The Stanford Prison Experiment

Thistlewaite and Wooldredge, Part 3 Chapter 5,  The Stanford Prison Experiment: What explains the coercive behavior of prison guards? pp. 362-372.

After reading the article answer the following questions in a discussion posting:

  1. Do you think the methodology used to complete this study was valid?
  2. What limitations do you see, if any, in the way the study was conducted?
  3. Given the increased emphasis placed on diversifying and professionalizing the correctional workforce and  do you feel the findings of the study are still valid today? Why or Why not?



the stanfoRd pRison expeRiment: What explains the CoeRCive behavioR of pRison guaRds?

Haney, C., C. Banks, and P. Zimbardo (1973). “Interpersonal Dynamics in a Simulated Prison.” International Journal of Criminology and Penology 1:69–97.

background According to the “dispositional hypothesis,” prison environments reflect the characteristics, behaviors, and attitudes of both the inmates and the guards (Haney et al. 1973). Guards lack empathy and are cruel to inmates because of their aggressive nature. Inmates, by virtue of their confinement, are criminals incapable of conforming to the rules and laws of society. Their destructive and reckless tendencies can only be controlled through physical and/or psychologi-cal coercion. This perspective is analogous to the importation theory used to explain inmates’ attitudes and behaviors as well as the idea that police work attracts certain types of individuals prone to abuse. A team of psychologists from Stanford University led by Philip Zimbardo devised a research experiment to test the dispositional hypothesis as an explanation of both guard and inmate behavior. The US Navy and Marine Corps funded the study because they were interested in explaining the conflict between guards and prisoners in their own military prisons. The obvi-ous approach would have been to gain access to a prison and observe the interactions and behav-iors of guards and inmates (ethnographic research), but the drawback to this approach is that inmates and guards are already part of the prison environment (Haney et al. 1973). It would have been difficult to separate out the effects of individual characteristics (of guards and inmates) from the effects of the prison environment on guards’ behaviors toward inmates. Instead, the research-ers came up with a clever alternative—construct a “mock” prison whose inhabitants would be selected from the general population. Researchers would randomly assign normal functioning adults to play the roles of “prisoners” and “guards,” and would then observe the interactions between the two groups. The “prison experiment” became one of the most controversial and criticized studies in the behavioral and social sciences.

the experiment

In the summer of 1971, Stanford University researchers Craig Haney, Curtis Banks, and Philip Zimbardo (1973) conducted an experiment to better understand the conflict between prison guards and inmates.8 A simulated prison was assembled in a 35-foot section of the basement of the psychology building at Stanford University. The goal was to create a “functional representa-tion” of a prison environment so on behavior. Ex-prisoners were consulted on the design. Laboratory rooms were turned into six-by-nine foot cells with black steel bars as doors. Each cell contained a cot, mattress, sheet, and pillow. A small room was set aside as the prison yard, and a tiny closet was designated to be the solitary confinement cell (which measured two-by-two-by-seven feet and had no lighting). Guard quarters were constructed in another section of the basement that consisted of a break room, a room for the prison warden and superintendent, and an interviewing area. The research subjects were recruited from a newspaper advertisement requesting male

volunteers for a psychological study of prison life. Seventy-five college students responded. The selection process consisted of questionnaires to assess family upbringing, physical and mental health, and prior involvement in crime. The researchers also interviewed each of the respondents. The 24 subjects selected were deemed to be mature, well-adjusted college students. Subjects were predominately white middle-class students with the exception of one Asian student. Subjects were randomly assigned to play either the role of “prisoner” or “guard” (this was literally done by flipping a coin). Random assignment would allow researchers to determine the influence of the prison environment separate from the characteristics of the individuals playing each role. Efforts were made to ensure that the participants did not know one another and each subject underwent a series of psychological tests prior to the start of the experiment. Four subjects were assigned as alternates, leaving nine guards and nine inmates. Zimbardo himself assumed the role of superin-tendent and one of his research assistants played the part of the prison warden. All of the details of the experiment were not disclosed to the subjects. Participants were

