On rethinking the psychology of tyranny: The BBC prison study

Philip G. Zimbardo[1]

Stanford University, USA

This commentary offers a critical evaluation of the scientific legitimacy of research generated by television programming interests. It challenges the validity of claims advanced by these researchers regarding the Stanford Prison Experiment (SPE) and highlights the biases, fallacies and distortions in this study conducted for BBC-TV that attempted a partial replication of my earlier experiment.

I have reluctantly accepted the editor’s invitation to prepare this commentary on my evaluation of Rethinking the psychology of tyranny: The BBC Prison Study, following my reviewer’s judgment of it as not acceptable for publication in any scientific journal. It is rarely productive to engage in such public debate without undermining the integrity of our discipline in the process. However, across their many publications, these authors have insisted on using the Stanford Prison Experiment (SPE) as their ‘straw situation’ to give visibility and conceptual legitimacy to their scientifically irresponsible ‘made-for-TV-study’. Thereby, they have forced me to counter their allegations in the BJSP, less to defend my research than to publicly highlight the inherent inadequacies, exaggerated claims and outright falsehoods being perpetrated by their insistence on making much ado about what should be nothing of scientific merit.

After briefly describing my privileged position in this matter, giving a synopsis of this study from my view point, mentioning some fundamental differences between the SPE and the BBC study, I will outline a short selection of reasons why I believe this alleged ‘social psychology field study’ is fraudulent and does not merit acceptance by the social psychological community in Britain, the United States or anywhere except in media psychology. A detailed, expansive chronology of the SPE will finally be published in my book, The Lucifer Effect: Why Good People Turn Evil (Random House/Ebury Press. London, 2006)

48 Philip G. Zimbardo My privileged vantage point

I was originally contacted by the BBC staff to be their primary consultant on this TV show in their attempted replication of the simulated prison experiment that I had conducted with Craig Haney and Curt Banks in 1971. I refused because I felt the reason for their wanting to make such a show was based on the dramatic demonstrations of prisoner abuse by the guards in our research. It might make interesting television, but is unethical to do so again either in academic research or in such media re-creations. When the producers indicated they were determined to proceed with this replication, I recommended that my British colleague, Mark McDermott, be hired to help organize an ethics committee with power to stop the study if similar abuses occurred; such an oversight group was formed and was active. I also was allowed to view their final 4-hour video production (and have made a detailed analysis of the events and reactions portrayed there). In addition, I have been in contact with several of the BBC study prisoners and one guard, have reviewed a web site that some participants created, and also spent a day in personal discussion with Phil ‘Bimmo’ Bimpson, the leader of the prisoner rebellion, about what really happened from his insider’s perspective.

Synopsis of the BBC attempted replication

Fifteen adult men (aged 22 to 44) from the United Kingdom were recruited by national advertisements for ‘The Experiment’ with headlines, ‘How well do you really know yourself?’ Those selected from 500 who answered the ad would take part in a ‘university-backed social science experiment to be shown on TV’ that promised ‘to change the way you think’.1 The volunteers were warned that there would be exposure to ‘exercise, tasks, hardship, hunger, solitude and anger’. Of the 15 men, 9 were randomly assigned (allegedly) to be prisoners, with 6 playing guards. They were given background questionnaires, and BBC staff visually evaluated some. A high-tech, goodlooking prison set was constructed at the George Lucas sound stage of BBC’s Elstree studios (Hertfordshire) to allow monitoring by the researchers, ethics committee and of course for optimal audio and video-recording at all times. The study was projected to last 10 days but was terminated a few days earlier. After considerable editing by the BBC producers, some of it forced by critical preview reactions of the former prisoners, the research was broadcast (May, 2002) on BBC television as a series of four 1-hour programmes entitled, ‘The Experiment’ (Koppel & Mirsky, 2002).[2] [3]

Once in the mock prison, prisoners were given orange jump suits, with lapel microphones attached, a prisoner number and their heads were shaved. Three of them occupied each of the attractive well-lit cells provided with books and games. The guards lived in separate quarters in military style uniforms. There were no pre-arranged guard rules governing what the prisoners must and must not do.

Basic differences between SPE and BBC-E

  • Our participants were young college students; theirs were older men from varied backgrounds.

Commentary 49

  • Our study began with simulated arrests by real local police; theirs began by volunteers arriving on their own.
  • Our study had an institutional authority hierarchy as well as planned sets of prisonlike corresponding agencies and events. We had a prison superintendent, warden and psychological counsellors arrayed above the prisoners and guards. Theirs had only the two experimenters prominently dominating the action. We had a parole board, disciplinary board, visits by a prison chaplain and legal public defender, parents and friends visiting nights and continued involvement of city and university police. They had none of that.
  • Our study began with the guards and warden formulating a set of coercive rules governing all aspects of prisoner behaviour. They had none.
  • In our study, video-recording was concealed and never apparent to participants, and none wore microphones to make them aware of being under surveillance. In theirs, it was obvious that everything was being recorded at all times.
  • In our study, the researchers rarely had direct contact with the prisoners or guards in their role as researchers, with minimal intrusion into the prisoner-guard dynamics. In theirs, the researchers continually intervened, made regular broadcasts into the prison facility, administered daily psychological assessments, arranged contests for the best prisoners to compete to become guards’ and as in all reality TV shows, created daily ‘confessionals’ for participants to talk directly to the camera about their feelings.
  • Our study was conceptualized around a psychology of imprisonment modelled after typical prisons in the US where staff and guards exert dominating influence over prisoners. Their study seemed to be based on a model of a commune where power is distributed and decisions negotiated democratically.
  • Our study randomly assigned participants to the two treatments. Their alleged ‘random assignment’ yielded two very different-seeming groups from the very start of their study.
  • In our study, some prisoners had to be released early because they could not take the abuse any more. In the BBC study, some guards quit early because they could not stand their abuse any longer.

Fast forward to the conclusion of the BBC-TV study

The rather remarkable conclusion of this simulated prison experience is that the prisoners dominated the guards! The guards became ‘increasingly paranoid, depressed and stressed and complained most of being bullied!’[4] Several of the guards could not take it any more and quit. The prisoners soon established the upper hand, working as a team to undermine the guards, to challenge and tease them, and even to limit access to their cells and get better food. Then, everyone got together and decided to form a peaceful ‘commune’, but several dissident prisoners disrupted it. The study was terminated a bit earlier than planned because a coup from a subgroup of prisoners was about to take place to create a more authoritarian prison-like atmosphere.

What is the external validity of such events in any real prison anywhere in the known universe? In what kind of prisons are the prisoners in charge? How could such an eventuality become manifest?

50 Philip G. Zimbardo

What went wrong with the BBC experiment?

In the design and execution of the SPEs my colleagues and I were invested in creating a general atmosphere that was a functional simulation of the psychological nature of being imprisoned, which included minimizing the voluntary experimental aspect of the situation to the participants. Over time, our guards and prisoners never referred to their experience as an experiment, it became prison-like in many ways. Not only were our guards always in control of the situation, many continually invented new ways to increasingly dominate the prisoners.

None of that transformative experience was ever evident in what I have seen and heard of the BBC experiment — for obvious reasons. From the start, everyone knew this was a made-for-TV study; that everything they did would be shown on national British television — for better or for worse. Next, the demands of good television production values — getting quality audio and video and well-lit close-ups even when a prisoner is in solitary confinement — undermine or at least interfere with scientific production values of creating ‘mundane realism’ within an admittedly artificial environment. Finally, for inexplicable reasons, the researchers made themselves a dominantly intrusive constant element into the research setting. Instead of being content to be just scientific observers, they became ‘players’, perhaps at the instigation of the producers, to make things happen each day that might be interesting. Their intrusions forced the participants to be constantly aware that what they were experiencing was just that ‘social science experiment to be shown on TV’ that had attracted their initial attention from the televised lure for participants.

I would like to elaborate these points in more detail and raise a few others. This final part of my Commentary is organized around the issues of: random assignment; failure to create the psychologically necessary situational conditions for the treatment variable to be meaningful; the Heisenberg indeterminacy principle; reality-concealing rhetoric, data cautions and the values conflict between TV and scientific research. [5]

Commentary 51

assignment that just happens to have all the big, rough and tough guys with tattoos as prisoners and the more effete guys as guards. I guess sometimes chance favours such an outcome that the BBC producers must have welcomed as did their TV cameramen.

  • Failure to create conditions for a prisoner mentality. The failure of the researchers even to try to sustain a prison-like psychological atmosphere, despite the million pound BBC-constructed prison setting, is not only surprising to me, but undercuts their specious attempts at meaningful comparisons with the SPE. In addition to calling out participants for their daily confessionals to their so-called ‘Diary Room’, the researchers themselves are always a dominating presence, figures and not ground in this jail. They make frequent announcements over the prison intercom, conduct daily psychological assessments and, even worse, they impose mindless ‘interventions’ during the study, clearly for TV values — in opposition to values of psychological science. At the end of the very first day, the research duo does something unimaginable in prisons anywhere, except perhaps one run by The Quakers for the Amish. They announce a competition: a prisoner will be transformed into a guard based on his good behaviour! This tactic was probably generated by BBC staff to stir up the pot that was filled with a very tepid stew at the start of videotaping. Surely it had no legitimate basis in the foundation of research on prison dynamics. This contest interrupts the development of participants becoming socialized into their new roles. Any sense of power differentials and categorical differences between the two groups was instantly muted and muddled. At a number of points throughout, the nature of ‘the experiment’ is explicit: a guard reminds the prisoners that the experimenters are watching them and had informed them about an incident in this cell; a prisoner reminds a guard that this is an experiment, and another tells a guard that they might be experimenting with you. There are many other failures to create essential testing conditions. Despite the dramatic anonymity-inducing tactic of shaving prisoners’ hair, there is no further attempt to create conditions of de-individuation. Although prisoners are given ID numbers, the guards never refer to them by number but by their names. Despite their common uniforms, those prisoners with flashy tattoos were allowed to wear undershirts that revealed them in all their glory for TV close-ups, an individuating tactic.
  • Heisenberg indeterminacy principle. Sometimes, the very act of measuring and recording a phenomenon changes its nature in unpredictable ways. It seemed evident that most of the participants knew that most of the times, what they said and did was under scrutiny by the experimenters and the TV producers. There are scenes in the privacy of prison cells where prisoners are not talking directly to each other — but into their lapel mikes, for good recordings. Anticipating the national screening of the unfolding events surely changed the behaviour of the chief guard, as noted, and probably for many others as well. Similarly, there are scenes from the solitary confinement chamber where prisoners are openly describing their thoughts and feelings, which of course alter the ‘solitary’ nature of such punitive confinement.
  • Reality concealing rhetoric. The authors reveal a penchant for framing wrong-headed decisions in high-sounding rhetoric. So their prisoner-into-guards alchemy competition is legitimized as the ‘permeability intervention’. Next they add what they call their ‘legitimacy intervention’, which means that prisoners are told that there really was not a good basis for having that prisoner become a guard. Sounds more like an ‘illegitimacy’ intervention, a post hoc tactic to explain away the guards’ racist decision not to accept into their ranks the better deserving black prisoner. Such an intervention may have also provoked inmate anger at the researchers for lying to them.

52 Philip G. Zimbardo

Recall that anger was one of the promised advertised reactions to be expected. Let the record also show that these researchers added a new prisoner to replace the guard- transformed one. Who do they select? Not as might be expected if science were relevant in this enterprise, a volunteer from their wait-listed subject pool of those randomly assigned to be prisoners. Instead, they cast into this TV movie someone chosen because he is an ‘experienced trade union official’. Why? Simply because, in their terms, he could use his skills ‘to envisage the achievement of a more equal set of social relations’. In a second phase of the study, ‘after the collapse of the prisoner- guard system’, the participants continued as a single self-governing ‘commune’, interrupted by ‘what they (the participants) perceived to be opposition from the experimenters’. These quotes are from the authors who are either unaware of, or not shamed, by such blatant imposition of experimenter bias into the research process.These actions also make evident to me that their underlying personal objective was to highjack this media opportunity to advance their evangelical worldview that rational people with free will can rise above situational forces to live in communal harmony, as long as they can sustain a social identity in accord with prosocial values and norms of their community. It is only when people are powerless and there is a breakdown of groups ‘that creates the conditions created under which tyranny can triumph’. In translation, The SPE created those evil conditions, and the BBC-TVexperiment created those holy conditions. Against their examples and quotes from historians of the Holocaust, we can readily counter with the current direct parallels between the SPE abuses and those of Iraqi prisoners by both American and British guards at Abu Ghraib and Basra prisons. One independent investigation of these abuses by the Schlesinger Committee makes explicit the relevance of the SPE that should have served as a forewarning to the military.

  • Data Cautions. The authors get one thing right in noting that given ‘the prolonged interaction between participants’ in their categorically different groups, the unit of statistical analysis should be the group and not the individual group member. However, all of their data analyses are conducted only at the level of individual analyses. But they do warn wary readers that their ‘statistical results need to be read with some caution’ (emphasis added). You cannot have it both ways, so maybe the wiser caution is for readers to dismiss all their results as not interpretable scientifically.
  • Media versus science values. There is much merit in presenting scientific research to the general public; I have been doing so for decades with the Discovering Psychology video series, as well as my participation with the BBC on programmes like 5 Steps to Tyranny, or in the recent neo-reality TV programme on British broadcasting, The Human Zoo. However, it is quite another matter to confuse scientific research and its values of objectivity and adherence to the cannons of scientific method with those values of public media that centre on entertainment and profit. In the case of this particular use of the media to create a scientific study, I find it seriously flawed for the reasons outlined, and many more that space limitations do not allow into this Commentary. My final example focuses on how can observers and psychological scientists ever know what the ‘real data’ are when those in charge of the media production and ultimately paying all the bills, are able to alter the data represented on their video-recordings?

When there is any major data selection or modification of the originally recorded

behaviour, it is critical that we know who was in charge of the selection, what priorities

were dominant — those of science or media. What we do know in this BBC experiment is

Commentary 53

that following an initial screening for the participants, the prisoners objected to the way they were being portrayed as stupid and less in charge than they were. They insisted that the film be re-edited to satisfy these objections, and it was! One prisoner, still unhappy with the final cut, told a reporter, ‘The TV show concentrates on the Big Brother aspect — the funny and sad moments — rather than the science. I think a lot of the more serious side has been cut out because it would make boring TV’.[6] Another participant voiced a similar complaint about what was shown on national television: ‘They said it was meant to be a thorough, scientific programme, but it just looked like a poor man’s Big Brother’.[7] Another insider’s perspective on what really happened diverges considerably from that offered by the authors of this BJSP article. He was one of the prisoner ringleaders of the end-game coup, and a visibly dominant prisoner throughout the BBC prison experience. ‘The prisoners won because they had organized themselves quicker than the guards; their subversive actions and organizational skills outwitted the guards who were disorganized in their new surroundings. They did not understand that they had to organize themselves and form a set of rules that they all agreed on. The prisoners had a common enemy, the guards, so they had only one target. The guards had many subversive individual groups to contend with. The prisoners escaped at night by force/not stealth. This caused the prisoners and guards to form a commune. I think the group is being exploited by the BBC for commercial gain. Me and my new friends in the group joined the experiment for the furtherance of science and not to be used as a form of cheap entertainment’. (Personal e-mail communication, 26 Feb. 2002; supplemented by in person visit, Glasgow 10 Oct. 2004, with Philip Bimpson).

In conclusion, I hate having to challenge my British colleagues in this manner since ultimately it ends discrediting social psychological research more than the researchers. I have made these and more criticisms known privately to journal editors in Britain and the US where this research has been submitted for multiple publications. Nevertheless, they have succeeded in getting their disputable research published at a time when much more significant and more methodologically sound research is being routinely rejected. I now felt obliged to pick up the gauntlet they have cast down not only against the SPE, but also in creating their artificial sub text of European social identity theory pitted against American role theory. Psychology advances by resolving meaningful conceptual challenges and embracing diverse perspectives. I think that the way this research was conducted and the manner in which it has been portrayed in this published article is not in the service of the best interests of our profession.


As a positive postscript to this Commentary, I want to thank these researchers for demonstrating a point that I have long argued in favour of as a means to reduce prisoner abuses, namely greater surveillance of guard-prisoner interactions. This BBC-TV research shows that such violence can be eliminated if all parties in a prison setting realize that their behaviour is open for scrutiny and evaluation.



Rethinking the psychology of tyranny: The BBC prison study

Stephen Reicher1[8] and S. Alexander Haslam2[9]

‘University of St. Andrews 2University of Exeter

This paper presents findings from the British Broadcasting Corporation (BBC) prison study — an experimental case study that examined the consequences of randomly dividing men into groups of prisoners and guards within a specially constructed institution over a period of 8 days. Unlike the prisoners, the guards failed to identify with their role. This made the guards reluctant to impose their authority and they were eventually overcome by the prisoners. Participants then established an egalitarian social system. When this proved unsustainable, moves to impose a tyrannical regime met with weakening resistance. Empirical and theoretical analysis addresses the conditions under which people identify with the groups to which they are assigned and the social, organizational, and clinical consequences of either doing so or failing to do so. On the basis of these findings, a new framework for understanding tyranny is outlined. This suggests that it is powerlessness and the failure of groups that makes tyranny psychologically acceptable.

In the introduction to his text on The Roots of Evil, Staub writes: ‘the widespread hope and belief that human beings had become increasingly ‘civilized’ was shattered by the events of the Second World War, particularly the systematic, deliberate extermination of six million Jews by Hitler’s Third Reich’ (1989, p. 3). The impact of this realization was as marked in academia, and more particularly within academic psychology, as it was in society at large. Indeed, it is arguable that the shadow of the Holocaust lies over the last half century of social psychology and, either indirectly or directly, informs many of the core issues that are of concern to the discipline’s practitioners: questions such as how we come to hate and to discriminate against members of other groups (Tajfel, Flament, Billig, & Bundy, 1971), how people come to see others as less human and less deserving than themselves (Leyens et al., 2003), how the seeds of authoritarianism, social dominance, and power abuse are sown and cultivated (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950), and — the question which concerns us most directly in this 2 Stephen Reicher and S. Alexander Haslam

paper — how we come to condone the tyranny of others or else act tyrannically ourselves.

While the literature in these various areas is both broad and varied, it is possible to identify at least one major trend. That is, there has been a shift away from explanations that focus on the individual characteristics of those who are prejudiced, discriminatory, or even genocidal, towards those that concentrate on the nature of group processes which can induce the most inoffensive of individuals to commit the most offensive of acts (Billig, 1978; Brown, 1965; Milgram, 1974; Sherif, 1966; Tajfel, 1982). In many cases, and certainly when it comes to the psychology of tyranny, theorists have taken the argument one step further and proposed not only that extreme antisocial behaviour must be analysed at the group level, but also that group psychology necessarily tends in the direction of extreme antisocial behaviour. While we strongly endorse the need for a group-level psychology of tyranny (which we define as an unequal social system involving the arbitrary or oppressive use of power by one group or its agents over another), we will take issue with the notion that groups per se are the root of the problem. Indeed, we will argue that powerful and effective groups provide an effective psychological bulwark against tyranny and that it is when groups prove ineffective that tyrannical forms of social organization begin to become attractive.

The equation of groups and tyranny has a long history both within social thought and within social psychology. Hannah Arendt (1998) describes the classical view that ‘the rule by many is not good’ and traces this back to Aristotle’s contention that collective rule leads to haphazardness, moral irresponsibility, and is but a disguised form of tyranny. More recently, such ideas were given substance by crowd psychologists such as Gustave LeBon (1895/1947) who argued that, through submergence in the crowd, individuals lose their individual identity and their sense of responsibility and hence become capable of barbaric and atavistic acts. The notion of submergence was directly transposed into the modern social psychological concept of de-individuation, which is seen to arise from anonymity within a group (Postmes & Spears, 1998; Reicher, Spears, & Postmes, 1995). As one influential de-individuation researcher has put it, ‘mythically, deindividuation is the ageless life force, the cycle of nature, the blood ties, the tribe, the female principle, the irrational, the impulsive, the anonymous chorus, the vengeful furies.’ (Zimbardo, 1969, p. 249).

Tyranny as role and power

Though well known as a de-individuation theorist, Zimbardo is better known for his work on the Stanford prison experiment (SPE; Haney, Banks, & Zimbardo, 1973; Zimbardo, 1989; Zimbardo, Maslach, & Haney, 1999). Indeed, this is one of the most famous social psychological investigations ever conducted, representing the culmination of a series of ‘classic’ field studies into the roots of extreme behaviours that were conducted in the aftermath of World War Two (Milgram, 1963; Sherif, 1956). Building on earlier studies, it played a critical part in cementing the shift that we have described from individual to group-level explanations of extreme behaviours (Banuazizi & Movahedi, 1975). Moreover, it was one of the few studies that not only addressed the issue of tyranny but, due to the power of the research paradigm, also produced direct evidence of tyrannical behaviour. While it is not, strictly speaking, a study of de-individuation, Zimbardo certainly used his general understanding of the group as a corrosive force to explain events in the SPE.

The psychology of tyranny                                                                                                                   3

The SPE is remembered for showing that, simply as a consequence of assigning college students the role of guard or prisoner, the former became increasingly brutal while the latter became passive and began to show signs of psychological disturbance. Such was the severity of these phenomena that the study, originally scheduled to last 2 weeks, had to be stopped after 6 days. Zimbardo and colleagues explained their findings by commenting that guard aggression ‘was emitted simply as a ‘natural’ consequence of being in the uniform of a ‘guard’ and asserting the power inherent in that role’ (Haney etal., 1973, p. 12). Thus, immersion in a group is seen to undermine the constraints that normally operate upon people when they act as individuals. In addition, when those groups have power at their disposal, this is believed to encourage extreme antisocial behaviour (Zimbardo, 1969).

Although these findings were significant in their own right, the impact of the SPE was as much ethical as theoretical (e.g. see Smith & Mackie, 2000, p. 49). Indeed, the very extremity of the results led many (including Zimbardo himself) to question the legitimacy of subjecting participants to such situations. The acceptability of conducting any sort of large-scale field interventions thus became a focus for vigorous debate (e.g. Herrera, 1997; Lindsay & Adair, 1990; Sieber, Iannuzzo, & Rodriguez, 1995; Smith & Richardson, 1983). Paradoxically, then, at the same time that the SPE marked the culmination of post-war field studies, it also led to their cessation.

