This is a famous "experiment" that "proved" institutions make people cruel. or at least that is the way the modern story is told.

But science relies on a pattern of experimentation. Did this "experiment" meet the test of scientific method?

This post will show how it did NOT do that.

The Stanford Prison Experiment was conducted by Philip Zimbardo. Zimbardo asserts (and there is a reason I note the word assert here) that the study was a continuation of the deindividuation study conducted by Stanley Milgram (Drury, Hutchens, Shuttlesworth, & White, 2012, p. 168). Zimbardo states that the experiment, referred to by Zimbardo as the SPE, was developed in an atmosphere of "situational revolution" in the field of psychology (Haney & Zimbardo, 1998, p. 709). Zimbardo further describes the purpose of the Stanford Prison Experiment as demonstrating that institutional environments influence, although Zimbardo's terminology is to "bend and twist", individuals that pass through the institution (Haney & Zimbardo, 1998, p. 710).

However, DeJong contends that the researchers did not have a clear hypothesis when beginning the Stanford Prison Experiment (1975, p. 1014). Banuazizi and Movahedi describe the methodology of the Stanford Prison Experiment as taking a random selection of "qualified" samples from a voluntary pool, then randomly dividing the selectees into two equal sets of roles; "prisoner" and "guard" (1975, p. 153). It should be noted that Haney et al nominally told the guards that preservation of order was their main role (Haney, Banks, & Zimbardo, 1973, p. 74), however, a deeper look at the experiment tells a different story. Haslam and Reicher claim that Zimbardo did not give the "guards" specific orders, but that he did give them a sense of how to behave (2012, p. 2). Gross quotes Zimbardo's instructions to the "guards" as to the creation of boredom, fear and arbitrariness within the "prisoner" population (2008, p. 22).

A research assistant took on the role of warden, while Zimbardo himself took the role of superintendant (Gross, 2008, p. 22). Banuazizi and Movahedi explain that this was to explore the interpersonal dynamics of a prison, however, a prison in which there was no dispositional difference between "prisoners" and "guards" (1975, p. 153). This lack of a dispositional difference is a point that will be returned to shortly.

Considering that Zimbardo had a predetermined purpose in mind for the experiment (Haney & Zimbardo, 1998, p. 709), and that there was no set hypothesis for the Stanford Prison Experiment, a closer look at Zimbardo's motivation must be examined. Zimbardo's bias can clearly be seen in Zimbardo's interview with Drury et al with his characterization of the death of Black Panther George Jackson during an armed escape attempt as "murder" (Drury, Hutchens, Shuttlesworth, & White, 2012, p. 161); six guards were taken hostage and three of those six guards lost their lives during the escape, a fact lost upon Zimbardo. Zimbardo did state that he "knew" of the "demonization" that took place in the prison system, however (Drury, Hutchens, Shuttlesworth, & White, 2012, p. 162)...even when he had to create that environment himself. Finally, Zimbardo's study criticized those "prisoners" that "took" the "guards'" side instead of "supporting" "rebellion" (Haney, Banks, & Zimbardo, 1973, p. 95).

It is unlikely that experiments similar to the Stanford Prison Experiment would be allowed today. In today's "sue happy" society, an institution whose IRB process sanctified a like experiment would be litigated to death by a student body horrified by "microaggressions".

If I were to proceed with such an experiment, I would do several things differently. To begin with, I would actually have a testable hypothesis. Secondly, I would not use the experiment as a vehicle for propaganda. Third, I would simulate a prison environment accurately, where the goal of the guards would be to maintain order, not to engender fear and boredom. The point in having the guards and prisoners share the same disposition would have to be modifed. In real life, people go to prison becuase they have been convicted for being unreasonable and unable to follow laws. I would have to design two sets of surveys, one for the prisoners and another for the guards, in which their attitudes. could be measured (as distinct groups) pre-experiment and post-experiment to see if the institution did create a change in personality. I may have even conducted the experiment by having a control group (either prisoner or guard) whose behavior was strictly controlled to represent real life, while measuring those attitudinal reponses changed in the experimental group.

Participant safety in the Stanford Prison Experiment was adequate. Food and shelter were provided and physical contact was prohibited. On a personal note, it seems that the "prisoners'" experience was much less stressful than Marine Corps boot camp, and I went through that before the Crucible was instituted. However, the Belmont standards arguably prohibit the intent to cause bordeom, fear, and a sense of arbitrariness through the "beneficence" standard (The Belmont Report, 1979, para. 14).

Gross discusses four relevant concepts that relate to the idea of situational attribution; groupthink, mob mentality, herd behavior, and emulation (2008, p. 25). I'd like to add the concept of pluralistic ignorance, which contends that people go along with ideas they don't accept personally, but because they think that everyone else accepts it, they themselves don't object. This can lead to a situation in which the majority of a group accept a situation they may personally be opposed to. I don't remember where I encountered this idea, but the Wiki entry can provide some introductory explanation (see comment for use of Wikipedia).