informed that they would be assigned to play either the role of guard or prisoner for two weeks in exchange for monetary compensation. Each subject signed a contract indicating their con-sent to participate and their acknowledgement of certain conditions of the experiment. Inmates were told that they would be provided minimal food, clothing, housing, and privacy. They were also informed that they might be subjected to verbal abuse. No other information was provided. Researchers wanted the subjects to enter into the experiment with only their own preconceived ideas about how inmates behaved in prison. Researchers went so far as to create a realistic prison induction process for the inmates. With

assistance from the Palo Alto City Police Department, participants assigned to the prisoner group were all arrested without warning at their place of residence. The public nature of the arrest was meant to humiliate them. Suspects were arrested for either burglary or armed robbery, they were read their Miranda rights, searched, handcuffed, and taken to the police station for booking. At the station, subjects were photographed, fingerprinted, and placed in a detention cell. Next, they were blindfolded and driven to the simulated prison. The prisoners were unaware of the location of the prison so they would think they were being taken to a “real” prison facility. After the inmates arrived, they were stripped of their clothing and sprayed for lice. Each prisoner was issued a uni-form that consisted of a simple “smock” with an inmate number on the front. Inmates were not permitted to wear anything under the smock and had to wear stocking caps on their heads (to sym-bolize having their heads shaved just like actual inmates). The inmates were issued rubber sandals as shoes, and each had to wear a small lock and chain around one ankle. The uniforms were not typical of what real inmates would wear, but the purpose of the uniform was the same: to strip the prison-ers of their old identities and remind them of their confined status. All inmates dressed the same in order to solidify their status as “inmates.” The entire induction process was designed to degrade and demoralize the inmates. Inmates were referred to only by their inmate numbers rather than by their names. The prisoners were to remain in their cells 24 hours a day for a two-week period. Subjects designated as prison guards were also issued standard uniforms to help create a common identity among them and to further remind the prisoners of their subordinate status.that researchers could observe the influence of this environment

The guards’ uniforms were very similar to what real prison guards might wear at the time of the study. Guards wore khaki shirts, pants, and mirrored sunglasses so that inmates could not make eye contact with them. They were given a whistle and a police nightstick as symbols of author-ity. The guards were assigned to eight-hour shifts with three guards working per shift. While

“off duty,” the guards were told to resume all normal activities. Guards attended an orientation meeting prior to the start of the experiment. They were provided with work assignments, pris-oner schedules, and procedures for completing reports. Researchers explained that the purpose of the experiment was to create a realistic prison environment. The guards were instructed to

“maintain the reasonable degree of order within the prison necessary for its effective functioning” (p. 7). No other instructions were provided. Researchers were purposely vague to make sure the guards brought with them only their own ideas about how to maintain control over a group of subordinates. After the subjects were placed in their “cells,” the warden welcomed the inmates and read

them the institutional rules that were written by the warden with input from the guards. Inmates had to memorize the rules and recite them back to the guards during the scheduled inmate counts. The daily routine of the inmates consisted of the following:

three meals (consisting of a bland diet) three toilet visits (all supervised) two hours of free time (to read or write letters) work assignments daily exercise two visiting periods per week

The routine became the inmate’s only indication of time, as there were no clocks or windows in the prison.


All of the interactions between guards and inmates were continuously observed. Both the war-den and superintendent were present at various times to document what took place and all of the events were recorded. Intercoms secretly recorded conversations between the inmates as well as between the guards. During the first day of the experiment, both groups appeared to have had some difficulty with their respective roles. The scheduled counts created an opportu-nity for the guards to assert their authority, and the inmates complied. When the inmates did not follow the rules, the guards required them to do “pushups” as punishment. Remember, the guards were not given any instructions on how to punish and so the pushups were their own idea. Punishments were also administered for disrespecting the guards because the guards felt they were entitled to respect by nature of their position. What happened on the second day completely surprised researchers. The day began with a revolt from the inmates. They removed their caps and inmate numbers and refused to leave their cells. Next, they barricaded their cells with their cots and verbally abused the guards. When the day shift arrived, the day shift guards became angry at the nightshift and blamed them for the rebellion. The guards immediately attributed the inmates’ behavior to lenient treatment by the guards during the previous night. Reinforcements were called in (the alternate guards) and the night shift guards stayed on duty to assist in quiet-ing the disturbance. The guards sprayed inmates with a fire extinguisher in order to gain access to their cells. Next, they proceeded to strip the inmates of their smocks and removed their cots. The alleged “leaders” of the revolt were taken to solitary confinement while the other inmates were subjected to verbal abuse and physical intimidation by the guards. The guards successfully ended the disturbance. Realizing that all nine guards could not remain on duty during a single shift, one of the