Accordingly, since the 1970s, social psychology has been increasingly dominated by laboratory experiments in which there is minimal or no interaction between participants and scant attention paid to the role of personal and group history or to the development of interactions over time (Bar Tal, 2004; Doosje, Spears, & Ellemers, 2002; Haslam & McGarty, 2001; Levine, 2003; Moreland, Argote, & Krishnan, 1996). Moreover, this unwillingness to undertake studies that create, manipulate and systematically investigate the effects of social environments on human interaction can be seen to have contributed to the increasing dominance of explanations based upon inherent and essentially unavoidable genetic, biological, or psychological propensities. It has also led to an increasing disjunction between the issues that motivate social psychological studies and the nature of those studies themselves. Research reports (and certainly most bids for research funding) typically start by alluding to large-scale topics such as oppression, discrimination, and genocide, but then go on to pursue an empirical strategy that seems very remote from the social realities of such phenomena (e.g. seeking to explain these phenomena in terms of individual-level subconscious processes from a cognitive or, more recently, neuroscientific perspective; Ito, Thompson, & Cacioppo, 2004). As Zimbardo (quoted in Brockes, 2001, p. 2) has argued, partly as a result of these trends, psychology has become increasingly marginal to, and marginalized from, debate surrounding important social issues.

In terms of the specific issue of tyranny, the ethical concerns that have placed the SPE ‘off-limits’ (with the exception of a partial replication by Lovibond, Mithiran, & Adams, 1979) have led to a situation in which the conclusions of that study have become almost inviolate and social psychological inquiry into tyranny has effectively ground to a halt. Barred from employing the power of the SPE paradigm, it is all but impossible to produce behaviours that are powerful enough to match those found by Zimbardo and his colleagues. Hence, even if researchers harbour doubts about the extreme situational determinism and negative views of the social group, which are used to explain these findings (and many do; e.g. see Ellemers, Spears, & Doosje, 1999; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Haslam, 2001), it has not been possible to produce data that 4 Stephen Reicher and S. Alexander Haslam

can give substance to those doubts and hence reopen scientific debate about the psychological bases of tyranny.

However, quite apart from the intrinsic importance of the topic, there are at least two sets of reasons why revisiting the SPE is long overdue. First, any assessment of the conclusions drawn from the SPE is inevitably limited by the fact that only a small proportion of the interactions in the study were recorded (because filming was intrusive and limited) and, of these, only a very small number are in the public domain. Moreover, observational data were never complemented by other data sources that would allow for controlled measurement of key behaviours and the psychological states seen to underlie them. At the very least, there is a need for a fuller and more transparent data set, which might progress empirically grounded and open debate about the psychological bases of tyranny.

However, even the limited amount of data that is available from the SPE casts doubt on the analytic conclusions that have been drawn from it. Where participants did behave in role, it is unclear whether, as Zimbardo and his colleagues claim, this was due to their ‘natural’ acceptance of role requirements or due to the leadership provided by the experimenters (Baron, 1984; Banuazizi & Movahedi, 1975). This is because during the study, the guards were given clear guidance as to how they should behave. Notably, when Zimbardo briefed his guards, he told them:

You can create in the prisoners feelings of boredom, a sense of fear to some degree, you can create a notion of arbitrariness that their life is totally controlled by us, by the system, you, me — and they’ll have no privacy. They’ll have no freedom of action, they can do nothing, say nothing that we don’t permit. We’re going to take away their individuality in various ways. In general what all this leads to is a sense of powerlessness (Zimbardo, 1989).

The importance of such guidance is demonstrated by the research of Lovibond et al. (1979) who conducted a study in which the guards were trained to respect the prisoners as individuals and to include them in decision-making processes. Under these conditions, the ensuing behaviour of both guards and prisoners was far less aggressive and extreme (Lovibond et al., 1979).

Yet, even with guidance, many of the participants in the SPE behaved out of role for much of the time (Baron, 1984). The available video material shows that both prisoners and guards challenged their roles not only at the start, but throughout the entire study. In the case of the guards, Zimbardo (1989) notes that, while some exploited their power, others sided with the prisoners and yet others were tough but fair. Such diversity sits uneasily with the notion that role acceptance is simply determined by the situation. It suggests that the emphasis on role acceptance and tyranny is one-sided and that there is a need to focus on (a) the conditions under which people do or do not assume their roles and (b) the balance between tyranny and resistance.

An alternative analysis: The social identity approach

It is not only that some of the data from the SPE appear to sit uneasily with a role account. Increasingly, the role account — and indeed the generally negative view that group membership leads to a loss of constraints on antisocial behaviour — is at odds with developments in group psychology. One of the most influential of these is the social identity approach (incorporating social identity theory; Tajfel, 1978, 1982; Tajfel & Turner, 1979; and self-categorization theory; Turner, 1985; Turner et al., 1987; Turner, Oakes, Haslam, & McGarty, 1994). According to this approach, people do not

The psychology of tyranny                                                                                                                   5

automatically act in terms of group memberships (or roles) ascribed by others. Rather, whether or not they do so depends upon whether they internalize such memberships as part of the self-concept (Turner, 1982).

Self-categorization theory in particular has argued that this act of self-definition in terms of group membership (social identification) forms the psychological basis of group behaviour and that the character of such behaviour depends upon the norms, values, and understandings that characterize the particular category in question (Turner, 1982, 1999). Thus, while members of certain groups may indeed use their power to act in discriminatory and oppressive ways, members of other groups may act more prosocially and use their power for constructive purposes (Pfeffer, 1981; Postmes & Spears, 1998). Moreover, even if some groups are tyrannical, group action is also the basis on which people gain the strength and confidence to resist, to challenge, and even to overthrow tyranny (Reicher, 1996; Tajfel, 1978).

Consistent with such emphasis, the greater part of early work informed by social identity theory has focused on the conditions under which people act to change inequalities between groups (e.g. Robinson, 1996). In broad terms, it is assumed that people who are positively valued by virtue of their group membership (e.g. members of dominant groups) would identify with and act in terms of the group. For people who are negatively valued by virtue of their group membership (e.g. members of subordinate groups), collective action is contingent upon two sets of factors in particular (Tajfel, 1978; Tajfel & Turner, 1979). The first relates to beliefs about one’s ability to advance through the social system despite one’s group membership (i.e. the permeability of category boundaries). The second concerns the perceived security of intergroup relations and comprises two further elements: the perceived fairness of intergroup inequalities (legitimacy) and their perceived stability. When relations are perceived to be insecure, this is characterized by the fact that individuals are aware of cognitive alternatives to the status quo and hence can envisage specific ways in which it could be changed.

Permeability affects whether people act individually or collectively, so that a belief that movement across boundaries is possible encourages strategies of individual mobility, but a belief that such movement is impossible encourages people to perceive themselves and act as group members (e.g. Ellemers, 1993; Wright, Taylor, & Moghaddam, 1990). Whether or not people then challenge inequality is also dependent upon intergroup relations being perceived as insecure. That is, people should be most inclined to resist domination when they see inequality as both illegitimate and unstable and can thus envisage cognitive alternatives to it (Turner & Brown, 1978; see also Ellemers, Wilke, & van Knippenberg, 1993; Tajfel, 1978; Tajfel & Turner, 1979).

It is important to stress that social identity theory does not constitute a comprehensive theory of domination and resistance. Most notably, it has little to say about the concomitants of identity processes (e.g. organizational and clinical factors), which may impact upon the ability of group members to act effectively. These are critical issues that we aim to investigate here. Nonetheless, the social identity approach provides a well-articulated and contemporary perspective from which to revisit the issues raised by the SPE: What are the psychological consequences of intergroup inequality? When do people seek to impose such inequality? And when do they resist it?

Before explaining how we addressed these issues, it is important to consider a second set of reasons for revisiting the SPE. These have less to do with the explanation of the findings themselves than with their broader social relevance. For Zimbardo and his colleagues, the results of the SPE were intended to bear directly on the nature of 6 Stephen Reicher and S. Alexander Haslam

prison regimes in the United States. Thus, they refer to the setting as a ‘simulated prison’ and claim that ‘this simulated prison environment developed into a psychologically compelling prison environment’ (Haney et al., 1973, p. 69). However, as Banuazizi and Movahedi (1975) point out, there are good reasons to doubt these claims, firstly, because there are critical features of the SPE that are very different from a real prison (e.g. participants know they have committed no crime and can ask to leave at any time) and, secondly, because even where there are material similarities, the two are very different phenomenologically (e.g. where the walls of a prison remind the inmates that they must be kept apart from ‘decent’ people, they remind the participants in the SPE that they are honourable participants in adventurous scientific research).

While these arguments are in themselves controversial (e.g. Thayer & Saarni, 1975), our point is that, even if they are accepted, they do not render the SPE practically irrelevant. For there is another level beyond that of phenomenal equivalence through which the study can (and is) claimed to have real-world implications. That is, as is the case with most psychological research, generalization is theoretically, not empirically, based (Haslam & McGarty, 2004; Turner, 1981). Thus, Zimbardo (e.g. 2001) uses his study to establish the theoretical claim that people ‘naturally’ assume roles, and then uses this theoretical analysis to explain a wide range of phenomena from prison behaviour to terrorism. Of course, the resultant behaviours will not be the same as those observed in the SPE, but will depend upon the precise nature of the roles and role requirements in the relevant domain of application.

It is precisely this ability to apply role theory to a broad range of domains, despite phenomenal and behavioural differences, that has ensured the impact of the SPE within and beyond psychology amongst those who have little interest in matters relating to prisons. It is the validity of this theoretical account that we wish to address in the present paper, not only for the empirical and theoretical reasons discussed above, but also because we feel that the traditional analysis of the SPE has profound and troubling social implications. If people cannot help but act in terms of assigned role, it implies that they have little choice, and hence little responsibility, for their social actions. This makes it more difficult to hold tyrants to account for what they do. Moreover, in communicating the message that resistance is futile, the analysis discourages the oppressed from attempting to challenge tyranny.

The BBC prison study

In December 2001, we conducted a major social psychological field study (possibly the largest such study in the three decades since the SPE) in conjunction with the documentaries unit of British Broadcasting Corporation (BBC; hence the title ‘the BBC prison study’). The study was independently designed, operationalized, run, and analysed by the authors. In this sense, it was an ordinary piece of scientific research, which went through all the normal scientific procedures (including ethical approval), and whose features were designed in relationship to the theoretical issues that concerned us. The contribution of the BBC was to coordinate and manage the logistical task of (a) creating the study environment (in line with our guidelines), (b) filming the study, and (c) preparing some of the resultant material for broadcast. In short, the project can be described as ‘original science filmed’. This made it a unique collaboration that was markedly distinct from ‘reality television’ enterprises in which a television company devises certain scenarios with issues of entertainment in mind and then invites academics to comment on them.

The psychology of tyranny 7

Over a period of 8 days, the study examined the behaviour of 15 men who were placed in a social hierarchy of guards and prisoners within a purpose-built environment. Their behaviour was video- and audio-recorded over the entire period, and this was complemented by daily psychometric and physiological measures. The video data were edited into four 1-hour long documentaries screened in May 2002 (Koppel & Mirsky, 2002).

The aim of the study was not to simulate a prison (which, as in the SPE, would have been impossible on ethical and practical grounds) but rather to create an institution that in many ways resembled a prison (but also other hierarchical institutions such as a school, an office, abarracks; see Morgan, 1979) as a site to investigate the behaviour of groups that were unequal in terms of power, status, and resources. What is critical, then, is not that the study environment replicated a real prison (which no such environment ever could), but that it created inequalities between groups that were real to the participants.

Similarly, our aim was not conduct an exact replication of the SPE (which, would also have been impossible for ethical reasons). Rather, it was to use a different system of intergroup inequality in order to revisit the conceptual issues raised by the SPE. We therefore do not invite comparison with the SPE in terms of the exact details of how people behaved but rather in terms of the ability of different explanatory frameworks to make sense of what happened. Do participants accept their roles uncritically? Do those accorded group power exercise it without constraint, and do those without group power accept their subordination without complaint? After all, if the process of role enactment is indeed ‘natural’, then it should apply in all cases and any exception is troubling for the overall claim. Do the concepts used by social identity and selfcategorization theorists provide a more satisfactory account of when people do (and do not) adopt the social positions ascribed to them?

Our study can thus be seen as an experimental case study of the behaviour of members in dominant or subordinate positions and of the developing relations between them. Unlike the SPE, it is not purely exploratory but rather is theoretically informed (by a social identity perspective). Hence, and again in contrast to the SPE, we included manipulations of theoretically relevant variables. Given the practical impossibility of running multiple sets of groups (due to the massive resources required in terms of equipment, personnel, and money), a time-series approach was adopted whereby interventions were introduced at predefined points in the study and their effects on the development of intra- and intergroup relations then analysed.

To summarize, the overall aims of our study were as follows:

(a)     To provide comprehensive and systematic data pertaining to the unfolding interactions between groups of unequal power and privilege.

(b)     To analyse the conditions under which people (i) define themselves in terms of their ascribed group memberships and act in terms of group identities, and (ii) accept or else challenge intergroup inequalities. Specifically, we predict that dominant group members will identify with their group from the start and impose their power. However, subordinate group members will only identify collectively and challenge intergroup inequalities to the extent that relations between groups are seen as impermeable and insecure.

(c)     To examine the relations between social, organizational and clinical factors in group behaviour.

(d)     To develop protocols that provide a practical and ethical framework for examining social psychological issues in large-scale studies.

8 Stephen Reicher and S. Alexander Haslam Method and ethics

The study was designed to create a hierarchical society in which people would live for up to 10 days. It was conducted within an institutional environment that was constructed inside Elstree Film Studios in north London. Prisoners were allocated to lockable 3-person cells that were located, together with showers, off a central atrium. This was separated by a lockable steel mesh fence from the guards’ quarters (a dormitory, bathroom, and mess room). A plan of the prison is presented in Figure 1.

Comprehensive details of the procedures are available in Haslam and Reicher (2002) or from the authors. What follows is a description of the key features of the study.


The study aimed to create a system of intergroup inequality that was meaningful but was not harmful to participants either physically or mentally. To ensure that no harm eventuated, details of the experimental set-up and planned manipulations were discussed with colleagues and submitted both to the University of Exeter’s ethics panel and to the Chair of The British Psychological Society’s Ethics Committee prior to the study being conducted. The novelty of the experimental manipulations and theoretical analysis constituted central components of the scientific case that was presented in order to justify the research. Moreover, the following safeguards were built into the study:

(a)    Potential participants went through 3-phase clinical, medical, and background screening to ensure that they were neither psychologically vulnerable nor liable to put others at risk (see below).

i____________ i


= sliding lockable grille Д = location of fixed camera

Figure 1. Plan of the prison.

The psychology of tyranny 9

(b)     Participants signed a comprehensive consent form. Amongst other things, this informed them that they may be subject to a series of factors — including physical and psychological discomfort, confinement, constant surveillance and stress — which may involve risk.

(c)     Two independent clinical psychologists monitored the study throughout, and had the right to see any participant at any time or to demand that any participant be removed from the study.

(d)     A paramedic was on constant standby in case of illness or injury.

(e)     On-site security guards were provided with detailed protocols clarifying when and how to intervene in cases of dangerous behaviours by participants.

(f)      An independent 5-person ethics committee — chaired by a British Member of Parliament — monitored the study throughout. This committee had the right to demand changes to the study’s set-up or to terminate it at any time.

Apart from minor ailments (blisters, etc.) that were treated by the paramedic, no interventions were necessary to address ethical concerns. After the study, the ethical committee published an independent report (McDermott, Opik, Smith, Taylor, & Wills, 2002) and characterized the conduct of the study as ‘exemplary’.

Selection of participants

Male participants were recruited through advertisements in the national press and through leaflets. Applicants went through three phases of screening. First, they completed a battery of psychometric tests that measured both social variables (authoritarianism, social dominance, modern racism) and clinical variables (depression, anxiety, social isolation, paranoia, aggressiveness, demotivation, self-esteem, self-harm, drug dependence). Second, they underwent a full weekend assessment by independent clinical psychologists. Third, medical and character references were obtained, and police checks were conducted.

For ethical reasons (noted above), we sought to include in the study only people who were well-adjusted and prosocial, scoring at low levels on all social and clinical measures. Additionally, we wished to ensure that the individual dispositions of our participants were such that, if the dynamics of the study produced antisocial actions in this sample (as in the SPE), then it could reasonably be supposed that they would have such an effect upon almost anybody in the population.

The screening reduced an initial pool of 332 applicants to 27 men (we recruited only men to ensure comparability with the SPE and to avoid ethical issues that would arise from placing men and women together in cells). The final sample of 15 was chosen to ensure diversity of age, social class, and ethnic background. They were randomly divided into two groups of 5 guards and 10 prisoners but in such a way as to ensure that the two groups were matched on key dimensions. More specifically, the 15 participants were first divided into five groups of 3 people who were as closely matched as possible on personality variables potentially implicated in tyranny: modern racism, authoritarianism and social dominance.[10] From each group of three, one participant was then randomly selected to be a guard (and the remaining two to be prisoners). This procedure 10 Stephen Reicher and S. Alexander Haslam

was conducted blind (i.e. the identities of the participants were not known to the experimenters). Note, however, that due to our experimental manipulations (explained below), the actual number of people in the two groups varied over time. Thus, at the start of the study, there were 5 guards but only 9 prisoners — the 10th being introduced at a later stage.

Data sources

The prison environment was designed in such a way that participants could be both video- and audio-recorded wherever they were. At all times, four channels were videorecorded and all audio channels were recorded. There was also daily psychometric testing. Measures were taken from a battery of scales. For the purposes of the present analyses the critical measures were:

(a)     social variables: social identification, awareness of cognitive alternatives, rightwing authoritarianism;

(b)     organizational variables: compliance with rules, organizational citizenship; and

(c)     clinical variables: self-efficacy, depression.

In order to minimize fatigue, not every measure in the full battery was administered every day. However, each was administered on multiple occasions to allow for an analysis of development over time. Finally, daily swabs of saliva were taken in order to ascertain cortisol levels (as an indication of stress; Laudat et al., 1988). For reasons of space, however, we will not consider the cortisol data here (see Haslam & Reicher, in press).


Five participants were invited to a hotel the evening before they entered the prison. On arrival, they were told that they would be guards in the study. They were shown the prison timetable — which included such elements as cleaning chores, work duties, prisoner roll calls, exercise time and a recreational hour — and were told that their responsibility was to ensure that the institution ran as smoothly as possible and that the prisoners performed all their tasks. The five guards were then asked to draw up a series of prison rules under headings provided by the experimenters and to draw up a series of punishments for rule violations.

The guards were given no guidance about how they should achieve their goals. The only limits on what they could do were a set of ethically determined ‘basic rights’ for prisoners. In particular (and as in the SPE), all participants were told that physical violence would not be tolerated (for rules, punishments, and rights, see Haslam & Reicher, 2002). Beyond this, however, it was stressed that the guards could act as they pleased.

On the morning of the study itself, the guards were taken in a blacked-out van to the prison (since this was meant to be their entire experiential world for the duration of the study, it was important that they could not imagine the outside). Once inside, they were given a full briefing by the experimenters on the prison layout and the resources available to them.

The guards had a series of means by which to enforce their authority, including keys to all doors inside the prison (including a punishment isolation cell), sole access to an

The psychology of tyranny 11

upper level, a ‘guards’ station’ with a surveillance system from which they could see into the prisoners’ cells, resources (including snacks and cigarettes) to use as rewards or withdraw as punishments — and, in addition, the ability to put prisoners on a bread and water diet. They also had far better conditions than the prisoners, including superior meals, extra supplies of drinks and snacks, superior living conditions and well-made uniforms as opposed to the prisoners’ uniform of a t-shirt printed with a 3-digit number, loose trousers and flimsy sandals. The prisoners also had their hair shaved on arrival.

After their briefing, the guards changed into their uniforms and practiced the procedure for admitting the prisoners. The nine prisoners then arrived one at a time. They were given no information apart from the prison rules, a list of prisoners’ rights, which was posted in their cells, and a very brief loudspeaker announcement from the experimenters. This introduced the permeability intervention (see below) and stressed that violence was not permissible.

Planned interventions


At their initial briefing, the guards were told that they had been selected on the basis of their reliability, trustworthiness and initiative as gleaned from pre-selection assessment scales. However, they were also told that while these scales were reasonably reliable, they were not perfect. In particular, the experimenters stated that it was possible that they had misassigned one or more of the prisoners. Hence, the guards were told that they should observe the behaviour of the prisoners to see if anyone showed guardlike qualities. If they did, they were told that there was provision for a promotion to be made on Day 3. This information was also announced to the prisoners over the loudspeaker. In the initial days of the study, participants were thus led to believe that movement between groups was possible (see also Ellemers, van Knippenberg, & Wilke, 1990; Wright et al., 1990). After the promotion of one prisoner to guard actually took place (the selection of the individual being made by the guards on the basis of a procedure suggested by the experimenters), the possibility of movement was removed by announcing that there would be no further promotions (or demotions).


Three days after the promotion, participants were to be informed by the experimenters that observations had revealed that there were in fact no differences between guards and prisoners on the key group-defining qualities. However, they would be told that it was impractical to reassign them and hence the groups would be kept as they were. Accordingly, whereas previously the group division had been legitimate, this would no longer be the case (Ellemers et al., 1993).

Cognitive alternatives

Within a day of the legitimacy intervention, a new prisoner was to be introduced. Although he was as naive as the others, he was chosen for this role (from the pool of 10 participants randomly assigned to be prisoners) because of his background as an experienced trade union official. On this basis, we expected that he would introduce a new perspective to the prison based on notions of group-based negotiation and collective- and equal rights (i.e. a perspective that suggested the existing regime was both illegitimate and changeable). It was also thought that he might provide skills 12 Stephen Reicher and S. Alexander Haslam

necessary to organize collective action (e.g. see Haslam, 2001). Hence, it was envisaged that his introduction would enable the prisoners (and the participants more generally) to envisage the achievement of a more equal set of social relations.


