rather focusing on the guards who were abusive. These and many other critiques were laid out by Erich Fromm, and his language is telling. Academics aren’t supposed to say “those guys are totally wrong and they’re probably effing idiots, to boot.” So, they say things like “there seems to be an unnecessary lack of precision when it should have been very easy” and “puzzling results” and “[that] is all the more regrettable” and “I doubt the experiment proved this thesis” and so on.
But wait, there’s more. We now know that most of the abuse came from only one guard, dubbed “John Wayne” in a misplaced reference to the movie Cool Hand Luke (a Paul Newman film). Some claim that this person has since admitted that he was both actively mimicking the warden from Cool Hand Luke (“what we’ve got here is … failure to communicate”) and was responding to positive feedback from his play-acting superiors (Zimbardo and his research assistant). I wasn’t able to verify this online, but that’s not terribly surprising because I think the source is dated investigative journalism. I did find an interview with Zimbardo and John Wayne, during which Zimbardo asserted that the other guards were “just as bad” as John Wayne because they didn’t stop him from acting this way. Well, then! Whatever you say, Professor! I rest my case.
What to Make of the Stanford Prison Experiment
I’ve heard both skeptics and psychologists (my fiancé is a clinical psych grad student) dismiss the Stanford Prison Experiment out-of-hand. They most certainly have two extremely strong points: (a) it is completely irreproducible, and (b) Zimbardo brazenly impacted the outcome of the experiment. I personally don’t think it’s appropriate to stop there, simply because the results were so striking and remain so well-known. Don’t get me wrong; I’m not crediting the results with any scientific legitimacy. I just think that they deserve a deeper investigation (translation: a harder smack-down).
And, damn, does that inquiry find some dirt! This experiment really was a catastrophe. From conception through execution and onto analysis, Zimbardo did just about as poor a job as a scientist can do. Probably the most shocking thing about this entire situation is that his reputation doesn’t seem to have been sullied whatsoever. On the contrary, he’s apparently one of the most highly respected psychologists out there. He was elected president of the American Psychological Association in 2002, for crying out loud! Now, I don’t know anything about his other pursuits and accolades, but part of me wonders how this wasn’t career suicide for the guy. (I’ll be sure to give my fiancé and her friends crap for this … silly soft scientists.)
One last point that warrants mentioning is some speculation a few people have made about Zimbardo’s motives. I’ve heard both Barbara Oakley (during an interview on the Point of Inquiry podcast) and Brian Dunning (on his Skeptoid podcast) suggest that Zimbardo’s childhood in the ghettos of New York City convinced him that horrible situations make good people do bad things. This does makes sense: if all your childhood friends grew up to be awful people, it would be nice to think that the terrible situations they were in caused their downfall (as opposed to something inherently bad about them). While this is an intriguing explanation for why he conducted such a biased experiment, I hesitate to agree with them here because it seems like idle speculation on their part.
Most importantly, worrying about Zimbardo’s specific biases is unnecessary. The results fail on their own lack of merit. The Stanford Prison Experiment was simply junk science. It’s a damn shame that one of the few truly famous scientific experiments is such bad science.