guards came up with a plan to use “psychological tactics” to keep the inmates in line. A “privilege” cell was set up for the three inmates believed to be the least responsible for the disturbance. These inmates were given back their clothes and beds, were allowed to wash and brush their teeth, and were given a special meal that was eaten in the presence of the other inmates. The guards were deliberately trying to break the inmates’ solidarity. The next day, the “good” inmates were returned to their cells and the “bad” inmates were taken to the privilege cell. This was all done to confuse the inmates and to trick the alleged leaders into thinking that their fellow inmates had turned on them and were now cooperating with the guards. Guards wanted to turn the inmates against each other in order to divert their aggression away from them (a plan that was devel-oped on their own). These actions also appeared to be increasing solidarity among the guards and their roles became more well-defined. The guards started to perceive the inmates as threats and, in response, began to intensify their efforts to control them. Guards no longer followed the daily routine. Bathroom visits became privileges dispensed at the whim of the guards. The guards went so far as to place buckets in the inmates’ cells to be used as toilets, and then they refused to empty them. One inmate had his smoking privileges taken away and others were denied access to their mail. Thirty-six hours into the experiment, one inmate started to exhibit intense emotional

symptoms such as uncontrollable crying and bouts of aggression. The researchers believed he was faking so he would be released early. Even some of the researchers assumed the roles of prison administrators by treating the inmate as a “con.” The prison consultant was brought in to inter-view the inmate. He called the inmate “weak” and explained that his treatment was mild com-pared to what real inmates received. The consultant then bargained with the inmate by telling him that if he continued with the experiment and served as an informant, the guards would leave him alone. He was sent back to his cell and told to think it over. At the next count, the inmate started screaming at the other inmates about how they could not leave or quit the experiment. His erratic behavior intensified and the guards were unable to bring him under control. The prison administrators finally realized his symptoms were real and released him. Day four was visiting day. The researchers were concerned that family members might find

the prison environment alarming and would be shocked at the appearance of the inmates, so they decided to manipulate the situation. The inmates were washed, allowed to shave, and fed a large meal. A nice-looking former Stanford cheerleader was brought in to play the role of greeter, and music was pumped in over the intercom during the visits. Despite the fact that the visitors were made to wait an hour to spend ten minutes with their family members, most cooperated. Some of the parents expressed concerns over the physical appearance of their sons, but the warden and superintendent alleviated their worries. One father was asked if he felt his son could handle the experiment and he indicated that he could. Most of the parents left with the impression that the experiment was all in fun and that their loved ones were well taken care of. After visiting hours were over, the guards picked up on a rumor that the inmate released

from the experiment on the third day was planning on returning to the prison with friends to help the other inmates escape. The guards called a meeting with the warden and superinten-dent to discuss strategies for preventing the escape. One idea was to bring in an informant to pose as an inmate to try and get information on the plan. The researchers also went so far as to try and have the inmates moved back to the Palo Alto jail, but the sheriff refused because of liability concerns. The researchers ended up dismantling the jail and moving the inmates to another floor of the building. If anyone from the outside showed up, they would be told the experiment ended and everyone had gone home. The only person who showed up was a col-league of one of the researchers who was curious about the experiment. The escape rumor turned out to be false. The jail was reconstructed and the inmates returned to their cells. The guards became furious after spending so much time dismantling and re-constructing the mock prison. They took out their frustration on the inmates who had made them look foolish. The guards punished the inmates by having them do strenuous exercises and more chores. Each count lasted over an hour and the inmates were subjected to continuous verbal abuse. Researchers decided to bring in a Catholic priest who had served as a prison chaplain in