The findings of the study can be divided into two phases. In the first phase, the guards failed to identify with each other as a group and to cohere collectively. By contrast, after the promotion on Day 3, the prisoners did increasingly identify as a group and work collectively to challenge the guards. This led to a shift of power and ultimately to the collapse of the prisoner-guard system. In the second phase, participants decided to continue as a single self-governing ‘commune’. However, they were unable to deal with internal dissent and lost confidence in the communal system. By the end of the study, they were increasingly disposed to tolerate a new and much more draconian system of inequality that some participants now wished to impose.

Results that pertain to these two phases will be presented in turn. For each phase, we combine a description based on the observational data with statistical analyses based on the quantitative measures. The statistical analyses are based on the data of individual participants within groups (excluding data from the participant who was promoted from prisoner to guard on Day 3 and from the prisoner who was introduced on Day 5). Given the interaction between participants, it could be argued that the group, rather than the individual group member, should be the unit of analysis here. For this reason, the present data were also analysed using methods advocated by McGarty and Smithson (2005), which do not assume (or require) independence of observations. These analyses confirmed the reliability of all the patterns reported below. However, for reasons of space and in light of the novelty of these alternative methods, we accord with general usage by presenting statistics that are commonly used even where there is interaction between participants (Hoyle, Georgesen, & Webster, 2001).

Phase 1: Rejecting inequality

Social identification

For the prisoners, the development of social identification was consistent with predictions. From the start, they were clearly dissatisfied by their inferior conditions. Initially though, as predicted in light of the permeability of group boundaries, many sought to improve their lot by displaying the individual qualities necessary for promotion as opposed to mounting a collective challenge to the guards. As a result, there was no shared identity among the prisoners and no consensus about how they should behave (see Photograph 1).

However, after the promotion, when group boundaries were impermeable and participants could only alter their position by changing the general prisoner-guard relationship, the prisoners began to develop a much stronger sense of shared social identity and to develop more consensual norms — particularly in relation to their treatment of the guards. This contrast between the pre- and post-promotion periods is exemplified in exchanges between participants in Cell 2. Before the promotion, two occupants of this cell, JEp2 and KMp, worked conscientiously and explicitly sought to improve their position by displaying behaviours required to become a guard. As JEp put

The subscripts p and g after participants’ initials denote their status as prisoner or guard, respectively.

The psychology of tyranny                                                                                                                 13

Photograph 1. Day 2: The prisoners comply with the guards’ regime.

it, ‘I’d like to be a guard because they get all the luxuries and we are not’. However, almost immediately after the promotion, all three occupants of this cell (i.e. including PPp) recognized that the only way to improve their position was to change the system. Accordingly, they began to discuss how they could achieve this together. If anyone expressed doubts about this objective, they were reminded of the collective inequalities in the prison and of the need for collective resistance. This is exemplified by the following interchange between the cellmates:

Extract 1

JEp: Hopefully we’ll get [TQg] in. That’s the person, he’s the target.

KMp: No. I mean obviously I think it’s going to be a lot of fun for us to do this but I don’t think [TQg] … I feel so … I just feel . . .

PPp: Listen, listen mate I, you’ve got to, you’ve got to start forgetting about other people’s feelings and what they’re doing because the days when you’re sitting here starving hungry and you’ve got fuck all and you’ve got nothing mate and you’ve got a ratty little bed and a stupid little blanket to sit under and they’re under there in their duvets, they’ve got everything they want and they’re not giving two fucks about you. So — think on and fuck them.

KMp: I think they do care about us. But guys I’m going to back you all the way. You should no’ doubt me.

For the prisoners, then, the promotion led to a perception of impermeability. This was accompanied by a shift from individual action and identification (i.e. a stress on what ‘I’ will do) to collective action and identification (i.e. discussion of what ‘we’ will do) and from compliance to conflict with the guards.

For the guards themselves, the results were very different. Moreover, the patterning of their social identification went against our predictions. Along lines reported by Zimbardo and his colleagues (Haney et al., 1973), we had expected that, from the outset, they would identify with what was a high-status and positively valued group within the prison. There was some evidence of this in the first day of the study, but it was also clear that several guards were wary of assuming and exerting their authority.

Consequently, some guards were always ambivalent about internalizing their assigned identity. Even to the extent that they overcame this ambivalence, the group as a whole could never reach consensus about its norms and priorities. In addition, over


14 Stephen Reicher and S. Alexander Haslam

time — particularly once the boundaries between-groups became impermeable and the prisoners had developed a stronger sense of shared identity — these cleavages were exacerbated and the fragility of the guards’ collective identity became more apparent.

However, as noted above, in contrast to the SPE, our systematic collection of quantitative data allowed us to complement behavioural observations that speak to these patterns with inferential statistical analysis (and hence to triangulate behavioural and psychometric data). To this end, social identification with prisoner and guard groups was measured every day by means of two 3-item scales containing items used widely in previous research (e.g. Doosje, Ellemers, & Spears, 1995; ‘I feel strong ties with the prisoners [guards] ’/‘I identify with the prisoners [guards]/I feel solidarity with the prisoners [guards])’. As with all other measures, responses were made on 7-item scales with appropriately labelled end-points (e.g. not at all, extremely).

Identification scores were computed by averaging responses on the three measures that comprised each scale and subtracting the out-group identification score from the ingroup identification score. Note too that, as with all other measures, data were collected early each morning. Accordingly, scores relate primarily to events of the previous day.

Given the small number of participants, the study inevitably has a low level of statistical power. In light of this, statistical significance was considered in conjunction with effect size (Cohen, 1977; Smithson, 2000). For all analyses, effects were only considered meaningful when, as well as being significant at conventional levels (a = .05), effect size was strong by Cohen’s (1977, p. 283) criteria (i.e. q2 > .14).

Figure 2 presents mean social identification scores as a function of participant group and time. As can be seen from this graph, the data here are consistent with the above observations. In the very first days of the study, the guards identified more strongly with their social position than the prisoners. However, as soon as they had to implement the disciplinary regime (from Day 2 onwards), the guards’ identification fell while that of the prisoners rose. Prisoner identification was also particularly high after the promotion (Day 4).

Statistical analysis of in-group identification scores was conducted by means of a 2 (participant group: guards, prisoners) X 6 (study phase: Day 1 to Day 6) ANOVA with repeated measures on the second factor. This analysis supported observational data in revealing a significant and large interaction between group and time, F(5, 55) = 3 05, p < .05; q2 = .22. Analyses of polynomial trends were used as the most practical way of decomposing this effect (Norusis, 1985). Consistent with observations, these showed

Study phase

The psychology of tyranny 15

that social identification among the prisoners increased linearly over time, t (7) = 2.46, p < .05. On the other hand, identification among the guards declined as the study progressed, but non-significantly, t(4) = – 0.77, ns.[11]

Security of intergroup relations

After the promotion, the normative consensus among prisoners led to effective organization based on both the expectation and the provision of mutual social support between group members (Haslam, O’Brien, Jetten, Vormedal, & Penna, 2005; House, 1981; Underwood, 2000; see Haslam & Reicher, in press, for a fuller treatment of this point). Conversely, the guards’ inability to agree on group norms and priorities meant that they could not trust others to represent them or act appropriately. As a consequence, they felt (and were) weak, inconsistent, and ineffective as a group.

These developments directly undermined the perceived legitimacy of the intergroup inequality as the guards showed no evidence of the qualities by which their selection was supposedly justified. Such developments also contributed directly to the emergence of cognitive alternatives; not only could the prisoners envisage changes to the existing hierarchy, but they also had a growing confidence in their ability to achieve them. Consider, for instance, the outcome of the first confrontation between prisoners and guards. The three prisoners of Cell 2 had orchestrated this mainly to see how the guards would respond when challenged. As part of a pre-arranged plan, JEp threw his lunch plate to the ground and demanded better food. As the guards tried to intervene, KMp and PPp then joined in with further demands for smoking rights and treatment for a blister. The guards were totally disunited in their response, with some wanting to take a disciplinarian line and others wanting to make concessions (see Photographs 2 and 3) . This continued until one guard, TQg, decided to resolve the situation by acceding to PPp’s requests for a cigarette in order to encourage the other prisoners to return to their cell.

After the incident, the prisoners back in their cell and the guards back in their mess expressed a growing realization that the system was open to change. However, they did so with a very different evaluative tone. The prisoners literally danced with joy. PPp exclaimed ‘that was fucking sweet’, to which KMp responded (admiringly) ‘you was fucking quality man’. Then all three cellmates exchanged ‘high fives’. The guards bickered despondently. TAg started by saying ‘This is only Day 4. They can see what happened today and now they know they can do whatever they want’. As BGg and then TQg weakly disagreed, TAg only insisted more, culminating in the claim that ‘It’s happened. I mean, come on. You know what that has done? That has lit the fuse on [PBp’s] arse’.

Hence, the induction of insecure relations between the groups was not dependent upon our interventions but was an emergent property of the intergroup dynamics. Correspondingly, planned interventions were not necessary for the prisoners to start challenging the guards. The challenge started mounting immediately after the promotion. Consequently, the legitimacy intervention was not implemented. However, because the new prisoner had already been told he would be brought in, this went ahead on Day 5 (and he was later withdrawn on Day 6). Given the situation, he did not so much suggest cognitive alternatives where none existed as suggest additional alternatives to the status quo. Most notably, he began to question the prisoner-guard


16 Stephen Reicher and S. Alexander Haslam

Photographs 2 and 3. Day 4: The prisoners start to challenge the guards’ regime.

division itself and, by questioning the legitimacy of certain aspects of the study (particularly the heat in the prison), he suggested both to his fellow cellmates (DDp and FCp) and to one of the guards (TQg) that the participants mount a united challenge to the experimenters. As the following extract shows, this provided new ways of envisaging how the system might work:

Extract 2

DMp: If this was a real-life situation . . .

TQg: Yes.

DMp: . . . and you were working in this kind of heat, then you as an employee could well go to the employer and say ‘The condition is unacceptable, I’m not prepared to work in it’. Now let’s treat this as a real-life situation. You and I — your group and the group I’m in — both have this problem of the heat. And if I’ve got to sleep in this, there is no way I will. And I, you know, won’t bear it. And I think collectively we should do something about it to the people who are running the experiment. Now you know in a normal, day-to-day, real-life situation, that’s what would happen.

TQg: Well, I am most impressed with your new-found kind of angle on this, which possibly shouldn’t come as a surprise to me. But I think that is a very very valid point you are making and I’m going to go along with it completely.

As a first practical step towards participant unity, DMp created a new negotiating structure, which brought guards and prisoners together on a basis that was far more equal than had hitherto been the case. The guards were eager and pleased to accept this arrangement, for even if they surrendered much of their hierarchical advantage by doing so, the new system confirmed their position in what promised to be a viable social order.

Quantitative confirmation of these patterns emerges on measures of participants’ awareness of cognitive alternatives to the status quo (Tajfel & Turner, 1979). These were administered on Days 1, 3, 4, and 6 using a 4-item scale (‘I cannot imagine the relationship between guards and prisoners being any different (reverse-scored)’/‘I think that the guards will always have more privileges than the prisoners (reverse-scored)’/‘I think that the relationship between guards and prisoners is likely to change’/‘I think that it would be possible for the prisoners to have more power than the guards)’. A single score was computed by averaging responses to these 4 items.

The psychology of tyranny                                                                                                                 17

Study phase

Figure 3. Awareness of cognitive alternatives as a function of assigned group and time.

Mean scores are presented in Figure 3 as a function of participant group and time. As can be seen from this graph, at the start of the study, prisoners and guards had an equal and relatively low sense of cognitive alternatives, but this increased as the study progressed. These patterns were confirmed by statistical analysis of cognitive alternatives data that was conducted by means of a 2 (participant group: guards, prisoners) X 4 (study phase: Days 1, 3, 4, 6) ANOVA with repeated measures on the second factor. This analysis revealed a significant and large main effect for time, F(3, 33) = 3 36, p < .05; h2 = .34. Consistent with the above observations, analysis of polynomial trends to decompose this effect indicated that participants’ sense of cognitive alternatives increased linearly over time, t (12) = 2.83, p < 05.

Acceptance of the unequal regime: Compliance and organizational citizenship Associated with the above effects, as the prisoners developed a sense of shared social identity that was defined in opposition to the guards, and as they became aware of the possibility of alternatives to existing status relations, they started to work actively against the regime that the guards were trying to uphold. This was manifest both in minor challenges to the guards’ status (e.g. insubordination during roll call) and in more overt challenges — including the incident alluded to above, in which a number of prisoners collectively protested about the quality of their food.

Data consistent with this analysis emerge from measures of (a) participants’ willingness to comply with authority and (b) their willingness to engage in acts of organizational citizenship (i.e. to do more than was asked of them in order to make the prison system work; Organ, 1988, 1997).

Compliance was measured on Days 1, 3 and 5 by means of a 2-item scale (‘I try to do what the guards want’/‘I try to comply with the guards’ rules’). Responses on these measures were averaged to provide a single score. The resultant mean scores are presented in Figure 4 as a function of participant group and time. As can be seen from this graph, in the early stages of the study, both guards and prisoners were willing to comply with the rules, but the prisoners became more reluctant to comply after the promotion.

These patterns were confirmed by statistical analysis of compliance scores that was conducted by means of a 2 (participant group: guards, prisoners) X 3 (study phase: Days 1, 3, 5) ANOVA with repeated measures on the second factor. This revealed a significant


18 Stephen Reicher and S. Alexander Haslam

□ Prisoners

□ Guards

Day 1                   Day 3                  Day 5

Study phase

Figure 4. Compliance with prison rules as a function of assigned group and time.






7 6 5 4 3 2 1

and large interaction between group and time; F(2,22) = 3.62, p < .05; r\2 = .32. Analyses of polynomial trends indicated that compliance on the part of the guards did not vary significantly over time, t (4) = – 0.12, ns, but that of the prisoners declined in a linear fashion, t (7) = – 4.48, p < .01. There was also evidence of a quadratic trend, t (7) = – 2.18, p < .07, suggesting that the prisoners’ decline in compliance was particularly marked after group boundaries had become impermeable.

Organizational citizenship (Organ, 1988, 1997) was measured on Days 2, 4, and 5 by means of a 3-item scale containing items used widely in previous research (e.g. Haslam, Powell, & Turner, 2000; Tyler & Blader, 2000; ‘I am willing to do more than is asked of me by the guards’/‘I will do whatever I can to help the guards’/‘Whenever possible I will try to make the guards’ work difficult’; reverse scored). A single score was computed by averaging responses to these 3 items.

The resultant mean scores are presented in Figure 5 as a function of participant group and time. As can be seen from this graph, the guards were always more willing than the prisoners to engage in citizenship behaviours that would help them run the regime. However, while the guards maintained their willingness to work for the regime across the three testing phases, over time the prisoners became much more reluctant to support the guards’ regime in this way. Again, this was particularly true after the promotion on Day 3.

□ Prisoners

□ Guards

Day 2                   Day 4                   Day 5

Study phase














CD [12]

The psychology of tyranny 19

These patterns were confirmed by statistical analysis of organizational citizenship scores performed by means of a 2 (participant group: guards, prisoners) X 3 (study phase: Days 2, 4, 5) ANOVA with repeated measures on the second factor. This analysis revealed a significant and large main effect for group, F(1,11) = 7.84,p < .01; h2 = .42, but this was qualified by a significant and large interaction between group and time; F(2, 22) = 5.10, p < .05; h2 = 32. Analyses of polynomial trends revealed that the guards’ willingness to display organizational citizenship did not vary across the three phases of testing, t(4) = 0.69, ns, but that of the prisoners declined linearly over time, t(7) = — 3.74, p < .01.

Collective self-efficacy and mental health

The disorganization of the guards and the unity of the prisoners did not just produce conditions where the latter challenged the former, it also led to the prisoners becoming progressively more dominant. Through planning and mutual support, they became increasingly more extreme and more successful in their efforts to undermine the guards’ control. By contrast, the guards’ attempts to impose their authority became increasingly unsuccessful, and they became more divided and mutually recriminatory. Moreover, as we have already illustrated with reference to observational data, the effectiveness of the prisoners in pursuing their collective goals led to strong positive affect while the inability of the guards to act collectively led to despondency.

These dynamics are reflected in quantitative data obtained on measures of (a) collective self-efficacy and (b) depression. Collective self-efficacy (Bandura, 1995) was measured on Days 2, 4, and 6 by means of a 5-item scale containing items used widely in previous research (e.g. Chen & Bliese, 2002; ‘My prison group is confident that we could deal efficiently with unexpected events’/‘My prison group can remain calm when facing difficulties because we can rely on our coping abilities’/‘My prison group can always manage to solve difficult problems if we try hard enough’/‘When my prison group is confronted with a problem, we can usually find several solutions’/‘My prison group can usually handle whatever comes our way’). A single score was computed by averaging responses to these 5 items.









Day 2

Day 4

Day 6

Study phase

Figure 6 presents the resultant mean collective self-efficacy scores as a function of participant group and time. As can be seen from this graph, these data mirror those obtained on measures of social identification, so that, at the start of the study, the guards


20 Stephen Reicher and S. Alexander Haslam

had a greater sense of collective self-efficacy than the prisoners. However, by Day 4, the prisoners’ self-efficacy had increased markedly and was now greater that that of the guards. This pattern was maintained at Day 6, although it is interesting to note that after the trade unionist was removed from the prison, both groups showed some decline in self-efficacy (seemingly because the opportunities for group-based negotiation and order that he presented were removed; see Eggins, Haslam, & Reynolds, 2002; Haslam, Eggins, & Reynolds, 2003). These patterns are consistent with our observational account — although the general decline in the guards’ sense of self-efficacy is less pronounced in these quantitative data.

Collective self-efficacy scores were statistically analysed by means of a 2 ( participant group: guards, prisoners) X 3 (study phase: Days 2, 4, 6) ANOVA with repeated measures on the second factor. This analysis revealed a significant main effect for time, F(2,22) = 6.08, p < .01; q2 = .36, that was conditioned by a significant and large interaction between group and time, F(2, 22) = 4.31,p < .05; q2 = .28. In line with the above observations, analyses of polynomial trends revealed that the prisoners’ sense of collective self-efficacy increased linearly over time, t(7) = 2.55, p < .05 — an effect that was also qualified by significant quadratic variation, t(7) = 4.25,p < .01. On the other hand, the collective self-efficacy of guards declined over time, although this effect was not significant, t(4) = – 0.90, ns.

Depression was measured during the screening process and then every day during the study (partly in order to monitor the ongoing welfare of participants). For this purpose, a 7-item scale was administered (‘In general, how has your mood been over the last few days?’/‘Do you ever feel low or depressed?’/‘Do you feel hopeless about the future?’/‘Do you have difficulty dealing with everyday problems?’/‘Are you self- confident?’/‘Do you think that you are a worthwhile person?’/‘Do you think about harming yourself?’). During the screening phase (N = 332), these items were found to form a reliable scale (a = .80) and so the items were averaged to form a single measure.

Mean depression scores are presented in Figure 7. From this graph, it can be seen that although overall levels of depression were low, they clearly varied as a function of participant group and time. Specifically, while the prisoners were more depressed than the guards at the start of the study, by its end, this situation had reversed. This pattern was confirmed by a 2 (participant group: guards, prisoners) X 7 (study phase: pre-test, Days 1 to 6) ANOVA with repeated measures on the second factor. This revealed a significant and large interaction between group and time; F(6,66) = 3 73, p < .01;


Study phase

The psychology of tyranny                                                                                                                 21

h2 = .25. Consistent with observations, polynomial contrasts showed that the prisoners’ depression decreased linearly over time, t (7) = – 4.22, p < .01. At the same time the guards’ depression increased, though not significantly, t (4) = 1.25, ns.

Combined impact

The combined outcome of these interrelated dynamics was that as the prisoners became increasingly aware of, and confident in, their collective identity, and as the guards’ confidence and collective self-efficacy declined, the prisoners developed and executed plans to take on and destroy the guards’ regime. The guards were divided, exhausted and demoralized. They were unable to organize themselves or the prisoners effectively. As a result, late on the evening of Day 6, the prisoners in Cell 2 broke out of their cell and occupied the guards’-quarters. At this point, the guards’ regime was seen by all to be unworkable and at an end (see Photograph 4).

Phase 2: Embracing inequality

After the collapse of the guard-prisoner hierarchy, all but two of the participants decided that they wanted the study to continue through the institution of a single selfgoverning commune. With FCp in the chair, they met with the experimenters and drew up the terms under which the commune would operate. Initially, the new system was highly effective. A number of participants who had been mutually hostile when they were divided into prisoners and guards formed strong and positive affective ties now they were recategorized part of a common group (Gaertner, Mann, Murrell, & Dovidio, 1989; Turner et al., 1987). Collectively, they performed their work tasks and chores with more effort and to a higher standard than at any other point in the study. However, those participants who had been centrally involved in challenging the old regime felt marginalized in this new system. At first, they simply failed to contribute to collective tasks. Later, they began to violate communal rules. Within a day, they were plotting to destroy the commune. The problems this created were exacerbated because the commune’s members had never developed procedures for dealing with dissidence, and hence they had no means of responding to threats to their social order.

By the morning of the commune’s second day, it was clear to many participants that the new social structure was in crisis. This situation was exacerbated when, by chance,

Photograph 4. Day 6: Cell 2 break out bringing the guards’ regime to an end.

22 Stephen Reicher and S. Alexander Haslam

the breakfast was of very poor quality. This was (incorrectly) taken as a sign that the experimenters disapproved of the commune system. It led to despondency among the commune’s supporters who felt, as one later put it, that they had the burden of responsibility for the system without the ability to make it work. Moreover, the emergent crisis was exploited by opponents of the commune, four of whom (one exguard and three ex-prisoners) formulated a plan to create a new and harsher guard- prisoner hierarchy. The nature and tone of this new regime was made abundantly clear in discussions about the form this would take. As PBp put it, ‘We want to be the guards and fucking make them toe the line, I mean on the fucking line. No fucking talking while you are eating. Get on with your food and get the fucking hell back to your cell’.