All of these concepts return to a central concept of conformity. Arendt introduces us to the idea of the "banality of evil" (Haslam & Reicher,2012, p. 2). Zimbardo also has introduced "The Lucifer Effect" which expands on his earlier idea. And yet, I don't think that either conformity or "the system" is a primary source of wrong behavior.

I always return to a central point in the way that I look at human behavior: Human behavior, including crime and sadism, is caused by the interaction of several factors, including distinct sub-categories of the old "nature vs nurture" argument. One of these factors is individual reaction to external factors (socialization, etc). In fact, I would say that "individual reaction" is a result of internal factors (self-control, resilience, etc). For example, what makes some youth resilient to criminal behavior in high risk environments? Conformity is just one of several factors that can can affect decision making. My own opinion is that behavior can be affected by situational attribution (an external factor) when the internal factors align with the external factors (such as the example of the one "guard" in the SPE mentioned in the bulk of the cites above who displayed "extreme" sadistic behavior), or when a culmination of external factors overwhelm internal factors...maybe an example of this would be Stockholm Syndrome.

In my own experience. I could see one example in which situational attribution could be used to describe my behavior, and one in which it did not. In the four years I spent in the Marine Corps, I emulated the behavior of those Marines I considered professional, and in four years of undergraduate study, I either ignored the academic environment or challenged those ideas I felt were wrong. The concept of situational attribution applied in one circumstance, and not in the other.

The researchers did not learn anything. They had a preconcieved notion of what they intended to do with the Stanford Prison Experiment. They set the standards for the "guards'" actions to ensure that result, and they did not simulate an actual prison environment.

There was also a major ethical problem with the study in the method the research was conducted to arrive at a preordained result. There may have been an additional issue with the study under the Belmont guidelines (which came later), but that is a matter of debate. The primary reason I personally do not feel that there was a humanitarian violation of ethics in the study in regards to the participants was that informed consent was given by the participants (The Belmont Report, 1979, para. 22). In my own opinion, this standard takes priority over the other Belmont principles as long as we are discussing legal and capable adults.

The use of Wikipedia as an academic source is discouraged. However, as long as one is aware of possible bias, Wiki can be used an an overview of a subject for someone that hasn't encountered that subject before. Wiki can also be used to gather citations, again keeping in mind that the citations may be cherry picked to support a particular point of view. Finally, I have found the Talk/Discussion tabs of Wiki useful (often, more useful than the subject page) in that as users of the site engage in "edit wars" over a page's contents, they often justify their edits with supporting documentation.

Sources

• Banuazizi, A., & Movahedi, S. (1975). Interpersonal dynamics in a simulated prison: A methodological analysis. American Psychologist, 30(2), 152–160. http://doi.org/http://dx.doi.org.southuniversity.libproxy.edmc.edu/10.1037/h0076835

• DeJong, W. (1975). Another look at Banuazizi and Movahedi’s analysis of the Stanford Prison Experiment. American Psychologist, 30(10), 1013–1015. http://doi.org/http://dx.doi.org.southuniversity.libproxy.edmc.edu/10.1037/0003-066X.30.10.1013

• Drury, S., Hutchens, S. A., Shuttlesworth, D. E., & White, C. L. (2012). Philip G. Zimbardo on his career and the Stanford Prison Experiment’s 40th anniversary. History of Psychology, 15(2), 161–170. http://doi.org/http://dx.doi.org.southuniversity.libproxy.edmc.edu/10.1037/a0025884

• Gross, B. (2008). Prison violence: Does brutality come with the badge? Forensic Examiner, 17(4), 21–27.

• Haney, C., Banks, C., and Zimbardo, P. (1973). Interpersonal dynamics in a simulated prison. International Journal of Criminology and Penology, 1. 69-97. Retrieved August 12, 2015 from http://www.ffst.unist.hr/_download/repository/Stanford_Prison_Experiment.pdf

• Haney, C., & Zimbardo, P. (1998). The past and future of U.S. prison policy: Twenty-five years after the Stanford Prison Experiment. American Psychologist, 53(7), 709–727. http://doi.org/http://dx.doi.org.southuniversity.libproxy.edmc.edu/10.1037/0003-066X.53.7.709

• Haslam, S. A., & Reicher, S. D. (2012). Contesting the “nature” of conformity: What Milgram and Zimbardo’s studies really show. PLoS Biology, 10(11), e1001426. http://doi.org/http://dx.doi.org.southuniversity.libproxy.edmc.edu/10.1371/journal.pbio.1001426

• National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.(1979). The Belmont report. Retrieved August 12, 2012 from http://www.hhs.gov/ohrp/humansubjects/guidance/belmont.html

• Gross, B. (2008). Prison violence: Does brutality come with the badge? Forensic Examiner, 17(4), 21–27.

• Pluralistic ignorance. (n.d.) in Wikipedia. Retrieved August 13, 2015 from https://en.wikipedia.org/wiki/Pluralistic_ignorance