order to provide feedback on the reality of the prison environment. The inmates were allowed to speak with the priest individually. The priest was quite surprised when the inmates introduced themselves by their number and not their actual name. He then proceeded to ask the inmates what they were doing to get out of prison. The inmates appeared confused by the question so the priest told them they needed a lawyer. He even offered to call their parents to tell them to get legal assistance. Some of inmates consented. One particular inmate refused to see the priest. He complained of feeling sick and requested to see a doctor. When he came out of his cell, he started crying uncontrollably. He was taken to a room to rest and was given some food. The guards became angry with this and made the other inmates start chanting that inmate #819 was a “bad inmate” (knowing that inmate #819 could hear this). The inmate became more upset and was asked if he wanted to go home. He responded that he couldn’t because the other inmates would think he was bad. He asked to go back to his cell. Zimbardo told the inmate that he was not bad, that this was just an experiment, not a real prison, and the inmate agreed to leave. The next day researchers decided to schedule a parole hearing. Some of the inmates were

brought before a parole board that consisted of college staff and graduate students. The prison consultant was also present. The inmates were asked if they would forfeit their pay in exchange for parole, and most said yes. When they were told to go back to their cells, however, they all complied. The last rebellious act on the part of the inmates occurred right before the experiment ended. The new inmate, brought in as an alternate, went on a hunger strike. The prison experi-ence was different for this inmate. He came in and immediately became immersed into the hostile environment created by the guards. The other inmates experienced a more gradual introduction to the environment. The others also told the new inmate that the prison was real because they were not free to leave. When he refused to eat, the guards placed him in solitary confinement and held him three times longer than the rules allowed. Instead of sympathizing with the inmate, the others labeled him a troublemaker. The guards took advantage of this and gave the other inmates a choice: give up their blankets and the inmate would be released from solitary confinement. The inmates chose to keep their blankets. On the fifth night, some visiting parents asked Zimbardo to contact a lawyer (just as the

priest had suggested) to help get their sons out of prison. An attorney showed up the next day and met with each of the prisoners. It was at this point that Zimbardo and the other researchers decided to end the experiment—less than one week after it began. The researchers felt that the simulated prison environment had become too real for both the guards and the prisoners. The inmates continued to become passive and withdrawn and showed signs of depression. Some of the guards became overzealous in their actions to control the inmates while the other (more humane) guards felt helpless to intervene. One of the interviewers hired to question the guards and inmates expressed strong objections to how the inmates were being treated and was shocked that the experiment had been allowed to continue as long as it did. The prisoners were elated to be released early but the guards expressed discontent that the study was over. During the entire experiment, no guard ever showed up late for work, called in sick, or requested overtime pay for the extra time on duty. The guards genuinely appeared to like their position of authority over the inmates.


Based on their observations, Zimbardo and his research team argued that they had successfully refuted the dispositional hypothesis. The subjects recruited to play the roles of guards and prison-ers were normal, middle-class adults put into an environment where the guards were given almost total authority over the prisoners. With little direction on how to exert and maintain that author-ity, the guards increasingly demonstrated negative and hostile behaviors. They quickly immersed themselves into their positions and relied on both physical and psychological techniques to con-trol the inmates. Three types of guards were observed. The “tough but fair” guard adhered to the prison rules consistently for all inmates. The “good guys” gave the inmates special privileges and never punished them. A third group of the guards were antagonistic and harsh toward the inmates. Rules were enforced arbitrarily and they seemed to enjoy coming up with new ways to exert power over the inmates. None of the psychological testing or personal interviews provided any insight into why the guards behaved differently. There were also observed differences in the coping styles of the prisoners. Initially, the

inmates all complied with the orders and followed the rules. This quickly changed when several inmates became rebellious and antagonistic toward the guards. The inmates eventually became more passive as the experiment progressed. Four prisoners experienced severe emotional dis-tress, and one even developed a psychosomatic rash. A few inmates actually tried to endure the conditions as best they could by cooperating at all times with the guards. The solidarity between inmates existing at the beginning of the study diminished quickly, as evidenced by their loss of personal identities (recall how each of them introduced themselves to the priest by their inmate number). All of the inmates expressed relief when the experiment was terminated early. The only personality trait uncovered in the psychological testing that appeared to be linked to differences in coping styles was that prisoners who tested high on a measure of authoritarianism seemed to adapt better to the environment. The researchers believed that their observations of the simulated prison environment