Shortly after breakfast, this group convened a meeting (see Photograph 5). Their leader berated the commune and its supporters and he introduced the idea of the new hierarchy. The supporters of the commune were largely passive in response. They looked despondent and listened in silence until he had nearly finished. During debriefings, a number of them acknowledged that, although they would not have openly endorsed such a hierarchy, they were less opposed to it than they had been previously and that they felt less repulsed by the idea of a strong social order in which someone else assumed responsibility for making the system work.

Again, these trends are supported by analysis of psychometric data. Most notably, this is apparent when one looks at the level of participants’ right-wing authoritarianism (Altmeyer, 1981, 1996). This was measured using 8 items abstracted from a 30-item scale administered during pre-testing (Reynolds, Turner, Haslam, & Ryan, 2001; ‘Things would go better if people talked less and worked harder’/‘It is better to live in a society in which the laws are vigorously enforced than to give people too much freedom’/‘People should always comply with the decisions of the majority’/‘You have to give up an idea when important people think otherwise’/‘There are two kinds of people: strong and weak’/‘What we need are strong leaders that the people can trust’/‘Our social problems would be solved if, in one way or another, we could get rid of weak and dishonest people’/‘People should always keep to the rules’). In pre-testing (N = 332), this scale was found to be reliable (a = .71) and to correlate very highly with the full 30-item scale (r = .97). Accordingly, scores on the above items were averaged to form a single measure. This was administered during the screening process and then on Days 1, 3 and 7.

Photograph 5. Day 8: New guards make the case for an authoritarian regime.


Day1             Day 3             Day 7

The psychology of tyranny                                                                                                                  23

Study phase

Figure 8. Right-wing authoritarianism as a function of assigned group and time.

In the first instance, it is relevant to examine levels of authoritarianism over time as a function of the groups to which participants were assigned by the experimenters. Data pertaining to this analysis are presented in Figure 8. These were analysed by means of a 2 (participant group: guards, prisoners) X 4 (study phase: pre-test, Days 1, 3, 7) ANOVA with repeated measures on the second factor. Consistent with qualitative observations, the only effect to emerge from this was a significant and large effect for time, F(3, 33) = 2.75, p < .05; h2 = .20. Analysis of polynomial trends revealed that there was a linear increase in all participants’ authoritarianism as the study progressed, t (12) = 2.61, p < .05.

Pre-test Day 1                      Day 3 Day 7

However, as a fascinating variant on the above analysis, it is also possible to look at authoritarianism as a function of the groups to which the remaining participants assigned themselves at the end of the study — that is, as participants who either supported the commune or who proposed setting up a new hierarchy (with themselves as the new guards within it). Data pertaining to this analysis are presented in Figure 9. These were again analysed by means of a 2 (participant group: new guards, new prisoners) X 4 (study phase: pre-test, Days 1, 3, 7) ANOVA with repeated measures on the second factor. The only effect to emerge from this analysis was a significant and large interaction between group and time; F(3, 36) = 3 07, p < .05; h2 = .20. Analysis of

Study phase


24 Stephen Reicher and S. Alexander Haslam

polynomial contrasts indicated that this interaction arose from the fact that the authoritarianism of those who sought to assume the guards’ role at the end of the study had declined non-significantly as the study progressed, t(3) = – 1.05, ns. However, the authoritarianism of the remaining participants had increased steadily over time, t(10) = 3.67, p < .01 (so that the two groups were demonstrating a very similar level of authoritarianism at the time that the new guards were actually seeking to take power).

Despite the general shift towards authoritarianism, and although it is probable that the new guards would have had the force to impose their regime in the face of weakening resistance, in this study such force was prohibited under the pre-established ethical guidelines. Hence, in our judgment, the study was gridlocked and had reached what we saw as a natural point of termination; the existing system was not working and the new system could not be imposed. Accordingly, the study was brought to a conclusion at noon on Day 8. However, the participants remained for a further day in order to undertake a series of structured debriefings designed to obtain and provide feedback on their experience, to explain the rationale for the study and to overcome any hostility between individuals deriving from events in the study.


The BBC prison study was designed to examine the factors that determine how people respond when a system of inequality is imposed upon them by others. At the start, almost all the participants rejected this system. However, by the end, they were close to instituting a new and more tyrannical social system. In addition to our original questions concerning the way in which people respond to a system of inequality that has been imposed upon them — do they accept it or do they resist it? — this raises a new and unexpected issue. What are the conditions under which people create a system of inequality for themselves?

We will argue that our findings concerning reactions to inequality cannot be explained through a general or ‘natural’ tendency to assume roles and assert power. Rather, they require an understanding of the conditions under which an externally imposed categorization becomes a subjective self-perception (Turner, 1982). The existing terms of social identity and self-categorization theories provide the basis for such an understanding, although it is necessary to elaborate upon notions of context in order to account for the failure of the guards to cohere as a group. Our findings concerning the creation of inequality lead us to more original theoretical conclusions. On the one hand, we suggest that groups are the basis for collective self-realization — that is, the creation of a social order based on shared values and norms (Drury & Reicher, 2005). However, where groups fail, we argue that people will be more inclined to accept the imposition of a social order by others, even where that violates their values and norms. Therefore, in contrast to those who explain tyranny and other extreme social phenomena in terms of the psychological dysfunctionality of groups, we interpret them in terms of the dysfunctionality of group failure.

Before we are in a position to develop these arguments, however, it is necessary to deal with four potential critiques of the study and its ability to say anything meaningful about the group processes relating to inequality and tyranny. The first is that the behaviour of participants was determined by the fact that they knew they were being observed by television cameras and that this renders the study so artificial as to have little or no general value. The second is that the effects we observed were a product of

The psychology of tyranny 25

participants’ personalities and that therefore the study has little to say about group processes. The third is that we failed to create real power differences or meaningful inequalities between groups and hence our study has little to say about the psychology of either power or inequality. The fourth is that the variables upon which our predictions (and hence our planned interventions) focused were not responsible for the effects obtained and, therefore, even if group and power processes were at play, we cannot be sure what they are.

Four critiques

The role of television

From their moment of first contact, volunteers knew that the BBC was involved in this project, and by the time that the study started, all participants were aware that they would be constantly observed by television cameras and that anything they did might be shown on national television. Although it was stressed throughout that this was a specialist scientific project that would not make ‘stars’ out of those involved, and although the screening process was used to exclude anyone who was motivated by the desire for publicity, this is a highly unusual situation and was bound to impact on behaviour — but what impact, and with what implications?

The most damning argument would be that participants were simply faking their behaviours for the cameras. We believe such an explanation to be implausible in light of the immense effort that would be required to continuously monitor and fake one’s own behaviour for nearly 9 days. Moreover, it would be much harder, if not impossible, to fake the psychometric and physiological data. However, even if participants were capable of such play-acting, one would still need to explain the complex pattern of results not only between groups, but also within groups across time. One might, for example, suggest that there is a certain discredit in being a tyrant and a certain glamour in being a rebel, which may explain why our guards were so mild and our prisoners so rebellious. But, why then did the prisoners become more rebellious after the system was made impermeable and insecure? Even more problematically, why did the participants move towards tyranny at the study’s end? To argue that our findings can be explained by suggesting that participants were merely play-acting or seeking ‘celebrity status’ is not only implausible but also unhelpful. In particular, it fails to explain (a) why people acted as they did when they did and (b) why their behaviour and attitudes clearly changed in particular (predicted) ways over the course of the study. By the same logic, surveillance cannot be the whole story, as it remained constant at the same time that behaviour itself was changing.

Having said this, we readily accept (see below) that the televising of the study is part of the story — particularly towards the start of the study (participants reported that they were acutely aware of the cameras on entering the ‘prison’, but that, as time went by, they increasingly forgot about them except during quiet moments; e.g. when seeing a camera move late at night in their cells). However, we suggest that the nature of this impact adds to, rather than detracts from, the richness and wider relevance of our findings. This is because although it may be unusual to be in a position where anything one does might be broadcast into millions of homes, this is an extreme example of something that is an increasingly common feature of our everyday lives — namely, surveillance. For most of our social lives, we are under observation and our behaviour can be examined by audiences who are not present (Reicher & Emler, 1985). Sometimes, the surveillance is a matter of surveillance cameras, scrutiny of computer

26 Stephen Reicher and S. Alexander Haslam

transactions, workplace review, or unobtrusive profiling (Lyon, 1994; Lyon & Zuriek, 1996). At other times, it is simply a matter of those who are with us in one place talking about our behaviour to other people in other places (e.g. in gossip, a universal human behaviour; Emler, 1994). However, whichever is the case, we can rarely ignore surveillance when we are alone and almost never when with others.

What the television cameras do, therefore, is to highlight in dramatic form an aspect of human experience, which is all too often overlooked in psychology. Indeed, it is arguable that the typical psychology experiment in which people are isolated and guaranteed absolute anonymity encounters greater problems of artificiality (Cronin & Reicher, 2006). Of course, the size and variety of audiences to which one is made accountable by television is much larger than we commonly experience. However, as we will discuss in more detail below, the fact of accountability, and the fact that our participants had to consider other contexts and other audiences even when acting within the context of the study, enhances rather than diminishes its relevance and meaningfulness. In this respect, we are reminded of Thayer and Saarni’s (1975) retort to those who impugn the validity of the SPE on the grounds that participants were guided by the expectations of others. This is summarized in the title of their paper: ‘demand characteristics are everywhere (anyway)’. Likewise, to those who criticize the BBC study on the grounds that it was broadcast, we would respond that surveillance is everywhere (anyway).

The role of personality

For those who have watched The Experiment (Koppel & Mirsky, 2002), it is hard not to be struck by the force of many of the characters who contributed to the various outcomes described above. In part, this has to do with the visual nature of the medium, where the concrete and visible impact of individuals is inevitably more salient than that of more intangible and ‘invisible’ elements such as group processes (Asch, 1952). Nonetheless, it could clearly be argued that all the observed effects were a product of strong (and weak) personalities and that the more powerful characters (in particular, JEp, PPp, and PBp) ended up as prisoners. One could argue that their personalities drove these individuals to overwhelm the guards and then, as they grew bored of the commune, to destroy it in turn.

The first response to such an argument is to remember that we matched the prisoners and guards on the obvious individual difference factors relating to tyranny: modern racism, authoritarianism, and social dominance. Nonetheless, one could still argue that there were other characteristics on which we did not match the groups. As a result, the ‘stronger’ individuals might still, by chance, have been allocated to the prisoner group. However, even if this were the case, as with our discussion of surveillance, is does not prove particularly helpful in explaining how events in the study unfolded. Most fundamentally, this due to the interrelated observations (a) that people’s ‘character’ on relevant dimensions appeared to change over the course of the study and (b) that relevant individual differences were much less apparent at the outset than they were as the study progressed.

We readily accept, however, that individual differences are part of the story (see Reicher & Haslam, in press; Turner, Reynolds, Haslam, & Venstra, in press). However, as with the analysis of leadership more generally (e.g. see Haslam, 2001; Haslam & Reicher, 2005; Reicher & Hopkins, 2001b; Turner & Haslam, 2001), the problem is that it is hard to explain changing patterns of behaviour in our study with reference to a blunt and constant factor such as ‘personality’. To claim that certain people were ‘rebellious’ does

The psychology of tyranny 1Л

not explain why rebellion was muted until after the promotion and fails to explain why those prisoners who rebelled against one form of authority (the guards) nonetheless remained deferential to another (the experimenters). In the case of JEp, for example, we see that before the promotion, he invested his energies in supporting the system and it was only after promotion was ruled out that he put them into undermining the system. Therefore, it is one thing to call him ‘forceful’, but to explain how that force was directed, we need to invoke systemic factors. Similarly, in the case of JE’s cellmate PPp, while he was opposed to the guards’ authority from the start of the study, his ability actually to undermine their regime was contingent on collective will and support, which only materialized after the promotion (in ways predicted by social identity theory; Tajfel & Turner, 1979; see below).

More generally, there were many points in the study where the prominence and impact of individuals depended upon their relationship to the group (Reicher & Haslam, in press). Thus, when they were seen to reflect shared identity and shared values, even the mildest of men came to the fore. This is apparent in the case of FCp, an environmentalist who had been quiet throughout the first 6 days of the study, but who became a leading and vocal figure in creating and managing the commune and without whose commitment, knowledge, and experience it may never have come into being.

Furthermore, it is also apparent that when there was no shared identity or shared values, even the most heroic of individuals was bound to fail. This was seen early in the study when FCg, made a strenuous effort to impose the guards’ authority and maintain strict discipline. However, as it became clear that he lacked the support of his fellow guards, he gave up, became increasingly passive, and withdrew into the background. Thus, the manifest differences in the forcefulness of prisoners and guards derived not from their inherent personalities but were instead an emergent product of the success or failure of their respective groups.

Again, this is not to dismiss the importance of individual variables. As we have repeatedly stressed, the skills, knowledge, values, and pre-existing commitments of individuals played a crucial part in facilitating, shaping (and indeed in blocking) group formation (Postmes, Baray, Haslam, & Morton, in press; Reicher, 2004; Reicher, Drury, Hopkins, & Stott, 2001; Reicher & Hopkins, 1996, 2001b, 2004). Far from being mindless, our participants were creative and thoughtful as group members, continuously striving to shape group and intergroup realities. However, if individuals played a key part in shaping groups, the converse is equally true — the ability of individuals to affect outcomes was dependent upon both the existence of groups and their prototypicality within the group (Turner, 1991). To put it slightly differently, individual agency was not destroyed by the group, but rather achieved through it (Reicher & Haslam, in press). Thus, while personality and other individual difference factors are an important aspect of our study (and we will consider them as part of our analysis below), they cannot substitute for that analysis (Asch, 1952; Brown, 1965; Sherif, 1966; Tajfel, 1978).

It follows that had different individuals participated in the study, we may well have observed quite different outcomes from those reported above. This is because many individual (and, indeed, chance) factors (such as the poor-quality breakfast on Day 8) impacted on the fate of groups and their subsequent trajectory in the study. However, as we explained in the Introduction, this is why we are not seeking to make empirical generalizations on the basis of findings alone but rather are attempting to make generalizations on the basis of the theoretical analysis that those findings support (Turner, 1981). Our claims to generality thus relate to (a) the effect of particular factors 28 Stephen Reicher and S. Alexander Haslam

(e.g. permeability, cognitive alternatives, roles) on group formation and (b) the consequences of group success and failure for both individuals and the system as a whole.

The reality of inequality and power

A third potential criticism of our study is that the results and, in particular, the failure of the guards to exert their authority (or the prisoners to accept it), reflects the twin facts that (a) the study context was simply too pallid for our participants to become engaged in it and (b) the guards had no authority to wield even if they wanted to. If, as Zimbardo and his colleagues argue, tyranny is a function of groups and power, then one might assert that our study lacked a meaningful degree of either and that we are therefore in no position to say anything worthwhile about tyranny.

Let us consider these issues of engagement and power in turn. In the case of the former, it is important to distinguish between engaging with the situation and engaging with one’s group. The evidence suggests that the prisoners experienced and resented their subordinate status from the start of the study. They disliked their food. They disliked being locked up. They disliked the boredom. The smokers in particular disliked being deprived of cigarettes. Indeed, after a day in his cell, PPp admitted that he was ‘falling for it [the experimental situation] hook, line and sinker’. At the first meal, several prisoners expressed outrage at the poor quality of the food, especially given the superior fare of the guards. The lack of group identification amongst prisoners in the early days cannot therefore be put down to apathy about their plight.

Equally, the guards engaged with the situation from the start. Their initial conversations focused on the power of the situation, the impact of wearing a guard’s uniform, and the dangers of becoming tyrannical. At their first meal, they also noted with discomfort the disparity in food quality and quantity. They then tried to ease this discomfort by offering their leftovers to some of the prisoners. Therefore, if the guards failed to act as a group, it was not because they did not care about the disparities in the study but, on the contrary, because they were troubled by them. Their disidentification did not reflect the fact that the study was not compelling, but rather that it was all too compelling (albeit in very different ways from the SPE).

The issues are similar when it comes to the matter of power. It is certainly true that the guards failed to exercise power. However, this was not because they had no power to exercise but precisely because they had so much. In discussion, the guards recognized the various options that were open to them. These included individual punishments, collective punishments, removal of privileges, extra tasks and roll calls and, most particularly, the power to promote a prisoner who would help them run the system in the way they wanted. However, their fear of the guard identity — of being authoritarian and of being seen as authoritarian — made them shun these options, to promote a prisoner who embodied their ambivalence (Herriot, 2002), and even to give away some of the sources of their power. In other words, the participants’ psychological state explains their failure to exercise power much more than their lack of power explains their psychological state (Turner, 2005). We will address the reasons for this state shortly. For now, we simply note that our participants did engage with a situation in which there were real inequalities of resources and power. Hence, the study can be viewed as a meaningful exploration of the psychology of unequal groups.

The impact of interventions and key variables

It may be that our study says something of interest, it may be that it says something interesting about group processes, it may even have something important to say about

The psychology of tyranny 29

tyranny and resistance, but what of our claim to have conducted an ‘experimental case study’ in which we investigated the impact of specific theoretically informed variables? Can our interventions really be described as operationalizations of permeability and cognitive alternatives? Are there not a host of other ways in which these interventions and their consequences can be interpreted, and does this fact not cast doubt on any particular theoretical gloss we might provide?

This is a reasonable concern. In the present study, there is certainly a disjunction between the simplicity of our variables and the complexity of our interventions. In particular, the promotion and the introduction of DMp could have had many effects, such as making participants uncertain about what might happen next and making them feel helpless and distrustful of the experimenters. However, such problems are not unique to the present study — not least because the theoretical status of independent variables is a potential concern in all experimental research (Haslam & McGarty, 2004). Indeed, whenever researchers claim that a concrete instance (a specific operationalization of an independent variable) reflects an abstract generality (the variable as it operates in the world at large), interpretation is involved and alternatives are possible (Billig, 1987). When the operationalization is as complex as ours, however, the space for argument is all the greater.

In experimental practice, this issue is normally addressed in two ways (Haslam & McGarty, 2003). The first is through the use of manipulation checks. As we have seen above (Figure 3), checks of this form provide some evidence consistent with our theoretical interpretation (specifically in relation to the existence of cognitive alternatives). Moreover, one of the advantages of having multiple data sources (i.e. behavioural as well as psychometric) is that we are able to see whether participants spontaneously reacted to our interventions in ways that were expected. We have provided qualitative evidence to suggest that they did. Specifically, after the promotion prisoners saw no point in placating the guards and realized that the only way to improve their position was to challenge the system (Extract 1). Equally, we saw that DMp transformed the way in which participants conceptualized their situation and their options (Extract 2). As with any manipulation check, such data support the notion that we did manipulate the variables of concern to us.

However, while manipulation checks exclude false positives, they cannot exclude all possible false negatives — that is, additional ways in which the manipulation may have had an impact. This is where the second way of reducing uncertainty about the status of independent variables comes into play. Here, it is not enough merely to argue for the importance of a particular independent variable (or a particular confound; e.g. personality, surveillance, the passing of time). It is also necessary to explain how that variable (or confound) might plausibly and comprehensively account for the patterns of observed results. That is, it is necessarily to have an integrated theoretical account of how the effects occurred. As we explained when introducing them, in this respect, our claims accord with a well-developed and extremely influential theoretical tradition (after Tajfel & Turner, 1979) and with a well-established and robust pattern of previous experimental findings (e.g. Ellemers, 1993; Wright et al., 1990). This theoretical and empirical tradition lends coherence and plausibility to our analysis.

Having said that, we certainly do not view our analysis as inviolate, and we fully accept that our account of the impact of particular interventions is open to argument. However, for critics who wish to challenge this account, it is incumbent upon them to do as we have and explain how an alternative understanding of our interventions provides a better understanding of the study’s findings in their entirety.

30 Stephen Reicher and S. Alexander Haslam

With this in mind, it is time to move beyond potential criticisms and to be explicit about what our explanation of events is. This involves three elements. The first relates to the conditions of social identification, the second addresses the consequences of social identification, and the third concerns reactions to group failure.

A social identity account of tyranny

The conditions of social identification

The simplest and clearest finding of our study is that people do not automatically assume roles that are given to them in the manner suggested by the role account that is typically used to explain events in the SPE (Haney et al., 1973; Zimbardo, 1989). Instead, there are a range of factors that determine whether people themselves identify with the social positions to which they are ascribed by others. Some of those factors operated in ways specified by social identity theory (Tajfel & Turner, 1979). Thus, the shift from permeability to impermeability of group boundaries had a strong impact on prisoner identification (along lines previously demonstrated in separate programmes of research by Ellemers and Wright and their colleagues; Ellemers, van Knippenberg, De Vries, & Wilke, 1988; Ellemers et al., 1990, 1993; Wright, 1997; Wright & Taylor, 1998, Wright et al., 1990; see also Lalonde & Silverman, 1994). Equally, the increasing insecurity of intergroup relations affected the willingness of the prisoners to work as a group in order to challenge the guards.

In addition, changing perceptions of security, and of the component of legitimacy in particular, help explain not only why the prisoners challenged the guards, but also why the participants as a whole did not challenge the experimenters, despite the adverse conditions that living in the prison imposed upon them (cf. Spears & Smith, 2001). Most of the time, when participants discussed the matter, they saw these conditions as a legitimate part of the study in which they had agreed to participate. On only two occasions were the conditions seen to violate that contractual legitimation — when the new prisoner, DMp, began to question the heat in the prison, and when one night, the experimenters failed to replenish basic supplies for the guards. On both occasions, this led to explicit consideration of ‘mutiny’.

However, if the study supported social identity theory in demonstrating the role of contextual factors in moderating the relationship between role and identity, it also shows the importance of extending the way in which we conceptualize the nature of context and its relation to human action. Thus, although the role of guard was positively valued in the immediate context of the prison, those assigned to this position were concerned with the possibility of negative evaluations by future audiences, and hence some of them were reluctant to identify with the role.