offered support for the contention that the prison environment, not the individuals who inhabit it, influences the behaviors of actual inmates and guards. Prisons were designed to be dehuman-izing, degrading, and to instill a sense of despair among the inmates being held against their will. Guards were put in a position of having to exercise control and maintain authority over the prisoners under conditions in which they were outnumbered and disrespected by the pris-oners. Inmates and guards became immersed in roles that were shaped by their environment. Some inmates adapted by lashing out against the guards through violence and verbal insults while others shut down emotionally and became depressed and submissive. Knowing that it was the responsibility of the guards to maintain control over the inmates, some guards exerted their authority in an aggressive and hostile manner. According to the researchers, this explained the volatile interactions that frequently occurred between guards and inmates in real prisons.



The researchers put forth great effort in creating their simulated prison, but an obvious ques-tion raised after the study was published was whether or not the environment was authentic.  The ecological validity of Zimbardo’s study depended on the extent to which his participants per-ceived the prison environment to be real. No matter how much Zimbardo and his team tried to simulate an authentic prison environment, the fact remained that the prison was not real. Certain conditions of prison could not be imitated. At no time were the inmates or guards in any physical danger, and the experiment was designed to last only two weeks. It was possible that the prisoners and guards had just put on a good show for the researchers by behaving in stereotypical ways because this was what was expected of them. Both groups knew their behaviors were being recorded. In response to the issue of ecological validity, researchers pointed out that the guards believed the experiment focused solely on the inmates. They did not know their conversations were recorded in the breakroom. Instead of using this time to get to know their fellow employ-ees, most of their conversations focused on work. They spoke about the inmates and discussed prison-related details. Interviews conducted after the experiment ended also revealed that there were interactions between the guards and prisoners that were not observed, such as trips to the bathroom. These unobserved interactions were reportedly more negative than those observed. The harassment continued and was often more intense. Also, the hostility of some of the guards intensified as the experiment progressed, suggesting that these guards might have become consumed in their roles. According to Zimbardo, the guards were not given any specific instruc-tions on how to exert control, yet they relied on physical and psychological coercion. Certain aspects of the prisoners’ daily routine such as watching television, reading, and even going to the bathroom became privileges that were granted or denied at the whim of the guards. The guards knew they had authority over the lives of the prisoners and they had no difficulty exhibiting that control. Several guards even reported that they found the control “exhilarating.” According to Erich Fromm (1973), however, Zimbardo’s experiment failed to adequately distinguish between

“behavior” and “character.” The guards were assigned a role, and while some behaved negatively toward the prisoners based on their beliefs regarding how guards should act, it was misleading to suggest that the guards enjoyed treating the prisoners poorly. Conversations between prisoners were also recorded without their knowledge. The pris-oners spent most of their time talking about their prison experiences. They complained to each other about the food, the guards, and the punishments. This was a group of students who did not know one another outside of the experiment and yet they had very little interest in learning about each other. The inmates introduced themselves by their numbers to the priest, and three inmates were willing to forfeit their pay to go home. The prisoners became passive and emotionally dis-traught as the experiment progressed. Despite the fact that the prisoners knew their incarceration was not real, researchers believed that the experience affected them. The loss of personal identity in conjunction with the arbitrary control and emasculation appeared to have had an impact on their behaviors. The helpless behaviors exhibited by the prisoners may have been manifestations of the initiation technique used in the study. Subjects were unknowingly arrested by real police officers and taken to a police station for booking. It would have been natural for them to be con-fused as to whether or not they were free to end the experiment at any time (Fromm 1973). There continues to be doubts as to whether or not the inmates and guards acted on their own accord. Carlo Prescott, the ex-inmate who served as a consultant for the experiment, wrote an article for the Stanford Daily in 2005 in which he reported that researchers had prompted some of the participants’ actions ( the Lie Of The Stanford Prison Experiment). Zimbardo’s sample of research subjects could also be criticized. His subjects were selected

from of group of volunteers who had answered an ad in the newspaper. Each underwent psycho-logical testing and an interview, yet 24 white male college students were selected to participate in the experiment. Study participants shared very few characteristics with real prison guards and inmates. It is questionable whether the results would have generalized to a larger, more diverse population. Most of the criticisms focused on alleged ethical violations in Zimbardo’s research. All of