It is here that the prospective televising of the study most clearly had an impact on the behaviour of participants and serves to raise important practical and theoretical points. Practically, it demonstrates how extreme behaviours can be restrained by rendering actors visible, and hence accountable, to broader or yet-to-be encountered audiences (Postmes, Spears, Lea, & Reicher, 2000; Reicher & Levine, 1994; Tedeschi, 1981; Tetlock, 1985).[13] Theoretically, it shows that context, for human beings, cannot

The psychology of tyranny                                                                                                                31

simply be assumed to be where we are in the here and now. People have imaginations that allow them to think of other times and places, and to evaluate their behaviour in relation to these. Indeed, surveillance has an impact on behaviour precisely through its capacity to make salient such alternative contexts and alternative audiences (Reicher & Haslam, 2003; Reicher & Levine, 1994).

Two points follow from this. First, under conditions of surveillance, individual differences relating to the different social commitments people have outside a given context have the capacity to impact upon behaviour within the context. Thus, in our study, the fact that TQg was publicly visible as a successful entrepreneur who owned a liberal enterprise goes some way towards explaining why he was particularly ambivalent about adopting the identity of guard. However, if individual factors were important here, then they operated in a paradoxical way. First, TQg’s very effectiveness outside the study is what rendered him ineffective within it. Second, once human imaginative capacities are acknowledged, forms of extreme situationism (such as the role account), which suggest that behaviour is always dominated by the present context, become untenable. Instead, it is important to take past and future contexts into account as well (Reicher & Hopkins, 2001a).

The consequences of social identification

One of the distinctive features of the present study was that its intensive nature allowed established disciplinary boundaries to be transcended so that it was possible to investigate the interrelationships between social, organizational and clinical variables. The results presented above point to the existence of such links and to their richness and complexity.

On the organizational side, the achievement of a common social identity was seen to be necessary in order for group members to trust others to act appropriately, to support others in their actions and to expect support from them in return (Haslam et al., 2005). In this way, the rudiments of effective organization (e.g. task differentiation, delegation, leadership, trust) were seen to be contingent upon shared social identification in a manner argued by a range of organizational theorists (e.g. Ellemers, De Gilder, & Haslam, 2004; Haslam, 2001; Haslam, van Knippenberg, Platow, & Ellemers, 2003; Hogg & Terry, 2001; Postmes, 2003; Smith, Tyler, & Huo, 2003; Turner & Haslam, 2001; Tyler & Blader, 2000). For this reason, the effectiveness of the prisoners, who were able to develop a sense of shared social identity, contrasted markedly with the ineffectiveness of the guards, who were not (Drury & Reicher, 2005).

On the clinical side, the increasing support amongst prisoners and their successful challenges to the guards contributed to mental states that were increasingly positive. On the other hand, the atomization of the guards and their failure to fulfil their collective tasks led to increasingly negative states. This is reflected in the depression data (and also on other measures not reported here including paranoia, anxiety and burnout; see Haslam & Reicher, in press). Although the nature of our quantitative data does not allow for analysis of causal relations between these variables (Haslam & McGarty, 2004), we can say with confidence that participants’ mental states evolved in relation to the social dynamics between groups, and that variables such as social support and collective selfefficacy are implicated in that relationship (e.g. as argued by Branscombe, Schmitt, & Harvey, 1999; Cohen & Wills, 1985; Contrada & Ashmore, 2000; Hall &Cheston, 2002; Haslam et al., 2005; Orford, 1992; Schwarzer, 2001; Terry, Carey, & Callan, 2001). Clearly though, the exact nature of the relationship merits further investigation.

32 Stephen Reicher and S. Alexander Haslam Reactions to group failure

There were two instances of group failure in the study. The first concerned the failure of the guards to fulfil their assigned task of making a hierarchical prison system work, and the second concerned the failure of the commune’s supporters to establish a viable egalitarian social order.

It is notable that in each case, group members became more willing to accept a system that promised to be viable even if it meant ceding some of the core principles of the group. On Day 5, after the introduction of the new prisoner, the guards were willing to cede their authority and accept a more equal social system and on Day 8 commune supporters were prepared to cede equality and tolerate a more hierarchical social system. Thus, rather than people ‘naturally’ preferring any given form of social order, it appears that, when group members fail to impose an order based on their own existing norms and values, they are willing to adapt those values (or to adopt new ones) in order to create a viable order rather than have no order at all.

In this regard, the fact that the participants’ authoritarianism increased significantly over time constitutes one of the most important findings of the study as a whole. Traditionally, authoritarianism has been viewed a stable individual difference variable that has the capacity to explain the emergence of hierarchical and tyrannical social structures (Adorno et al., 1950; Altmeyer, 1981, 1996; for reviews see Billig, 1978; Brown, 1965; Duckitt, 1994). However, in contrast to this analysis, the present study illustrates that authoritarianism is a variable outcome of social structure (see also Oakes, Haslam, & Turner, 1994; Reynolds et al., 2001). More specifically, we see that authoritarian solutions — and the personalities that would promote them — appeared more attractive after attempts to make democracy work were seen to have failed.

Against this argument, it might be suggested that the system we observed moved towards tyranny simply because the more authoritarian individuals were waiting for an opportunity to impose their preferred system on others (indeed, a similar point could be made in relation to the SPE). However, along lines intimated above, two points speak against this conclusion and against related arguments couched in terms of individual personality-based dynamics. The first is that tyranny arose out of the social (dis)order of the prison, and authoritarianism (like other dimensions of raw personality) was clearly not a straightforward determinant of that (dis)order. The second is that although it may be true that those who proposed the new regime were those who had initially been more disposed towards authoritarianism, their influence and leadership was clearly contingent upon their views having become more representative of the views of the participants as a whole as these had evolved over the course of the study (for more in-depth treatment of this issue see Haslam & Reicher, 2005; Reicher & Haslam, in press; see also Haslam, 2001; Haslam & Platow, 2001; Navas, Morales, & Moya, 1992; Reicher & Hopkins, 1996, 2001b; Turner, 1991; Turner & Haslam, 2001; Turner et al., in press). Indeed, the interaction between final group membership and study phase suggests that, if anything, the authoritarianism of the guards in the vaunted new regime had declined over time, whereas that of other participants was increasing (see Figure 9). It also is worth adding that previous research has been unable to observe dynamics of this form at work or to generate empirical support for arguments such as these, precisely because it has been unable to examine the impact of evolving group history on both individual psychology and social structure (Haslam & McGarty, 2001; Levine, 2003).

The psychology of tyranny 33

Conclusion: Rethinking the relationship between groups, power and tyranny

At one level, our study confirms the findings of the SPE. It shows that an understanding of collective conflict and tyranny cannot be achieved simply by looking at individuals but requires an analysis of group processes and intergroup relations. In this sense, we agree with Zimbardo (and many others; e.g. Asch, 1952; Sherif, 1966; Tajfel, 1978) that such phenomena can only be explained through group-level analysis. Our disagreement with prior analysis of the SPE thus relates to the nature of group processes and of the conditions under which they lead to social pathologies.

As almost every psychology student (and an unusually large proportion of the general public) knows, the message of the SPE is that the toxic combination of groups and power leads to tyranny. The implications of the BBC prison study are different. In common with recent theoretical developments in social psychology, they contest the premise that group behaviour is necessarily uncontrolled, mindless and antisocial (Ellemers et al., 1999; Oakes et al., 1994; Postmes et al., 2000; Reicher, 1982, 2001; Spears, Oakes, Ellemers, & Haslam, 1997; Turner, 1999). In contrast, the results of the BBC prison study suggest that the way in which members of strong groups behave depends upon the norms and values associated with their specific social identity and may be either anti- or prosocial (Jetten, Spears, & Manstead, 1997).

However, based on the present data, we would argue that failing groups almost inevitably create a host of problems for their own members and for others. These problems have a deleterious impact on organization, on individuals’ clinical state, and — most relevant here — on society. For it is when people cannot create a social system for themselves that they will more readily accept extreme solutions proposed by others. It is when groups lack the power to exercise choice that an authoritarian ideology that promises to create order for them appears more seductive. In short, it is the breakdown of groups and powerlessness that creates the conditions under which tyranny can triumph (for related arguments see Kanter, 1979; Pfeffer, 1981; Reynolds & Platow, 2003)

We would argue that as well as being consistent with contemporary thinking in social psychology (e.g. after Tajfel & Turner, 1979), this analysis also articulates more closely than Zimbardo’s original role account with the analysis of tyranny put forward by researchers in other academic disciplines. Most notably, it accords with influential analyses proposed by modern historians (e.g. Abel, 1986; Gellately, 2001; Hobsbawm, 1995; Rees, 2002). Consider, for instance, Hobsbawm’s account of the conditions that gave rise to the fall of the Weimar republic and the emergence of Nazism in 1930s Germany:

The optimal conditions for the triumph of the ultra-right were an old state and its ruling mechanisms which could no longer function; a mass of disenchanted, disoriented and disorganized citizens who no longer knew where their loyalties lay; strong socialist movements threatening or appearing to threaten social revolution, but not actually in a position to achieve it…. These were the conditions that turned movements of the radical right into powerful, organized and sometimes uniformed and paramilitary force (1995, p. 127).

We would also argue that this analysis can be used to make sense of what happened in the SPE when one looks more closely at the events that unfolded there. As we have noted, that study, like ours, appears to have started off with prisoners threatening to become ascendant over the guards. Things changed when Zimbardo intervened in such a way as to lead the prisoners to believe that they could not leave the study. At this point they became disoriented as to their position — in Zimbardo’s own words, they 34 Stephen Reicher and S. Alexander Haslam

experienced ‘role confusion’ (Zimbardo, 1989). They ceased to support each other against the guards. They collapsed as a group and this allowed tyrannical guards to prevail. Thus, whereas in the SPE, the failure of the prisoners allowed an existing tyranny to be consolidated, in the BBC study, the failure of the commune paved the way for the emergence of a new tyranny.

These, of course, are large and controversial claims. However, they have important theoretical implications along with considerable practical implications. They point to new ways of thinking both about the nature of group psychology and about the psychological underpinnings of tyranny. Because the scope of these analyses is so large, it would be both presumptuous and foolish to expect them to be accepted on the basis of a single data set — especially one reflecting the complexities of the present study. Yet what is true of our study is largely true of the SPE. As we have argued, one of the great problems arising from that study was that it made strong claims (that had an enormous impact on public consciousness) but then led to debate being closed off because further research was declared to be ethically unacceptable.

Hence, irrespective of our results and of our analysis, we would claim that one of the significant achievements of the BBC prison study is to show that, if sufficient care is taken, it is possible to run powerful and influential field studies into social processes that are also ethical. In a field increasingly dominated by reductionist accounts of human behaviour, such studies can restore balance by demonstrating the impact of systematic variations in social relations upon human behaviour. The richness, immediacy and relevance of their findings can also help reconnect psychologists with policy makers. For these reasons, we offer our conclusions not so much as a final word on the matter, but more in the hope of stimulating researchers to renew their interest in the important debates that our discipline can, and must, advance.

Consensus Statement on the Stanford Prison Experiment and BBC Prison Study

Professors Craig Haney, Alex Haslam, Stephen Reicher, and Philip Zimbardo

We, the undersigned researchers who conducted the Stanford Prison Experiment (Philip Zimbardo and Craig Haney) and BBC Prison Study (Alex Haslam and Stephen Reicher), recognize that our studies, results, and public statements have engendered strong debate and, at times, misunderstanding within and beyond psychology. In an effort to promote constructive scientific dialogue, we are therefore releasing this consensus statement to highlight common ground and clarify our views on the research in question.

First, we jointly believe that it is critically important to develop a scientific understanding of toxic human behavior, including brutality and the abuse of authority and power.

Second, we regard the Stanford Prison Experiment and BBC Prison Study as valid studies and valuable resources for advancing such understanding. At the same time, we recognize that both investigations have methodological limitations and are best viewed as one-trial demonstration studies rather than traditional experiments.

Third, the BBC Prison Study differs from the Stanford Prison Experiment in essential ways and is not a direct replication of the earlier study. We do, however, see value in comparing the different outcomes of the two studies (and others) as a means of advancing conceptual understanding.

Fourth, the behaviors observed in the Stanford Prison Experiment and BBC Prison Study were a function of many factors, including roles, norms, leadership, social identification, group pressure, and individual differences, not all of which are necessarily mutually exclusive. It is only natural that explanations of social behavior will be complex and multifactorial.

Fifth, we believe in open science and welcome the public release of information that aids in the interpretation of these and other studies.

Sixth, we encourage others to investigate, discuss, and teach about the roots of toxic behavior and effective ways to prevent it.

Seventh, we regret instances in which our statements appeared to involve ad hominem criticisms or used intemperate language. Although it is legitimate to debate the accuracy, comprehensiveness, and meaning of research reports, we have no definitive evidence that any signatory of this statement committed scientific fraud or deliberately misled others about their research findings.

Eighth, we hope that future discussions and debates about our research and other studies of toxic behavior will be open, collegial, and respectful of differing points of view.

Craig Haney

Distinguished Professor of Psychology University of California, Santa Cruz

Alexander Haslam

Professor of Social and Organizational Psychology University of Queensland

Stephen Reicher

Wardlaw Professor of Psychology University of St. Andrews

Philip Zimbardo

Professor Emeritus of Psychology Stanford University

August 27, 2018


Statement by Craig Haney, Ph.D., J.D.

Distinguished Professor of Psychology and University of California Presidential Chair, 2015-2018

I’m dismayed to see the recent criticism of the Stanford Prison Experiment. I am not sure there is a single study in all of psychology about which more has been written over the years. Unfortunately, in part because of what appear to be pre-existing biases by the authors and the ad hominem nature of their criticisms, this new writing adds little if anything of value to that larger literature. The criticisms themselves contain a number of inaccuracies, misrepresentations, and misinterpretations. I think Phil Zimbardo has responded appropriately, thoughtfully, candidly, and persuasively to them and I have little to add, except to clarify some issues that I am in a special position to address.

The written descriptions that Phil, Curt Banks, and I provided of what went on in the study were accurate reflections of what we saw and experienced. We had the benefit of just having witnessed the events that transpired in the study, in the context in which they occurred, and had each other and our notes to rely on. Our goal was to faithfully report and validly interpret what had happened and why. I stand by the things I have written about the study, including the things that Phil Zimbardo and I have written about it together. I was the senior graduate student on the project, involved in the initial conceptualization of the study, all aspects of its planning, and, except for the last day when I was called to the East Coast for a family emergency, I was on site as much or more as anyone on the research team. Among other things, I can attest to the fact that the guards were not «just following orders» but rather were responding to the situational forces, role demands, and contextual cues that were built into the prison-like environment that had been created. In fact, I also can attest to the fact that, at least in my case, they certainly were not following any orders to be harsh and tough; some of them actually pushed back against me when I instructed them to treat the prisoners more humanely (even though they obviously knew I was one of the researchers).

In addition, I was the last person on the research team who saw the first prisoner who broke down, just 36 hours into the study. (I have refrained over the years from using his name in print, out of respect for his privacy; he has publicly identified himself, but I will continue to refer to him as «Prisoner 8612.») Interestingly, although it is common knowledge that I was the researcher who released Prisoner 8612, and therefore was the only percipient witness to his emotional state at the time of his release, I was never directly interviewed by any of the recent critics of the study about what actually happened. Prisoner 8612 was in genuine emotional distress when I saw him on the second night of the study–at times he was shaking, screaming, and crying. He seemed out of control, helpless, and despondent. I did not believe then and I do not believe now that he was “faking.” I treated him respectfully and compassionately, took him to a room outside the “prison yard” to talk with him, and gave him an opportunity to rest and hopefully calm down, while I tried to reach Phil Zimbardo. The possibility of a prisoner actually breaking down was not something that we had anticipated or planned for and we had no protocol for handling it. This was a grievous error on our part, one of several that we have candidly acknowledged (most stemming from the fact that, as researchers, we had placed ourselves in the midst of an environment whose power we underestimated). In any event, when I could not contact Phil, I returned to Prisoner 8612 and asked him if he wanted to stay in the study. It was clear to me that he was still very upset and, when he said he wanted to go home, I made the decision to release him and did. What he says about those events now, and why, is his business. But I accurately described what transpired and always have.

Notwithstanding the misinterpretations others have given it, the Stanford Prison Experiment stands as a dramatic demonstration of several fundamental psychological truths. The truths include the fact that a potentially very destructive dynamic is created when one group of people is given near-total power over another group of derogated others. Especially when it occurs in an otherwise powerful, dehumanizing environment such as a prison or prison-like environment, that dynamic can become punitive and abusive relatively quickly, lead people to engage in harsh and even destructive behavior from which they might otherwise have refrained, and facilitate their becoming inured to the suffering of others who bear the brunt of their mistreatment. It also can instill deep feelings of helplessness, degradation, despondency, and despair in the persons who are on the other side of this equation.

Of course, the Stanford Prison Experiment alone did not “prove” these fundamental truths. Rather it has endured over the years as a vivid illustration of them–in part because we selected the participants in the study on the basis of their psychological health and conventional “normality” (so that whatever happened could not be attributed to their alleged underlying “pathological” traits or dispositions), in part because they were all randomly assigned to their respective roles (underscoring the arbitrariness of the power dynamic that was created), and in part because of how quickly the situation devolved into one of extreme mistreatment. But the truths themselves have been repeatedly empirically documented in a wide range of other studies in social psychology and across various social scientific disciplines more generally. Other researchers–Asch, Bandura, Goffman, Milgram, Mischel, Rosenhan, and Ross among others–have provided similarly astute empirical observations and theoretical explanations about the power of certain kinds of situations to significantly modify and transform human behavior.

For me, these truths are more than textbook statements or laboratory findings. Unlike the recent critics, I have spent many thousands of hours inside actual prisons, all too often documenting the operation of these very dynamics at work and assessing the toll that they take on prisoners (and, although it is much more difficult to study, on correctional staff members as well). By almost any measure, the conditions created in the “Stanford County Prison” represented a pallid, almost benign representation of the ones that exist inside actual prisons and jails. Nonetheless, I have seen the basic lessons of the Stanford Prison Experiment play out again and again–not always, and not in exactly the same way–but often enough to confirm the wisdom and real world application of those fundamental truths in actual correctional facilities and, by extension, in institutional settings that resemble them. Over the years, I also have had numerous correctional officials and line staff acknowledge to me the validity of the insights that were generated by the Stanford Prison Experiment and the importance of trying to resist the institutional pressures to succumb (whether they were successful in doing so or not).

In the writing I have done about the Stanford Prison Experiment, including writing co-authored with Phil Zimbardo, neither he nor I have ever suggested that these fundamental truths should be used as blanket justifications or excuses for the mistreatment of others, as these recent critics suggest. Rather quite the opposite. Of course, if it can be shown that powerful situational forces have played a significant role in causing or influencing someone s behavior, then those forces should be taken into account (as the law, in its wisdom, provides). However, in my own professional work, I have spent decades urging prison systems to fundamentally “change the situation” in order to minimize if not eliminate abuse and mistreatment and to alleviate the suffering that occurs inside as a result. Toward that end, I have repeatedly recommended that prisons and jails introduce and implement greater levels of accountability inside correctional environments themselves and make these settings more accessible and transparent. This is the only way that outside ethical and legal norms and standards of humane treatment can be effectively brought to bear and that meaningful oversight and intervention can be introduced and maintained in places where they are often lacking.

To reduce the message of the Stanford Prison Experiment to the “suggestion that all it takes to make us enthusiastic sadists is a jumpsuit, a billy club, and the green light to dominate our fellow human beings,” as one of the recent critics has said, is a profound misinterpretation of the study, one that distorts and trivializes its results. It also does a disservice to this and similar studies that underscore the way that powerful, dehumanizing situations can and often do negatively shape and transform human behavior. This is a difficult, uncomfortable message for some to accept, with significant implications for social, legal, and correctional change. I believe this is why, at least in some quarters, the study continues to be fiercely debated after all these years. But the need or desire to reject the message does not justify these motivated attacks on the messengers.

June 25, 2018


Interpersonal Dynamics in a Simulated Prison

CRAIG HANEY, CURTIS BANKS AND PHILIP ZIMBARDO Department of Psychology, Stanford University,

California 94305, U.S.A.

Interpersonal dynamics in a prison environment were studied experimentally by designing a functional simulation of a prison in which subjects role-played prisoners and guards for an extended period of time. To assess the power of the social forces on the emergent behaviour in this situation, alternative explanations in terms of pre-existing dispositions were eliminated through subject selection. A homogeneous, “normal” sample was chosen after extensive interviewing and diagnostic testing of a large group of volunteer male college students. Half of the subjects were randomly assigned to role-play prison guards for eight hours each day, while the others role-played prisoners incarcerated for nearly one full week. Neither group received any specific training in these roles.

Continuous, direct observation of behavioural interactions was supplemented by video-taped recording, questionnaires, self-report scales and interviews.. All these data sources converge on the conclusion that this simulated prison developed into a psychologically compelling prison environment. As such, it elicited unexpectedly intense, realistic and often pathological reactions from many of the participants. The prisoners experienced a loss of personal identity*1 and the arbitrary control of their behaviour which resulted in a syndrome of passivity, dependency, depression and helplessness. In contrast, the guards (with rare exceptions) experienced a marked gain in social power, sta,tus and group identification which made role-playing rewarding.

The most dramatic.pf the coping behaviour utilised by half of the prisoners in adapting to this stressful situation was the development of acute emotional disturbance–severe enough to warrant their early release. At least a third of the guards were judged to have become far more aggressive and dehumanising toward the prisoners than would ordinarily be predicted in a simulation study. Only a very few of the observed reactions to this experience of imprisonment could be attributed to personality trait differences which existed before the subjects began to play their assigned roles.


After he had spent four years in a Siberian prison the great Russian novelist Dostoevsky commented, surprisingly, that his time in prison had created in him a deep optimism about the ultimate future of mankind because, as he put it, if man could survive the horrors of prison life he must surely be a «creature who could withstand anything”. The cruel irony which Dostoevsky overlooked is that the reality of prison bears witness not only to the resilience and adaptiveness of the men who tolerate life within its walls, but as well to the “ingenuity” and tenacity of those who devised and still maintain our correctional and reformatory systems.

Nevertheless, in the century which has passed since Dostoevsky’s imprisonment, little has changed to render the main thrust of his statement less relevant.