the details of the experiment were not disclosed to the participants. The fake arrests that took place at the beginning of the study were a complete surprise. This was intentional because the researchers wanted the prisoners to go through the same type of induction as real criminals. The vagueness of the instructions left the guards under the impression that the researchers were only interested in studying the behaviors of the prisoners, and researchers made no attempt to correct this because they wanted the guards to feel comfortable playing their roles. Full disclosure is not always ideal when conducting research but it becomes an important consideration when making sure that subjects willingly provide their informed consent to participate. Zimbardo himself admit-ted he became so caught up in his role of prison superintendent that his objectivity may have been compromised. When confronted with the rumor of the impending prison break, he helped devise a strategy to stop it. Despite the fact that Zimbardo was assigned a role to play in the experiment, as a researcher he should have allowed the guards to come up with their own plan. The most significant ethical issue surrounding the Stanford Prison experiment was the

harm inflicted upon the subjects. Researchers wanted to explore the influence of a prison envi-ronment on the behaviors of individuals assigned to play the roles of prisoners and guards, but they never anticipated the extent to which both groups would become absorbed in those roles. The prisoners suffered harm and became so emotionally distraught that the experiment had to be terminated after only six days. The guards became immersed in their roles and demonstrated a willingness to use physical and psychological coercion to control the prisoners. Their position of authority carried with it a sense of power that they freely exercised. No one questioned their tactics or intervened to stop the abuse. While the experiment was taking place, over fifty curious individuals visited the prison and observed what was going on. No one voiced any concerns or disapproval until the sixth day when a colleague of Zimbardo showed up and expressed shock and condemnation at what was taking place. Before the subjects left the experiment, everyone participated in discussion sessions to allow the guards and prisoners to openly express their thoughts and feelings. The researchers also participated by questioning the moral and ethical concerns of some of their actions. It is important to point out that the researchers followed all guidelines specified by the Stanford Human Ethics Committee which granted approval of the study. Predicting harm can be difficult, but the experiment should have been terminated once the researchers started to observe the prisoners’ distress and the guards’ willingness to inflict suffering.

significance and subsequent Research

The Stanford Prison experiment became widely cited primarily because of the alleged ethical vio-lations. Numerous social science research texts cite the study as an example of unethical research, and the experiment has never been replicated in the United States for this reason. In 2002, the British Broadcasting Corporation produced a reality television show based on the original prison experiment. Fifteen participants were assigned to play the roles of prisoners and guards in a sim-ulated prison constructed in a television studio. Two psychologists, Alex Haslam and Stephen Reicher, watched over the project, and an independent ethics committee was on hand to supervise as well. In this experiment, the prisoners formed a unified front against the guards. Prisoners mocked and ridiculed the guards and refused to comply with their orders (Reicher and Haslam 2006). This study met a similar fate as the original in that it was terminated early after participants started to exhibit severe emotional symptoms (although it was the guards exhibiting these symptoms and not the inmates). The Stanford Prison experiment gained further notoriety in 2004 when the news media

reported on the torture of Iraqi inmates held by US soldiers at Baghdad’s Abu Ghraib prison. Images of inmates being abused and humiliated left many wondering how American soldiers could be responsible for such behavior. The military placed the blame on a few dishonorable guards, but Zimbardo provided another explanation. In fact, he testified as an expert witness for one of the soldiers who was court-martialed. Zimbardo believed the abuse stemmed from some of the same “situational forces” that existed in his simulated prison (Zimbardo 2004). The total power exercised by guards over the inmates, a lack of oversight and accountability, and an envi-ronment where inmates were viewed as less than human helped to explain the guards’ actions. The Stanford Prison experiment provided valuable insight into the dynamic roles of