Although we have passed through periods of enlightened humanitarian reform, in which physical conditions within prisons have improved somewhat and the rhetoric of rehabilitation has replaced the language of punitive incarceration, the social institution of prison has continued to fail. On purely pragmatic grounds, there is substantial evidence that prisons in fact neither «rehabilitate” nor act as a deterrent to future crime–in America, recidivism rates upwards of 75% speak quite decisively to these criteria. And, to perpetuate what is additionally an economic failure, American taxpayers alone must provide an expenditure for “corrections” of 1.5 billion dollars annually. On humanitarian grounds as well, prisons have failed: our mass media are increasingly filled with accounts of atrocities committed daily, man against man, in reaction to the penal system or in the name of it. The experience of prison undeniably creates, almost to the

point of cliche, an intense hatred and disrespect in most inmates for the                         ;;

authority and the established order of society into which they will eventually        .

return. And the toll which it takes on the deterioration of human spirit for those    . …. ..

who must administer it, as well as for those upon whom it is inflicted, is incalculable.

Attempts to provide an explanation of the deplorable condition of our penal system and its dehumanising effects upon prisoners and guards, often focus upon what might be called the dispositional hypothesis. While this explanation is rarely expressed explicitely, it is central to a prevalent non-conscious ideology:

that the state of the social institution of prison is due to the “nature” of the                                                                           ;i;

people who administer it, or the “nature” of the people who populate it, or

both. That is, a major contributing cause to despicable conditions, violence, ‘•                                                                  : Щ

brutality, dehumanisation and degradation existing within any prison can be                                                                        T

traced to some innate or acquired characteristic of the correctional and inmate

population. Thus on the one hand, there is the contention that violence and

brutality exist within prison because guards are sadistic, uneducated, and                                   ;

insensitive people. It is the “guard mentality”, a unique syndrome of negative

traits which they bring into the situation, that engenders the inhumane

treatment of prisoners. Or, from other quarters comes the argument that

violence and brutality in prison are the logical and predictable result of the

involuntary confinement of a collective of individuals whose life histories are, by definition, characterised by disregard for law, order and social convention and a concurrent propensity for impulsiveness and aggression. Logically, it follows that these individuals, having proved themselves incapable of functioning satisfactorily within the “normal” structure of society, cannot do so either inside the structure provided by prisons. To control such men as these, the argument continues, whose basic orientation to any conflict situation is to react with physical power or deception, force must be met with force, and a certain number of violent encounters must be expected and tolerated by the public.

The dispositional hypothesis has been embraced by the proponents of the prison status quo (blaming conditions on the evil in the prisoners), as well as by its critics (attributing the evil to guards and staff with their evil motives and deficient personality structures). The appealing simplicity of this proposition localises the source of prison riots, recidivism and corruption in these “bad seeds” and not in the conditions of the “prison soil». Such an analysis directs attention away from the complex matrix of social, economic and political forces which combine to make prisons what they are–and which would require complex, expensive, revolutionary solutions to bring about any meaningful change. Instead, rioting prisoners are identified, punished, transferred to maximum security institutions or shot, outside agitators sought and corrupt officials suspended–while the system itself goes on essentially unchanged, its basic structure unexamined and unchallenged.

However, a critical evaluation of the dispositional hypothesis cannot be made directly through observation in existing prison settings, since such naturalistic observation necessarily confounds the acute effects of the environment with the chronic characteristics of the inmate and guard populations. To separate the effects of the prison environment per se from those attributable to a priori dispositions of its inhabitants requires a research strategy in which a “new” prison is constructed, comparable in its fundamental social-psychological milieu to existing prison systems, but entirely populated by individuals who are undifferentiated in all essential dimensions from the rest of society.

Such was the approach taken in the present empirical study, namely, to create a prison-like situation in which the guards and inmates were initially comparable and characterised as being “normal-average”, and then to observe the patterns of behaviour which resulted, as well as the cognitive, emotional and attitudinal reactions which emerged. Thus, we began our experiment with a sample of individuals who did not deviate from the normal range of the general population on a variety of dimensions we were able to measure. Half were randomly assigned to the role of “prisoner”, the others to that of “guard”, neither group having any history of crime, emotional disability, physical handicap nor even intellectual or social disadvantage.

The environment created was that of a “mock” prison which physically constrained the prisoners in barred cells and psychologically conveyed the sense of imprisonment to all participants. Our intention was not to create a literal


simulation of an American prison, but rather a functional representation of one. For ethical, moral and pragmatic reasons we could not detain our subjects for extended or indefinite periods of time, we could not exercise the threat and promise of severe physical punishment, we could not allow homosexual or racist practices to flourish, nor could we duplicate certain other specific aspects of prison life. Nevertheless, we believed that we could create a situation with sufficient mundane realism to allow the role-playing participants to go beyond the superficial demands of their assignment into the deep structure of the characters they represented. To do so, we established functional equivalents for the activities and experiences of actual prison life which were expected to produce qualitatively similar psychological reactions in our subjects–feelings of power and powerlessness, of control and oppression, of satisfaction and frustration, of arbitrary rule and resistance to authority, of status and anonymity, of machismo and emasculation. In the conventional terminology of experimental social psychology, we first identified a number of relevant conceptual variables through analysis of existing prison situations, then designed a setting in which these variables were made operational. No specific hypotheses were advanced other than the general one that assignment to the treatment of “guard” or “prisoner” would result in significantly different reactions on behavioural measures of interaction, emotional measures of mood state and pathology, attitudes toward self, as well as other indices of coping and adaptation to this novel situation. What follows is the mechanics of how we created and peopled our prison, what we observed, what our subjects reported, and finally, what we can conclude about the nature of the prison environment and the experience of imprisonment which can account for the failure of our prisons.



The effects of playing the role of “guard” or “prisoner” were studied in the context of an experimental simulation of a prison environment. The research design was a relatively simple one, involving as it did only a single treatment variable, the random assignment to either a “guard” or “prisoner” condition. These roles were enacted over an extended period of time (nearly one week) within an environment which was physically constructed to resemble a prison. Central to the methodology of creating and maintaining a psychological state of imprisonment was the functional simulation of significant properties of “real prison life” (established through information from former inmates, correctional personnel and texts).

The “guards” were free with certain limits to implement the procedures of induction into the prison setting and maintenance of custodial retention of the “prisoners”. These inmates, having voluntarily submitted to the conditions of this total institution in which they now lived, coped in various ways with its


stresses and its challenges. The behaviour of both groups of subjects was observed, recorded and analysed. The dependent measures were of two general types: transactions between and within each group of subjects, recorded on video and audio tape as well as directly observed; individual reactions on questionnaires, mood inventories, personality tests, daily guard shift reports, and post experimental interviews.


The 21 subjects who participated in the experiment were selected from an initial pool of 75 respondents, who answered a newspaper advertisement asking for male volunteers to participate in a psychological study of “prison life” in return for payment of $15 per day. Those who responded to the notice completed an extensive questionnaire concerning their family background, physical and mental health history, prior experience and attitudinal propensities with respect to sources of psychopathology (including their involvement in crime). Each respondent who completed the background questionnaire was interviewed by one of two experimenters. Finally, the 24 subjects who were judged to be most stable (physically and mentally), most mature, and least involved in anti-social behaviour were selected to participate in the study. On a random basis, half of the subjects were assigned the role of “guard”, half to the role of “prisoner”.

The subjects were normal, healthy males attending colleges throughout the United States who were in the Stanford area during the summer. They were largely of middle class socio-economic status, Caucasians (with the exception of one Oriental subject). Initially they were strangers to each other, a selection precaution taken to avoid the disruption of any pre-existing friendship patterns and to mitigate against any transfer into the experimental situation of previously established relationships or patterns of behaviour.

This final sample of subjects was administered a battery of psychological tests on the day prior to the start of the simulation, but to avoid any selective bias on the part of the experimenter-observers, scores were not tabulated until the study was completed.

Two subjects who were assigned to be a “stand-by” in case an additional “prisoner” was needed were not called, and one subject assigned to be a “stand-by” guard decided against participating just before the simulation phase began–thus, our data analysis is based upon ten prisoners and eleven guards in our experimental conditions.


Physical aspects of the prison

The prison was built in a 35-ft section of a basement corridor in the psychology building at Stanford University. It was partitioned by two fabricated walls, one of which was fitted with the only entrance door to the cell block, the other



contained a small observation screen. Three small cells (6×9 ft) were made from converted laboratory rooms by replacing the usual doors with steel barred, black painted ones, and removing all furniture.

A cot (with mattress, sheet and pillow) for each prisoner was the only furniture in the cells. A small closet across from the cells served as a solitary confinement facility; its dimensions were extremely small (2 x 2 x 7 ft) and it was unlit.

In addition, several rooms in an adjacent wing of the building were used as guards’ quarters (to change in and out of uniform or for rest and relaxation), a bedroom for the “warden” and “superintendent”, and an interview-testing room. Behind the observation screen at one end of the “yard” was video recording equipment and sufficient space for several observers.

Operational details

The “prisoner” subjects remained in the mock-prison 24 hours per day for the duration of the study. Three were arbitrarily assigned to each of the three cells; the others were on stand-by call at their homes. The “guard” subjects worked on three-man, eight-hour shifts; remaining in the prison environment only during their work shift, going about their usual lives at other times.

Role instruction

All subjects had been told that they would be assigned either the guard or the prisoner role on a completely random basis and all had voluntarily agreed to play either role for $15.00 per day for up to two weeks. They signed a contract guaranteeing a minimally adequate diet, clothing, housing and medical care as well as the financial remuneration in return for their stated “intention” of serving in the assigned role for the duration of the study.

It was made explicit in the contract that those assigned to be prisoners should expect to be under surveillance (have little or no privacy) and to have some of their basic civil rights suspended during their imprisonment, excluding physical abuse. They were given no other information about what to expect nor instructions about behaviour appropriate for a prisoner role. Those actually assigned to this treatment were informed by phone to be available at their place of residence on a given Sunday when we would start the experiment.

The subjects assigned to be guards attended an orientation meeting on the day prior to the induction of the prisoners. At this time they were introduced to the principal investigators, the “Superintendent” of the prison (P.G.Z.) and an undergraduate research assistant who assumed the administrative role of “Warden”. They were told that we wanted to try to simulate a prison environment within the limits imposed by pragmatic and ethical considerations. Their assigned task was to “maintain the reasonable degree of order within the prison necessary for its effective functioning”, although the specifics of how this



duty might be implemented were not explicitly detailed. They were made aware of the fact that while many of the contingencies with which they might be confronted were essentially unpredictable (e.g. prisoner escape attempts), part of their task was to be prepared for such eventualities and to be able to deal appropriately with the variety of situations that might arise. The “Warden” instructed the guards in the administrative details, including: the work-shifts, the mandatory daily completion of shift reports concerning the activity of guards and prisoners, the completion of “critical incident” reports which detailed unusual occurrences and the administration of meals, work and recreation programmes for the prisoners. In order to begin to involve these subjects in their roles even before the first prisoner was incarcerated, the guards assisted in the final phases of completing the prison complex–putting the cots in the cells, signs on the walls, setting up the guards’ quarters, moving furniture, water coolers, refrigerators, etc.

The guards generally believed that we were primarily interested in studying the behaviour of the prisoners. Of course, we were equally interested in the effect which enacting the role of guard in this environment would have on their behaviour and subjective states.

To optimise the extent to which their behaviour would reflect their genuine reactions to the experimental prison situation and not simply their ability to follow instructions, they were intentionally given only minimal guidelines for what it meant to be a guard. An explicit and categorical prohibition against the use of physical punishment or physical aggression was, however, emphasised by the experimenters. Thus, with this single notable exception, their roles were relatively unstructured initially, requiring each “guard” to carry out activities necessary for interacting with a group of “prisoners” as well as with other “guards” and the “correctional staff”.


In order to promote feelings of anonymity in the subjects each group was issued identical uniforms. For the guards, the uniform consisted of: plain khaki shirts and trousers, a whistle, a police night stick (wooden batons) and reflecting sunglasses which made eye contact impossible. The prisoners’ uniform consisted of loosely fitting muslin smocks with an identification number on front and back. No underclothes were worn beneath these “dresses”. A chain and lock were placed around one ankle. On their feet they wore rubber sandals and their hair was covered with a nylon stocking made into a cap. Each prisoner was also issued a toothbrush, soap, soapdish, towel and bed linen. No personal belongings were allowed in the cells.

The outfitting of both prisoners and guards in this manner served to enhance group identity and reduce individual uniqueness within the two groups. The khaki uniforms were intended to convey a military attitude, while the whistle and night-stick were carried as symbols of control and power. The prisoners’


uniforms were designed not only to deindividuate the prisoners but to be humiliating and serve as symbols of their dependence and subservience. The ankle chain was a constant reminder (even during their sleep when it hit the other ankle) of the oppressiveness of the environment. The stocking cap removed any distinctiveness associated with hair length, colour or style (as does shaving of heads in some “real” prisons and the military). The ill-fitting uniforms made the prisoners feel awkward in their movements, since these dresses were worn without undergarments, the uniforms forced them to assume unfamiliar postures, more like those of a woman than a man–another part of the emasculating process of becoming a prisoner.

Induction procedure

With the cooperation of Palo Alto City Police Department all of the subjects assigned to the prisoner treatment were unexpectedly “arrested” at their residences. A police officer charged them with suspicion of burglary or armed robbery, advised them of their legal rights, handcuffed them, thoroughly searched them (often as curious neighbours looked on) and carried them off to the police station in the rear of the police car. At the station they went through the standard routines of being fingerprinted, having an identification file prepared and then being placed in a detention cell. Each prisoner was blindfolded and subsequently driven by one of the experimenters and a subject-guard to our mock prison. Throughout the entire arrest procedure, the police officers involved maintained a formal, serious attitude, avoiding answering any questions of clarification as to the relation of this “arrest” to the mock prison study.

Upon arrival at our experimental prison, each prisoner was stripped, sprayed with a delousing preparation (a deodorant spray) and made to stand alone naked for a while in the cell yard. After being given the uniform described previously and having an I.D. picture taken (“mug shot”), the prisoner was put in his cell and ordered to remain silent.

Administrative routine

When all the cells were occupied, the warden greated the prisoners and read them the rules of the institution (developed by the guards and the warden). They were to be memorised and to be followed. Prisoners were to be referred to only by the number on their uniforms, also in an effort to depersonalise them.

The prisoners were to be served three bland meals per day, were allowed three supervised toilet visits, and given two hours daily for the privilege of reading or letterwriting. Work assignments were issued for which the prisoners were to receive an hourly wage to constitute their $15 daily payment. Two visiting periods per week were scheduled, as were movie rights and exercise periods. Three times a day all prisoners were lined up for a “count” (one on each guard

work-shift). The initial purpose of the “count” was to ascertain that all prisoners were present, and to test them on their knowledge of the rules and their I.D. numbers. The first perfunctory counts lasted only about 10 minutes, but on each successive day (or night) they were spontaneously increased in duration until some lasted several hours. Many of the pre-established features of administrative routine were modified or abandoned by the guards, and some were forgotten by the staff over the course of the study.

Data collection (dependent measures)

The exploratory nature of this investigation and the absence of specific hypotheses led us to adopt the strategy of surveying as many as possible behavioural and psychological manifestations of the prison experience on the guards and the prisoners. In fact, one major methodological problem in a study of this kind is defining the limits of the “data”, since relevant data emerged from virtually every interaction between any of the participants, as well as from subjective and behavioural reactions of individual prisoners, guards, the warden, superintendent, research assistants and visitors to the prison. It will also be clear when the results are presented that causal direction cannot always be established in the patterns of interaction where any given behaviour might be the consequence of a current or prior instigation by another subject and, in turn, might serve as impetus for eliciting reactions from others.

Data collection was organised around the following sources:

(1)                      Videotaping. About 12 hours of recordings were made of daily, regularly occurring events, such as the counts and meals, as well as unusual interactions, such as a prisoner rebellion, visits from a priest, a lawyer and parents, Parole Board meetings and others. Concealed video equipment recorded these events through a screen in the partition at one end of the cell-block yard or in a conference room (for parole meetings).

(2)                Audio recording. Over 30 hours of recordings were made of verbal interactions between guards and prisoners on the prison yard. Concealed microphones picked up all conversation taking place in the yard as well as some within the cells. Other concealed recordings were made in the testing-interview room on selected occasions–interactions between the warden, superintendent and the prisoners’ Grievance Committee, parents, other visitors and prisoners released early. In addition, each subject was interviewed by one of the experimenters (or by other research associates) during the study, and most just prior to its termination.

(3)                Rating scales. Mood adjective checklists and sociometric measures were administered on several occasions to assess emotional changes in affective state and interpersonal dynamics among the guard and prisoner groups.

(4)                Individual difference scales. One day prior to the start of the simulation all subjects completed a series of paper and pencil personality tests. These tests


were selected to provide dispositional indicators of interpersonal behaviour styles–the F scale of Authoritarian Personality [1], and the Machiavellianism Scale [2]–as well as areas of possible personality pathology through the newly developed Comrey Personality Scale [3]. The subscales of this latter test consist of:

(a)                    trustworthiness

(b)                    orderliness

(c)                    conformity

(d)                    activity

(e)                    stability

(f)                     extroversion

(g)                    masculinity

(h)                    empathy

(5)                Personal observations. The guards made daily reports of their observations after each shift, the experimenters kept informal diaries and all subjects completed post-experimental questionnaires of their reactions to the experience about a month after the study was over.

Data analyses presented problems of several kinds. First, some of the data was subject to possible errors due to selective sampling. The video and audio recordings tended to be focussed upon the more interesting, dramatic events which occurred. Over time, the experimenters became more personally involved in the transaction and were not as distant and objective as they should have been. Second, there are not complete data on all subjects for each measure because of prisoners being released at different times and because of unexpected disruptions, conflicts and administrative problems. Finally, we have a relatively small sample on which to make cross-tabulations by possible independent and individual difference variables.

However, despite these shortcomings some of the overall effects in the data are powerful enough to reveal clear, reliable results. Also some of the more subtle analyses were able to yield statistically significant results even with the small sample size. Most crucial for the conclusions generated by this exploratory study is the consistency in the pattern of relationships which emerge across a wide range of measuring instruments and different observers. Special analyses were required only of the video and audio material, the other data sources were analysed following established scoring procedures.

Video analysis

There were 25 relatively discrete incidents identifiable on the tapes of prisoner-guard interactions. Each incident or scene was scored for the presence of nine behavioural (and verbal) categories. Two judges who had not been involved with the simulation study scored these tapes. These categories were defined as follows:


Question. All questions asked, requests for information or assistance (excluding rhetorical questions).

Command. An order to commence or abstain from a specific behaviour, directed either to individuals or groups. Also generalised orders, e.g. “Settle down”.

Information. A specific piece of information proffered by anyone whether requested or not, dealing with any contingency of the simulation.

Individuating reference. Positive: use of a person’s real name, nickname or allusion to special positive physical characteristics. Negative: use of prison number, title, generalised «you” or reference to derogatory characteristic.

Threat. Verbal statement of contingent negative consequences of a wide variety, e.g. no meal, long count, pushups, lock-up in hole, no visitors, etc.

Deprecation insult. Use of obscenity, slander, malicious statement directed toward individual or group, e.g. “You lead a life of mendacity” or “You guys are really stupid.”

Resistance. Any physical resistance, usually prisoners to guards, such as holding on to beds, blocking doors, shoving guard or prisoner, taking off stocking caps, refusing to carry out orders.

Help. Person physically assisting another (i.e. excludes verbal statements of support), e.g. guard helping another to open door, prisoner helping another prisoner in cleanup duties.

Use of instruments. Use of any physical instrument to either intimidate, threaten, or achieve specific end, e.g. fire extinguisher, batons, whistles.

Audio analysis

For purposes of classifying the verbal behaviour recorded from interviews with guards and prisoners, eleven categories were devised. Each statement made by the interviewee was assigned to the appropriate category by judges. At the end of this process for any given interview analysis, a list had been compiled of the nature and frequencies of the interviewee’s discourse. The eleven categories for assignment of verbal expressions were:

Questions. All questions asked, requests for information or assistance (excluding rhetorical questions).

Informative statements. A specific piece of information proffered by anyone whether requested or not, dealing with any contingency of the simulation.

Demands. Declarative statements of need or imperative requests.

Requests. Deferential statements for material or personal consideration.

Commands. Orders to commence or abstain from a specific behaviour, directed either to individuals or groups.

Outlook, positive/negative. Expressions of expectancies for future experiences or future events; either negative or positive in tone, e.g. “I don’t think I can make it” v. «I believe I will feel better.”


Criticism. Expressions of critical evaluation concerning other subjects, the experimenters or the experiment itself.

Statements of identifying reference, deindividuating/individuating. Statements wherein a subject makes some reference to another subject specifically by allusion to given name or distinctive characteristics (individuating reference), or by allusion to non-specific identity or institutional number (deindividuating reference).

Desire to continue. Any expression of a subject’s wish to continue or to curtail participation in the experiment.

Self-evaluation, positive I negative. Statements of self-esteem or selfdegradation, e.g. “I feel pretty good about the way I’ve adjusted” v. “I hate myself for being so oppressive.”

Action intentions, positive/negative including “intent to aggress». Statements concerning interviewees’ intentions to do something in the future, either of a positive, constructive nature or a negative, destructive nature, e.g. “I’m not going to be so mean from now on” v. “I’ll break the door down.”



Although it is difficult to anticipate exactly what the influence of incarceration will be upon the individuals who are subjected to it and those charged with its maintenance (especially in a simulated reproduction), the results of the present experiment support many commonly held conceptions of prison life and validate anecdotal evidence supplied by articulate ex-convicts. The environment of arbitrary custody had great impact upon the affective states of both guards and prisoners as well as upon the interpersonal processes taking place between and within those role-groups.

In general, guards and prisoners showed a marked tendency toward increased negativity of affect and their overall outlook became increasingly negative. As the experiment progressed, prisoners expressed intentions to do harm toothers more frequently. For both prisoners and guards, self-evaluations were more deprecating as the experience of the prison environment became internalised.