inmates and guards and how the prison environment can shape these roles. During the experi-ment, the guards relied primarily on coercive tactics to force inmates to comply with their orders even though they had not been instructed to do so. Research that followed explored differences in how correctional officers exercise their control over inmates. John Hepburn (1985) surveyed cor-rectional officers in an effort to determine the extent to which officers rely on particular “bases of power” in order to gain inmate compliance. Three hundred and sixty correctional officers completed surveys designed to measure five different power bases: legitimate, coercive, reward, expert, and referent. “Legitimate” power is derived by nature of the correctional officer’s position of legal authority. Officers are also in a position to exercise “coercive” power because they can punish inmates who refuse to comply with their orders. Alternately, they can distribute “rewards” to inmates for their compliance by recommending particular housing or work assignments or by ignoring minor rule violations. “Expert” power is based on an inmate’s perception that officers possess some special knowledge that make them experts in their jobs (such as being able to solve problems that inmates are unable to solve). Finally, inmates who comply with officers because they respect them are subject to “referent” power. Hepburn found that most officers believed inmates complied primarily because of legitimate and expert power, despite the fact that inmates typically did not acknowledge officers’ rightful authority over them and usually did not perceive officers as skilled professionals. Coercive and reward power were ranked last among officers as their primary means to gaining an inmate’s cooperation. According to Hepburn, this finding was not surprising because inmates usually did not perceive the punishments handed out by guards as being separate from their general incarceration experience, not to mention that the availability of rewards issued by officers was limited. Zimbardo and his research team identified three types of guards from their observations

of the guards’ interactions with inmates. Other researchers have created similar typologies. Kelsey Kauffman (1988) identified five types of correctional officers based upon his interviews with 60 correctional officers from different prisons. This typology reflected the officers’ attitudes toward each other and the inmates. The “pollyanna” officers displayed positive attitudes toward other officers and the inmates. The “white hats” held favorable attitudes toward inmates but not for their fellow officers. “Hard asses,” on the other hand, were unsympathetic toward inmates but held other officers in high regard. “Burnouts” displayed negative attitudes toward both inmates and other officers. The last type included the “functionaries,” described as “indifferent” toward the inmates and officers. According to Kauffman, officers entered their careers as either pollyan-nas, white hats, or hard asses, and then some of these officers became burnouts or functionaries later in their career. This suggested that the occupational experiences of the officers shaped their attitudes.The “guard” participants in Zimbardo’s prison experiment were all white, college educated males. Correctional officers in the early 1970s were typically white, uneducated, con-servative males who lived in the rural areas where most prisons were located (Philliber 1987). To more fully understand the influence of correctional officer characteristics and attitudes toward inmates, James Jacobs and Lawrence Kraft (1978) conducted a study to determine if there was a relationship between correctional officers’ race and their attitudes toward inmates. They admin-istered a survey to a sample of 252 in-service guards from two prisons in Illinois: Stateville and Joliet. Both prisons were maximum-security facilities located within seven miles of each other. Three-fourths of the inmate populations in both prisons were Black, and most of the prison-ers came from the Chicago area (the prisons were situated 35 miles southwest of the city). At the time the survey was administered, 12 percent of the correctional officers employed by the Illinois Department of Corrections were Black and most of them were working at Stateville and Joliet. The survey was administered to guards while they attended the Correctional Academy between the summer of 1974 and fall of 1975. One hundred and sixty-five white guards and sixty-six Black guards completed the survey. Researchers were also interested in whether or not there were any racial differences in attitudes based on age and years of experience. Overall, Black prison guards did not express attitudes that were more sympathetic toward inmates com-pared to white guards, although responses to certain survey questions did indicate less sympathy by Black guards compared to whites. No racial differences were found in response to a question designed to measure social distance: “How similar are correctional officers and inmates?” Most of the guards, including the younger Black guards, indicated that inmates “try to take advantage of officers whenever they can.” Black guards were less likely to agree that most inmates were

“decent people,” although these guards also indicated that a smaller number of inmates actu-ally “belong in prison” compared to the preferences of white guards. There was no difference in responses to the item “most inmates lack morals.” Researchers expected the Black guards with less experience to display the most empathic attitudes toward inmates, but this was not revealed in their responses. Many of them indicated that “inmates take advantage of officers whenever they can,” although less experienced Black guards also felt that fewer inmates should be in prison and that inmates had not been given too many constitutional rights. Less experienced white guards displayed the most negative attitudes toward inmates. Black guards (and particularly less experienced Black guards) were more likely than white guards to agree with the statement,