Overt behaviour was generally consistent with the subjective self-reports and affective expressions of the subjects. Despite the fact that guards and prisoners were essentially free to engage in any form of interaction (positive or negative, supportive or affrontive, etc.), the characteristic nature of their encounters tended to be negative, hostile, affrontive and dehumanising. Prisoners immediately adopted a generally passive response mode while guards assumed a very active initiating role in all interactions. Throughout the experiment, commands were the most frequent form of verbal behaviour and, generally, verbal exchanges were strikingly impersonal, with few references to individual identity. Although it was clear to all subjects that the experimenters would not

Interpersonal dynamics in a simulated prison                                                                            81

permit physical violence to take place, varieties of less direct aggressive behaviour were observed frequently (especially on the part of guards). In lieu of physical violence, verbal affronts were used as one of the most frequent forms of interpersonal contact between guards and prisoners.

The most dramatic evidence of the impact of this situation upon the participants was seen in the gross reactions of five prisoners who had to be released because of extreme emotional depression, crying, rage and acute anxiety. The pattern of symptoms was quite similar in four of the subjects and began as early as the second day of imprisonment. The fifth subject was released after being treated for a psychosomatic rash which covered portions of his body. Of the remaining prisoners, only two said they were not willing to forfeit the money they had earned in return for being “paroled”. When the experiment was terminated prematurely after only six days, all the remaining prisoners were delighted by their unexpected good fortune. In contrast, most of the guards seemed to be distressed by the decision to stop the experiment and it appeared to us that had become sufficiently involved in their roles so that they now enjoyed the extreme control and power which they exercised and were reluctant to give it up. One guard did report being personally upset at the suffering of the prisoners and claimed to have considered asking to change his role to become one of them–but never did so. None of the guards ever failed to come to work on time for their shift, and indeed, on several occasions guards remained on duty voluntarily and uncomplaining for extra hours–without additional pay.

The extremely pathological reactions which emerged in both groups of subjects testify to the power of the social forces operating, but still there were individual differences seen in styles of coping with this novel experience and in degrees of successful adaptation to it. Half the prisoners did endure the oppressive atmosphere, and not all the guards resorted to hostility. Some guards were tough but fair (“played by the rules”), some went far beyond their roles to engage in creative cruelty and harassment, while a few were passive and rarely instigated any coercive control over the prisoners.

These differential reactions to the experience of imprisonment were not suggested by or predictable from the self-report measures of personality and attitude or the interviews taken before the experiment began. The standardised tests employed indicated that a perfectly normal emotionally stable sample of subjects had been selected. In those few instances where differential test scores doTdiscriminate between subjects, there is an opportunity to, partially at least, discern some of the personality variables which may be critical in the adaptation to and tolerance of prison confinement.

Intitial personality and attitude measures

Overall, it is apparent that initial personality-attitude dispositions account for an extremely small part of the variation in reactions to this mock prison experience. However, in a few select instances, such dispositions do seem to be correlated with the prisoners’ ability to adjust to the experimental prison environment.


Comrey scale

The Comrey Personality Inventory [3] was the primary personality scale administered to both guards and prisoners. The mean scores for prisoners and guards on the eight sub-scales of the test are shown in Table 1. No differences between prisoner and guard mean scores on any scale even approach statistical significance. Furthermore, in no case does any group mean fall outside of the 40 to 60 centile range of the normative male population reported by Comrey.

Table 1. Mean scores for prisoners and guards on eight Comrey subscales

Scale Prisoners Guards
Trustworthiness–high score indicates belief in the basic honesty and good intentions of others X = 92.56 X = 89.64
Orderliness–extent to which person is meticulous and concerned with neatness and orderliness X = 75.67 X = 73.82
Conformity–indicates belief in law enforcement, acceptance of society as it is, resentment of nonconformity in others X = 65.67 X = 63.18
Activity–liking for physical activity, hard work, and exercise X = 89.78 X= 91.73
Stability–high score indicates calm, optimistic, stable, confident individual X = 98.33 X = 101.45
Extroversion–suggests outgoing, easy to meet person X = 83.22 X= 81.91
Masculinity–“people who are not bothered by crawling creatures, the sight of blood, vulgarity, who do not cry easily and are not interested in love stories” X = 88.44 X = 87.00
Empathy–high score indicates individuals who are sympathetic, helpful, generous and interested in devoting their lives to the service of others X = 91.78 X= 95.36


Table 2. Mean scores for “Remaining” v. «Early released” prisoners on Comrey subscales

Scale Remaining prisoners Early released prisoners Mean difference
Trustworthiness 93.4 90.8 +2.6
Orderliness 76.6 78.0 -1.4
Conformity 67.2 59.4 +7.8
Activity 91.4 86.8 +4.6
Stability 99.2 99.6 -0.4
Extroversion 98.4 76.2 +22.2
Masculinity 91.6 86.0 +5.6
Empathy 103.8 85.6 +17.2



Table 2 shows the mean scores on the Comrey sub-scales for prisoners who remained compared with prisoners who were released early due to severe emotional reactions to the environment. Although none of the comparisons achieved statistical significance, three seemed at least suggestive as possible discriminators of those who were able to tolerate this type of confinement and those who were not. Compared with those who had to be released, prisoners who remained in prison until the termination of the study: scored higher on conformity (“acceptance of society as it is”), showed substantially higher average scores on Comrey’s measure of extroversion and also scored higher on a scale of empathy (helpfulness, sympathy and generosity).


The F-scale is designed to measure rigid adherence to conventional values and a submissive, uncritical attitude towards authority. There was no difference between the mean score for prisoners (4.78) and the mean score for guards (4.36) on this scale.

Again, comparing those prisoners who remained with those who were released early, we notice an interesting trend. This intra-group comparison shows remaining prisoners scoring more than twice as high on conventionality and authoritarianism (X = 7.78) than those prisoners released early (X = 3.20). While the difference between these means fails to reach acceptable levels of significance, it is striking to note that a rank-ordering of prisoners on the F-scale correlates highly with the duration of their stay in the experiment [rs= 0.898, F< 0.005). To the extent that a prisoner was high in rigidity, in adherence to conventional values, and in the acceptance of authority, he was likely to remain longer and adjust more effectively to this authoritarian prison environment.


There were no significant mean differences found between guards (X- 7.73) and prisoners (X = 8.77) on this measure of effective interpersonal manipulation. In addition, the Mach Scale was of no help in predicting the likelihood that a prisoner would tolerate the prison situation and remain in the study until its termination.

This latter finding, the lack of any mean differences between prisoners who remained v. those who were released from the study, is somewhat surprising since one might expect the Hi Mach’s skill at manipulating social interaction and mediating favourable outcomes for himself might be acutely relevant to the simulated prison environment. Indeed, the two prisoners who scored highest on the Machiavellianism scale were also among those adjudged by the experimenters to have made unusually effective adapatations to their confinement. Yet, paradoxically (and this may give the reader some feeling for the anomalies we encountered in attempting to predict in-prison behaviour from personality


measures), the other two prisoners whom we categorised as having effectively adjusted to confinement actually obtained the lowest Mach scores of any prisoners.

Video recordings

An analysis of the video recordings indicates a preponderance of genuinely negative interactions, i.e. physical aggression, threats, deprecations, etc. It is also clear that any assertive activity was largely the prerogative of the guards, while prisoners generally assumed a relatively passive demeanour. Guards more often aggressed, more often insulted, more often threatened. Prisoners, when they reacted at all, engaged primarily in resistance to these guard behaviours.

For guards, the most frequent verbal behaviour was the giving of commands and their most frequent form of physical behaviour was aggression. The most frequent form of prisoners’ verbal behaviour was question-asking, their most frequent form of physical behaviour was resistance. On the other hand, the most infrequent behaviour engaged in overall throughout the experiment was “helping”–only one such incident was noted from all the video recording collected. That solitary sign of human concern for a fellow occurred between two prisoners.

Although question-asking was the most frequent form of verbal behaviour for the prisoners, guards actually asked questions more frequently overall than did prisoners (but not significantly so). This is reflective of the fact that the overall level of behaviour emitted was much higher for the guards than for the prisoners. All of those verbal acts categorised as commands were engaged in by guards. Obviously, prisoners had no opportunity to give commands at all, that behaviour becoming the exclusive «right” of guards.

Of a total 61 incidents of direct interpersonal reference observed (incidents in which one subject spoke directly to another with the use of some identifying reference, i.e. “Hey, Peter”; “you there”, etc.), 58 involved the use of some deindividuating rather than some individuating form of reference. (Recall that we characterised this distinction as follows: an individuating reference involved the use of a person’s actual name, nickname or allusion to special physical characteristics, whereas a deindividuating reference involved the use of a prison number, or a generalised “you”–thus being a very depersonalising form of reference.) Since all subjects were at liberty to refer to one another in either mode, it is significant that such a large proportion of the references noted involved were in the deindividuating mode (Z = 6.9, P <0.01). Deindividuating references were made more often by guards in speaking to prisoners than the reverse (Z = 3.67, P < 0.01). (This finding, as all prisoner-guard comparisons for specific categories, may be somewhat confounded by the fact that guards apparently enjoyed a greater freedom to initiate verbal as well as other forms of behaviour. Note, however, that the existence of this greater “freedom” on the part of the guards is itself an empirical finding since it was not prescribed


a priori.) It is of additional interest to point out that in the only three cases in which verbal exchange involved some individuating reference, it was prisoners who personalised guards.

A total of 32 incidents were observed which involved a verbal threat spoken by one subject to another. Of these, 27 such incidents involved a guard threatening a prisoner. Again, the indulgence of guards in this form of behaviour was significantly greater than the indulgence of prisoners, the observed frequencies deviating significantly from an equal distribution of threats across both groups (Z = 3.88, P < 0.01).

Guards more often deprecated and insulted prisoners than prisoners did of guards. Of a total of 67 observed incidents, the deprecation-insult was expressed disproportionately by guards to prisoners 61 times; (Z = 6.72, P< 0.01).

Physical resistance was observed 34 different times. Of these, 32 incidents involved resistance by a prisoner. Thus, as we might expect, at least in this reactive behaviour domain, prisoner responses far exceeded those of the guards (Z = 5.14, P< 0.01).

The use of some object or instrument in the achievement of an intended purpose or in some interpersonal interaction was observed 29 times. Twenty- three such incidents involved the use of an instrument by a guard rather than a prisoner. This disproportionate frequency is significantly variant from an equal random use by both prisoners and guards (Z = 316, P < 0.01).

Over time, from day to day, guards were observed to generally escalate their harassment of the prisoners. In particular, a comparison of two of the first prisoner-guard interactions (during the counts) with two of the last counts in the experiment yielded significant differences in: the use of deindividuating references per unit time (Xt{ =0.0 and Xtг =5.40, respectively; t= 3.65, P<0.10); the incidence of deprecation-insult per unit time (Xti =0.3 and Zf2 = 5.70, respectively; f = 3.16, P<0.10). On the other hand, a temporal analysis of the prisoner video data indicated a general decrease across all categories over time: prisoners came to initiate acts far less frequently and responded (if at all) more passively to the acts of others–they simply behaved less.

Although the harassment by the guards escalated overall as the experiment wore on, there was some variation in the extent to which the three different guard shifts contributed to the harassment in general. With the exception of the 2.30 a.m. count, prisoners enjoyed some respite during the late night guard shift (10.00 p.m. to 6.00 a.m.). But they really were “under the gun” during the evening shift. This was obvious in our observations and in subsequent interviews with the prisoners and was also confirmed in analysis of the video taped interactions. Comparing the three different guard shifts, the evening shift was significantly different from the other two in resorting to commands; the means being 9.30 and 4.04, respectively, for standardised units of time {t = 2.50, P< 0.05). In addition, the guards on this “tough and cruel” shift showed more than twice as many deprecation-insults toward the prisoners (means of 5.17 and

2.29, respectively, P< 0.20). They also tended to use instruments more often than other shifts to keep the prisoners in line.

Audio recordings

The audio recordings made throughout the prison simulation afforded one opportunity to systematically collect self-report data from prisoners and guards regarding (among other things) their emotional reactions, their outlook, and their interpersonal evaluations and activities within the experimental setting. Recorded interviews with both prisoners and guards offered evidence that: guards tended to express nearly as much negative outlook and negative self-regard as most prisoners (one concerned guard, in fact, expressed more negative self-regard than any prisoner and more general negative affect than all but one of the prisoners); prisoner interviews were marked by negativity in expressions of affect, self-regard and action intentions (including intent to aggress and negative outlook).

Analysis of the prisoner interviews also gave post hoc support to our informal impressions and subjective decisions concerning the differential emotional effects of the experiment upon those prisoners who remained and those who were released early from the study. A comparison of the mean number of expressions of negative outlook, negative affect, negative self-regard and intentions to aggress made by remaining v. released prisoners (per interview) yielded the following results: prisoners released early expressed more negative expectations during interviews than those who remained (t = 2.32, Z><0.10) and also more negative affect (f=2.17, /^ < 0.10); prisoners released early expressed more negative self-regard, and four times as many “intentions to aggress” as prisoners who remained (although those comparisons fail to reach an acceptable level of significance).

Since we could video-record only public interactions on the “yard”, it was of special interest to discover what was occurring among prisoners in private. What were they talking about in the cells–their college life, their vocation, girl friends, what they would do for the remainder of the summer once the experiment was over. We were surprised to discover that fully 90% of all conversations among prisoners were related to prison topics, while only 10% to non-prison topics such as the above. They were most concerned about food, guard harassment, setting up a grievance committee, escape plans, visitors, reactions of prisoners in the other cells and in solitary. Thus, in their private conversations when they might escape the roles they were playing in public, they did not. There was no discontinuity between their presentation of self when under surveillance and when alone.

Even more remarkable was the discovery that the prisoners had begun to adopt and accept the guards’ negative attitude toward them. Half of all reported private interactions between prisoners could be classified as non-supportive and non-cooperative. Moreover, when prisoners made evaluative statements of or


expressed regard for, their fellow prisoners, 85% of the time they were uncomplimentary and deprecating. This set of observed frequencies departs significantly from chance expectations based on a conservative binominal probability frequency (P< 0.01 for prison v. non-prison topics; P<0.05 for negative v. positive or neutral regard).

Mood adjective self-reports

Twice during the progress of the experiment each subject was asked to complete a mood adjective checklist and indicate his current affective state. The data gleaned from these self-reports did not lend themselves readily to statistical analysis. However, the trends suggested by simple enumeration are important enough to be included without reference to statistical significance. In these written self-reports, prisoners expressed nearly three times as much negative as positive^ffect. Prisoners roughly expressed three times as much negative affect as guards.Guardsexpressed slightly more negative than positive affect. While prisoners expressed about twice as much emotionality as did guards, a comparison of mood self-reports over time reveals that the prisoners showed two to three times as much mood fluctuation as did the relatively stable guards. On the dimension of activity-passivity, prisoners tended to score twice as high, indicating twice as much internal “agitation” as guards (although, as stated above, prisoners were seen to be markedly less active than guards in terms of overt behaviour).

It would seem from these results that while the experience had a categorically negative emotional impact upon both guards and prisoners, the effects upon prisoners were more profound and unstable.

When the mood scales were administered for a third time, just after the subjects were told the study had been terminated (and the early released subjects returned for the debriefing encounter session), marked changes in mood were evident. All of the now “ex-convicts” selected self-descriptive adjectives which characterised their mood as less negative and much more positive. In addition, they now felt less passive than before. There were no longer any differences on the sub-scales of this test between prisoners released early and those who remained throughout. Both groups of subjects had returned to their pre- experimental baselines of emotional responding. This seems to reflect the situational specificity of the depression and stress reactions experienced while in the role of prisoner.

Representative personal statements

Much of the flavour and impact of this prison experience is unavoidably lost in the relatively formal, objective analyses outlined in this paper. The following quotations taken from interviews, conversations and questionnaires provide a more personal view of what it was like to be a prisoner or guard in the “Stanford County Prison” experiment.



“They [the prisoners] seemed to lose touch with the reality of the experiment–they took me so seriously.”

«… I didn’t interfere with any of the guards’ actions. Usually if what they were doing bothered me, I would walk out and take another duty.”

“. . . looking back, I am impressed by how little I felt for them . . .”

«. . . They [the prisoners] didn’t see it as an experiment. It was real and they were fighting to keep their identity. But we were always there to show them just who was boss.”

«… I was tired of seeing the prisoners in their rags and smelling the strong odours of their bodies that filled the cells. I watched them tear at each other, on orders given by us.”

“. . . Acting authoritatively can be fun. Power can be a great pleasure.”

«. . . During the inspection, I went to cell 2 to mess up a bed which the prisoner had made and he grabbed me, screaming that he had just made it, and he wasn’t going to let me mess it up. He grabbed my throat, and although he was laughing I was pretty scared. I lashed out with my stick and hit him in the chin (although not very hard) and when I freed myself I became angry.”


«. . . The way we were made to degrade ourselves really brought us down and that’s why we all sat docile towards the end of the experiment.”

”… I realise now (after it’s over) that no matter how together I thought I was inside my head, my prison behaviour was often less under my control than I realised. No matter how open, friendly and helpful I was with other prisoners I was still operating as an isolated, self-centred person, being rational rather than compassionate.”

”… I began to feel I was losing my identity, that the person I call

——————————————————— , the person who volunteered to get me into this prison (because

it was a prison to me, it still is a prison to me, I don’t regard it as an experiment or a simulation . . .) was distant from me, was remote until finally I wasn’t that person, I was 416. I was really my number and 416 was really going to have to decide what to do.”

“I learned that people can easily forget that others are human.”

Debriefing encounter sessions

Because of the unexpectedly intense reactions (such as the above) generated by this mock-prison experience, we decided to terminate the study at the end of six days rather than continue for the second week. Three separate encounter sessions were held, first, for the prisoners, then for the guards and finally for all participants together. Subjects and staff openly discussed their reactions and strong feelings were expressed and shared. We analysed the moral conflicts posed by this experience and used the debriefing sessions to make explicit alternative courses of action that would lead to more moral behaviour in future comparable situations.

Follow-ups on each subject over the year following termination of the study revealed the negative effects of participation had been temporary, while the personal gain to the subjects endured.

Conclusions and Discussion

It should be apparent that the elaborate procedures (and staging) employed by the experimenters to insure a high degree of mundane realism in this mock prison contributed to its effective functional simulation of the psychological dynamics operating in “real” prisons. We observed empirical relationships in the simulated prison environment which were strikingly isomorphic to the internal relations of real prisons, corroborating many of the documented reports of what occurs behind prison walls.

The conferring of differential power on the status of “guard” and “prisoner” constituted, in effect, the institutional validation of those roles. But further, many of the subjects ceased distinguishing between prison role and their prior self-identities. When this occurred, within what was a surprisingly short period of time, we witnessed a sample of normal, healthy American college students fractionate into a group of prison guards who seemed to derive pleasure from insulting, threatening, humiliating and dehumanising their peers–those who by chance selection had been assigned to the “prisoner” role. The typical prisoner syndrome was one of passivity, dependency, depression, helplessness and self-deprecation. Prisoner participation in the social reality which the guards had structured for them lent increasing validity to it and, as the prisoners became resigned to their treatment over time, many acted in ways to justify their fate at the hands of the guards, adopting attitudes and behaviour which helped to sanction their victimisation. Most dramatic and distressing to us was the observation of the ease with which sadistic behaviour could be elicited in individuals who were not “sadistic types” and the frequency with which acute emotional breakdowns could occur in men selected precisely for their emotional stability.

Situational v. dispositional attribution

To what can we attribute these deviant behaviour patterns? If these reactions had been observed within the confines of an existing penal institution, it is probable that a dispositional hypothesis would be invoked as an explanation. Some cruel guards might be singled out as sadistic or passive-aggressive personality types who chose to work in a correctional institution because of the outlets provided for sanctioned aggression. Aberrant reactions on the part of the inmate population would likewise be viewed as an extrapolation from the prior social histories of these men as violent, anti-social, psychopathic, unstable character types.

Existing penal institutions may be viewed as natural experiments in social control in which any attempts at providing a causal attribution for observed behaviour hopelessly confound dispositional and situational causes. In contrast, the design of our study minimised the utility of trait or prior social history explanations by means of judicious subject selection and random assignment to roles. Considerable effort and care went into determining the composition of thу final subject population from which our guards and prisoners were drawn. Through case histories, personal interviews and a battery of personality tests, the subjects chosen to participate manifested no apparent abnormalities, anti-social tendencies or social backgrounds which were other than exemplary. On every one of the scores of the diagnostic tests each subject scored within the normal-average range. Our subjects then, were highly representative of middle- class, Caucasian American society (17 to 30 years in age), although above average in both intelligence and emotional stability.

Nevertheless, in less than one week their behaviour in this simulated prison could be characterised as pathological and anti-social. The negative, anti-social reactions observed were not the product of an environment created by combining a collection of deviant personalities, but rather, the result of an intrinsically pathological situation which could distort and rechannel the behaviour of essentially normal individuals. The abnormality here resided in the psychological nature of the situation and not in those who passed through it. Thus, we offer another instance in support of Mischel’s [4] social-learning analysis of the power of situational variables to shape complex social behaviour. Our results are also congruent with those of Milgram [5] who most convincingly demonstrated the proposition that evil acts are not necessarily the deeds of evil men, but may be attributable to the operation of powerful social forces. Our findings go one step further, however, in removing the immediate presence of the dominant experimenter-authority figure, giving the subjects-as-guards a freer range of behavioural alternatives, and involving the participants for a much more extended period of time.

Despite the evidence favouring a situational causal analysis in this experiment, it should be clear that the research design actually minimised the effects of individual differences by use of a homogenous middle-range subject population. It did not allow the strongest possible test of the relative utility of the two types of explanation. We cannot say that personality differences do not have an important effect on behaviour in situations such as the one reported here. Rather, we may assert that the variance in behaviour observed could be reliably attributed to variations in situational rather than personality variables. The inherently pathological characteristics of the prison situation itself, at least as functionally simulated in our study, were a sufficient condition to produce aberrant, anti-social behaviour. (An alternative design which would maximise the potential operation of personality or dispositional variables would assign subjects who were extreme on pre-selected personality dimensions to each of the two experimental treatments. Such a design would, however, require a larger subject population and more resources than we had available.)