“Black guards get along better with inmates than whites do.” Most of the Black and white offi-cers overall disagreed with the statement, “Correctional officers should be rough with inmates occasionally to let them know who is boss,” but more of the less experienced Blacks agreed with this statement. Guards were also asked about the number of disciplinary reports they issued each week. Black guards indicated they issued more reports compared to white guards. While the responses to most questions indicated that Black guards did not have more favorable atti-tudes toward inmates, Blacks expressed a preference for job assignments that involved greater contact with the inmates. White guards, on the other hand, more often indicated preferences for assignments with less contact with inmates. Jacobs and Kraft (1978) speculated that Black guards might be more comfortable interacting with inmates compared to white guards. Nancy Jurik published a fairly comprehensive study of correctional officers’ attitudes in 1985. Jurik sur-veyed 179 correctional officers from a prison located in the western United States. The prison housed both male and female inmates in medium-and minimum-security units. Jurik exam-ined both organizational and individual officer characteristics to determine the influence of each on correctional officers’ attitudes toward inmates. Contrary to Jacobs and Kraft, she found that minority correctional officers displayed more positive attitudes toward inmates compared to white officers. The difference in findings could have been explained by regional differences between the samples of guards (Jacobs and Kraft surveyed guards in the Midwest), differences in security level (Jacobs and Kraft surveyed guards from maximum-security prisons only), and/ or the inclusion of other minority groups. Jurik also uncovered differences in attitudes based on age, but opposite to what she expected in that older guards responded with more positive attitudes toward their work. Male and female guards displayed similar attitudes, and there were no significant differences in attitudes based on level of education. Two organizational measures were found to maintain significant relationships with correctional officers’ attitudes. More years of experience corresponded to more negative attitudes toward inmates whereas officers work-ing in minimum-security prisons held more favorable attitudes. This finding was inconsistent with research by Carol Smith and John Hepburn (1979) who found the opposite relationship between security level and attitudes.Survey research has uncovered some support for the dispo-sitional hypothesis (i.e., the prison environment reflects the characteristics and attitudes guards bring with them to the job), but the influence of prison environments in shaping these attitudes cannot be ignored. Craig Haney (2008), one of the researchers who assisted Zimbardo with his experiment, believes this to be particularly true with the modern supermax prisons. Inmates are kept in an extremely deprived environment with little activity and social interaction. This envi-ronment is not only detrimental to the inmates, it also increases the potential for correctional officers to abuse their power. Officers perceive supermax inmates to be immune from the pains of imprisonment, and they consider confinement in these prisons (built to house the “worst of the worst”) as justification for using any amount of force against inmates. The Stanford Prison experiment remains one of the most controversial studies ever con-ducted in behavioral research. The study even attracted the attention of film makers. A German film titled Das Experiment (released in 2001) was based on Zimbardo’s experiment, and an American version appeared in theaters in 2010. The notoriety stems primarily from the ethical violations of the researchers, but findings from the experiment still contributed substantially to our understanding of the conflict between correctional officers and inmates.


Fromm, E. (1973). The Anatomy of Human Destructiveness. New York: Holt, Rinehart, and Winston.

Duffee, D. (1974). “The Correction Officer Subculture and Organizational Change.” Journal of Research in Crime and Delinquency 11:155–72.

Haney, C. (2008). “A Culture of Harm: Taming the Dynamics of Cruelty in Supermax Prisons.” Criminal Justice and Behavior 35:956–84.

Hepburn, J. (1985). “The Exercise of Power in Coercive Organizations: A Study of Prison Guards.” Criminology 23:145–64.

Jacobs, J., and H. Retsky (1975). “Prison Guard.” Urban Life 4:5–29.

Jacobs, J., and L. Kraft (1978). “Integrating the Keepers: A Comparison of Black and White Prison Guards in Illinois.” Social Problems 25:304–18.

Jurik, N. (1985). “Individual and Organizational

Determinants of Correctional Officer Attitudes Toward Inmates.” Criminology 23:523–39.

Kauffman, K. (1988). Prison Officers and Their World. Cambridge, MA: Harvard UP.

Klofas, J. (1984). “Reconsidering Prison Personnel: New Views of the Correctional Officer Subculture.” International Journal of Offender Therapy and Comparative Criminology 28:169–75.