The failure of personality assessment variables to reliably discriminate the various patterns of prison behaviour, guard reactions as well as prisoner coping styles is reminiscent of the inability of personality tests to contribute to an understanding of the psychological differences between American P.O.W.s in Korea who succumbed to alleged Chinese Communist brain-washing by “collaborating with the enemy” and those who resisted [6]. It seems to us that there is little reason to expect paper-and-pencil behavioural reactions on personality tests taken under “normal” conditions to generalise into coping behaviours under novel, stressful or abnormal environmental conditions. It may be that the best predictor of behaviour in situations of stress and power, as occurs in prisons, is overt behaviour in functionally comparable simulated environments.

In the situation of imprisonment faced by our subjects, despite the potent situational control, individual differences were nevertheless manifested both in coping styles among the prisoners and in the extent and type of aggression and exercise of power among the guards. Personality variables, conceived as learned behaviour styles can act as moderator variables in allaying or intensifying the impact of social situational variables. Their predictive utility depends upon acknowledging the inter-active relationship of such learned dispositional tendencies with the eliciting force of the situational variables.

Reality of the simulation

At this point it seems necessary to confront the critical question of “reality” in the simulated prison environment: were the behaviours observed more than the mere acting out assigned roles convincingly? To be sure, ethical, legal and practical considerations set limits upon the degree to which this situation could approach the conditions existing in actual prisons and penitentiaries. Necessarily absent were some of the most salient aspects of prison life reported by criminologists and documented in the writing of prisoners [7, 8]. There was no involuntary homosexuality, no racism, no physical beatings, no threat to life by prisoners against each other or the guards. Moreover, the maximum anticipated “sentence” was only two weeks and, unlike some prison systems, could not be extended indefinitely for infractions of the internal operating rules of the prison.

In one sense, the profound psychological effects we observed under the relatively minimal prison-like conditions which existed in our mock prison make the results even more significant and force us to wonder about the devastating impact of chronic incarceration in real prisons. Nevertheless, we must contend with the criticism that the conditions which prevailed in the mock prison were too minimal to provide a meaningful analogue to existing prisons. It is necessary to demonstrate that the participants in this experiment transcended the conscious limits of their preconceived stereotyped roles and their awareness of the artificiality and limited duration of imprisonment. We feel there is abundant evidence that virtually all of the subjects at one time or another experienced reactions which went well beyond the surface demands of role-playing and penetrated the deep structure of the psychology of imprisonment.

Although instructions about how to behave in the roles of guard or prisoner were not explicitly defined, demand characteristics in the experiment obviously exerted some directing influence. Therefore, it is enlightening to look to


circumstances where role demands were minimal, where the subjects believed they were not being observed, or where they should not have been behaving under the constraints imposed by their roles (as in “private” situations), in order to assess whether the role behaviours reflected anything more than public conformity or good acting.

When the private conversations of the prisoners were monitored, we learned that almost all (a full 90%) of what they talked about was directly related to immediate prison conditions, that is, food, privileges, punishment, guard harassment, etc. Only one-tenth of the time did their conversations deal with their life outside the prison. Consequently, although they had lived together under such intense conditions, the prisoners knew surprisingly little about each other’s past history or future plans. This excessive concentration on the vicissitudes of their current situation helped to make the prison experience more oppressive for the prisoners because, instead of escaping from it when they had a chance to do so in the privacy of their cells, the prisoners continued to allow it to dominate their thoughts and social relations. The guards too, rarely exchanged personal information during their relaxation breaks. They either talked about “problem prisoners”, or other prison topics, or did not talk at all. There were few instances of any personal communication across the two role groups. Moreover, when prisoners referred to other prisoners during interviews, they typically deprecated each other, seemingly adopting the guards’ negative attitude.

From post-experimental data, we discovered that when individual guards were alone with solitary prisoners and out of range of any recording equipment, as on the way to or in the toilet, harassment often was greater than it was on the “Yard”. Similarly, video-taped analyses of total guard aggression showed a daily escalation even after most prisoners had ceased resisting and prisoner deterioration had become visibly obvious to them. Thus guard aggression was no longer elicited as it was initially in response to perceived threats, but was emitted simply as a “natural” consequence of being in the uniform of a “guard” and asserting the power inherent in that role. In specific instances we noted cases of a guard (who did not know he was being observed) in the early morning hours pacing the “Yard” as the prisoners slept–vigorously pounding his night stick into his hand while he “kept watch” over his captives. Or another guard who detained an “incorrigible” prisoner in solitary confinement beyond the duration set by the guards’ own rules and then he conspired to keep him in the hole all night while attempting to conceal this information from the experimenters who were thought to be too soft on the prisoners.

In passing, we may note an additional point about the nature of role-playing and the extent to which actual behaviour is “explained away” by reference to it. It will be recalled that many guards continued to intensify their harassment and aggressive behaviour even after the second day of the study, when prisoner deterioration became marked and visible and emotional breakdowns began to occur (in the presence of the guards). When questioned after the study about their persistent affrontive and harrassing behaviour in the face of prisoner


emotional trauma, most guards replied that they were “just playing the role” of a tough guard, although none ever doubted the magnitude or validity of the prisoners’ emotional response. The reader may wish to consider to what extremes an individual may go, how great must be the consequences of his behaviour for others, before he can no longer rightfully attribute his actions to “playing a role” and thereby abdicate responsibility.

When introduced to a Catholic priest, many of the role-playing prisoners referred to themselves by their prison number rather than their Christian names. Some even asked him to get a lawyer to help them get out. When a public defender was summoned to interview those prisoners who had not yet been released, almost all of them strenuously demanded that he “bail” them out immediately.

One of the most remarkable incidents of the study occurred during a parole board hearing when each of five prisoners eligible for parole was asked by the senior author whether he would be willing to forfeit all the money earned as a prisoner if he were to be paroled (released from the study). Three of the five prisoners said, “yes”, they would be willing to do this. Notice that the original incentive for participating in the study had been the promise of money, and they were, after only four days, prepared to give this up completely. And, more suprisingly, when told that this possibility would have to be discussed with the members of the staff before a decision could be made, each prisoner got up quietly and was escorted by a guard back to his cell. If they regarded themselves simply as “subjects” participating in an experiment for money, there was no longer any incentive to remain in the study and they could have easily escaped this situation which had so clearly become aversive for them by quitting. Vet, so powerful was the control which the situation had come to have over them, so much a reality had this simulated environment become, that they were unable to see that their original and singular motive for remaining no longer obtained, and they returned to their cells to await a “parole” decision by their captors.

The reality of the prison was also attested to by our prison consultant who had spent over 16 years in prison, as well as the priest who had been a prison chaplain and the public defender who were all brought into direct contact with out simulated prison environment. Further, the depressed affect of the prisoners, the guards’ willingness to work overtime for no additional pay, the spontaneous use of prison titles and I.D. numbers in non role-related situations all point to a level of reality as real as any other in the lives of all those who shared this experience.

To understand how an illusion of imprisonment could have become so real, we need now to consider the uses of power by the guards as well as the effects of such power in shaping the prisoner mentality.

Pathology of power

Being a guard carried with it social status within the prison, a group identity (when wearing the uniform), and above all, the freedom to exercise an unprecedented degree of control over the lives of other human beings. This

control was invariably expressed in terms of sanctions, punishment, demands and with the threat of manifest physical power. There was no need for the guards to rationally justify a request as they do in their ordinary life and merely to make a demand was sufficient to have it carried out. Many of the guards showed in their behaviour and revealed in post-experimental statements that this sense of power was exhilarating.

The use of power was self-aggrandising and self-perpetuating. The guard power, derived initially from an arbitrary label, was intensified whenever there was any perceived threat by the prisoners and this new level subsequently became the baseline from which further hostility and harassment would begin. The most hostile guards on each shift moved spontaneously into the leadership roles of giving orders and deciding on punishments. They became role models whose behaviour was emulated by other members of the shift. Despite minimal contact between the three separate guard shifts and nearly 16 hours a day spent away from the prison, the absolute level of aggression as well as more subtle and «creative” forms of aggression manifested, increased in a spiralling function. Not to be tough and arrogant was to be seen as a sign of weakness by the guards and even those “good” guards who did not get as drawn into the power syndrome as the others respected the implicit norm of never contradicting or even interfering with an action of a more hostile guard on their shift.

After the first day of the study, practically all prisoner’s rights (even such things as the time and conditions of sleeping and eating) came to be redefined by the guards as “privileges” which were to be earned for obedient behaviour. Constructive activities such as watching movies or reading (previously planned and suggested by the experimenters) were arbitrarily cancelled until further notice by the guards–and were subsequently never allowed. “Reward”, then became granting approval for prisoners to eat, sleep, go to the toilet, talk, smoke a cigarette, wear glasses or the temporary diminution of harassment. One wonders about the conceptual nature of “positive” reinforcement when subjects are in such conditions of deprivation, and the extent to which even minimally acceptable conditions become rewarding when experienced in the context of such an impoverished environment.

We might also question whether there are meaningful non-violent alternatives as models for behaviour modification in real prisons. In a world where men are either powerful or powerless, everyone learns to despise the lack of power in others and in oneself. It seems to us, that prisoners learn to admire power for its own sake–power becoming the ultimate reward. Real prisoners soon learn the means to gain power whether through ingratiation, informing, sexual control of other prisoners or development of powerful cliques. When they are released from prison, it is unlikely they will ever want to feel so powerless again and will take action to establish and assert a sense of power.

The pathological prisoner syndrome

Various coping strategies were employed by our prisoners as they begaoto react to their perceived loss of personal identity and the arbitrary control of their lives. At first they exhibited disbelief at the total invasion of their privacy, constant surveillance and atmosphere of oppression in which they were living. Their next response was rebellion, first by the use of direct force, and later with subtle divisive tactics designed to foster distrust among the prisoners. They then tried to work within the system by setting up an elected grievance committee. When that collective action failed to produce meaningful changes in their existence, individual self-interests emerged. The breakdown in prisoner cohesion was the start of social disintegration which gave rise not only to feelings of isolation but deprecation of other prisoners as well. As noted before, half the prisoners coped with the prison situation by becoming extremely disturbed emotionally–as a passive way of demanding attention and help. Others became excessively obedient in trying to be “good” prisoners. They sided with the guards against a solitary fellow prisoner who coped with his situation by refusing to eat. Instead of supporting this final and major act of rebellion, the prisoners treated him as a trouble-maker who deserved to be punished for his disobedience. It is likely that the negative self-regard among the prisoners noted by the end of the study was the product of their coming to believe that the continued hostility toward all of them was justified because they “deserved it” [9]. As the days wore on, the model prisoner reaction was one of passivity, dependence and flattened affect.

Let us briefly consider some of the relevant processes involved in bringing about these reactions.

Loss of personal identity. Identity is, for most people, conferred by social recognition of one’s uniqueness, and established through one’s name, dress, appearance, behaviour style and history. Living among strangers who do not know your name or history (who refer to you only by number), dressed in a uniform exactly like all other prisoners, not wanting to call attention to one’s self because of the unpredictable consequences it might provoke–all led to a weakening of self identity among the prisoners. As they began to lose initiative and emotional responsivity, while acting ever more compliantly, indeed, the prisoners became deindividuated not only to the guards and the observers, but also to themselves.

Arbitrary control. On post-experimental questionnaires, the most frequently mentioned aversive aspect of the prison experience was that of being subjugated to the apparently arbitrary, capricious decisions and rules of the guards. A question by a prisoner as often elicited derogation and aggression as it did a rational answer. Smiling at a joke could be punished in the same way that failing to smile might be. An individual acting in defiance of the rules could bring punishment to innocent cell partners (who became, in effect, “mutually yoked controls»), to himself, or to all.

As the environment became more unpredictable, and previously learned assumptions about a just and orderly world were no longer functional, prisoners ceased to initiate any action. They moved about on orders and when in their cells rarely engaged in any purposeful activity. Their zombie-like reaction was the functional equivalent of the learned helplessness phenomenon reported by Seligman and Groves [10]. Since their behaviour did not seem to have any contingent relationship to environmental consequences, the prisoners essentially gave up and stopped behaving. Thus the subjective magnitude of aversiveness was manipulated by the guards not in terms of physical punishment but rather by controlling the psychological dimension of environmental predictability [11].

Dependency and emasculation. The network of dependency relations established by the guards not only promoted helplessness in the prisoners but served to emasculate them as well. The arbitrary control by the guards put the prisoners at their mercy for even the daily, commonplace functions like going to the toilet. To do so, required publicly obtained permission (not always granted) and then a personal escort to the toilet while blindfolded and handcuffed. The same was true for many other activities ordinarily practised spontaneously without thought, such as lighting up a cigarette, reading a novel, writing a letter, drinking a glass of water or brushing one’s teeth. These were all privileged activities requiring permission and necessitating a prior show of good behaviour. These low level dependencies engendered a regressive orientation in the prisoners. Their dependency was defined in terms of the extent of the domain of control over all aspects of their lives which they allowed other individuals (the guards and prison staff) to exercise.

As in real prisons, the assertive, independent, aggressive nature of male prisoners posed a threat which was overcome by a variety of tactics. The prisoner uniforms resembled smocks or dresses, which made them look silly and enabled the guards to refer to them as “sissies” or “girls”. Wearing these uniforms without any underclothes forced the prisoners to move and sit in unfamiliar, feminine postures. Any sign of individual rebellion was labelled as indicative of “incorrigibility” and resulted in loss of privileges, solitary confinement, humiliation or punishment of cell mates. Physically smaller guards were able to induce stronger prisoners to act foolishly and obediently. Prisoners were encouraged to belittle each other publicly during the counts. These and other tactics all served to engender in the prisoners a lessened sense of their masculinity (as defined by their external culture). It follows then, that although the prisoners usually outnumbered the guards during line-ups and counts (nine v. three) there never was an attempt to directly overpower them. (Interestingly, after the study was terminated, the prisoners expressed the belief that the basis for assignment to guard and prisoner groups was physical size. They perceived the guards were “bigger”, when, in fact, there was no difference in average height or weight between these randomly determined groups.)

In conclusion, we believe this demonstration reveals new dimensions in the social psychology of imprisonment worth pursuing in future research. In addition, this research provides a paradigm and information base for studying alternatives to existing guard training, as well as for questioning the basic operating principles on which penal institutions rest. If our mock prison could generate the extent of pathology it did in such a short time, then the punishment of being imprisoned in a real prison does not “fit the crime” fo


most prisoners–indeed, it far exceeds it! Moreover, since prisoners and guards are locked into a dynamic, symbiotic relationship which is destructive to their human nature, guards are also society’s prisoners.

Shortly after our study was terminated, the indiscriminate killings at San Quentin and Attica occurred, emphasising the urgency for prison reforms that recognise the dignity and humanity of both prisoners and guards who are constantly forced into one of the most intimate and potentially deadly encounters known to man.


This research was funded by an ONR grant: N00014-67-A-Oil2-0041 to Professor Philip G. Zimbardo.

The ideas expressed in this paper are those of the authors and do not imply endorsement of ONR or any sponsoring agency. We wish to extend our thanks and appreciation for the contributions to this reasearch by David Jaffe who served as “warden” and pre-tested some of the variables in the mock prison situation. In addition, Greg White provided invaluable assistance during the data reduction phase of this study. Many others (most notably Carolyn Burkhart, Susie Phillips and Kathy Rosenfeld), helped at various stages of the experiment, with the construction of the prison, prisoner arrest, interviewing, testing, and data analysis–we extend our sincere thanks to each of these collaborators. Finally, we wish especially to thank Carlo Prescott, our prison consultant, whose personal experience gave us invaluable insights into the nature of imprisonment.

Statement by Christina Maslach, Ph.D.

Professor of the Graduate School and Former Vice Provost for Undergraduate Education, University of California-Berkeley

There were many people who witnessed what took place during the Stanford Prison Experiment (SPE)– not just the participants and research team, but various visitors, including academic colleagues, invited experts with experience of working in prisons, family members, and staff who helped with the prison logistics (e.g., delivering food for the prisoners). All of these people observed the operation of the study, so I find it surprising that recent critics are claiming that such a long-running, widely observed, and well-documented study was built on fakery and lies. The critics are very far away from the truth.

For the record, I was one of the visitors. I was not a part of the research team that designed and carried out the study, and I did not know much about it beforehand. I had just completed my doctorate at Stanford, and I was in the midst of packing up everything for my move to the University of California, Berkeley, where I would begin my new job as an assistant professor of psychology. I did know everyone on the research team, and I was also in a dating relationship with Phil. So, when he asked me to help out near the end of the study, particularly with doing interviews, I agreed to visit the prison and see what was going on.

As I have stated and written elsewhere, I was really upset by the dehumanizing way the guards were treating the prisoners as they took them to the bathroom before going to bed in their cells. Indeed, I felt sick to my stomach and could not watch the line of hooded prisoners being led down the hall. As it turned out, I was the only one who felt that way. None of the other researchers or visitors had the same reaction, and some even teased me a bit, questioning my ability as a psychologist if I could not look at these unusual scenes of human behavior. Phil was especially annoyed by my unwillingness to be interested in what I was seeing, and it was that moment that came as a great shock to me. How could Phil and I have such totally different viewpoints of the world around us, and such different responses and judgments of it? I had thought that I knew this man well, and now he was like a stranger to me–it felt as though we were on opposite sides of a deep chasm, and that chasm was too serious an issue for me to simply ignore it or run away from it. I had to figure out what was going wrong, so I confronted Phil, and we had a major argument that went on for several hours. Eventually, Phil came to agree with what I was saying, and he decided to end the study in the morning.

That morning I did do interviews, but now they were “exit” interviews. I was also a participant-observer of all the group meetings and debriefings that took place throughout the day. The ex-prisoners were really happy and celebrating that the study was over; the ex-guards were quiet and seemed unsure of what would happen next. When everyone came together as a single group, there were lots of interesting moments and questions for each other. For example, some of the former prisoners were convinced that the people selected to be guards were the biggest and tallest ones. However, when we had everyone stand up, it became evident that there was no significant difference in average height between prisoners and guards. The difference was in the mindset, not in objective measures.

There was a lot of deep debate about whether people’ s behavior was a true reflection of their “real self” or just the enactment of a role. For example, if a guard tripped a prisoner while leading him down the hall to the bathroom, was he just behaving like a guard, or was he going beyond the role because he was actually a mean person? Clearly, different people had varying perceptions on all of this, and the conversations about these issues–and how people came to terms with each other about them–were amazing to hear. In no way was any of this experience “false.” It was deeply meaningful to everyone who went through it, and I think the critics do a serious disservice to all of the participants by denying what occurred and what was learned because of it.

Finally, I want to note that there are many other people who have witnessed similar experiences in their own lives; that similarity is why people continue to resonate so much to the SPE. It has been 47 years since the study took place, and during that time I ‘ve been told hundreds of stories which began, “something like that happened to me.” These stories have come from students in my classes, neighbors, co-workers, friends, and even strangers. Some stories have taken place in jails or hospitals, others in schools or the military, and still others in family settings. And the details range from prisoner-like experiences of being bullied or abused to guard-like experiences of going too far and treating someone else badly.

There are important lessons to be learned from the SPE, and in my view, the study makes an important and enduring contribution to our understanding of human behavior.

June 26, 2018



The mind is a formidable jailer

The following article is the result of a research project at Stanford University conducted by Philip G. Zimbardo, professor of psychology, and three graduate-student colleagues: W. Curtis Banks, Craig Haney and David Jaffe.

In prison, those things withheld from and denied to the prisoner become precisely what he wants most of all.

–Eldridge Cleaver, “Soul on Ice.”

Our sense of power is more vivid when we break a man’s spirit than when we win his heart.

–Eric Hoffer, “The Passionate State of Mind.”

Every prison that men build is built with bricks of shame,/ and bound with bars lest Christ should see how men their brothers maim.

–Oscar Wilde, “The Ballad of Reading Gaol.”

Wherever anyone is against his will that is to him a prison.    –Epictetus, “Discourses.»

The quiet of a summer morning in Palo Alto, Calif.,- was shattered fey a screeching squad car siren as police swept through the city picking up college students in a surprise mass arrest Each suspect was charged with a felony, warned of his constitutional rights, spread-eagled against the car,

searched, handcuffed and carted off in the back seat of the squad car to the police station for booking.

After fingerprinting and the preparation of identification forms for his “jacket” (central information file), each prisoner was left isolated in a detention cell to wonder what he had done to get himself into this mess. After a while, he was blindfolded and transported to the “Stanford County Prison.» Here he began the process of becoming a prisoner–stripped naked, skin-searched, deloused and issued a uniform, bedding, soap and towel.

The warden offered an impromptu welcome:

“As you probaby know, I’m your warden. All of you have shown that you are unable to function outside in the real world for one reason or another–that somehow you lack the responsibility of good citizens of this great country. We of this prison, your correctional staff, are going to help you learn what your responsibilities as citizens of this country are. Here are the rides. Sometime in the near future there will be a copy of the rules posted in each of the cells. We expect you to know them and to be able to recite them by number. If you follow all of these rules and keep your hands clean, repent for your misdeeds and show a proper attitude of penitence, you and I will get along just fine.”

There followed a reading of the 16 basic rules of prisoner conduct. “Rule Number One: Prisoners


must remain silent during rest periods, after lights are out, during meals and whenever they are outside the prison yard. Two: Prisoners must eat at mealtimes and only at mealtimes. Three: Prisoners must not move, tamper, deface or damage walls, ceilings, windows, doors, or other prison property.

. . . Seven: Prisoners must address each other by their ID number only. Eight: Prisoners must address the guards as ‘Mr. Correctional Officer.’. . . Sixteen: Failure to obey any of the above rules may result in punishment.”

By late afternoon these youthful “first offenders” sat in dazed silence on the cots in their barren cells trying to make sense of the events that had transformed their lives so dramatically.

If the police arrests and processing were executed with customary detachment, however, there were some things that didn’t fit. For these men were now part of a very unusual kind of prison, an experimental mock prison, created by social psychologists to study the effects of imprisonment upon volunteer research subjects. When we planned our two-week-long simulation of prison life, we sought to understand more about the process by which people called “prisoners” lose their liberty, civil rights, independence and privacy, while those called “guards” gain social power by accepting the responsibility for controlling and managing the lives of their dependent charges.

Why didn’t we pursue this research in a real prison? First, prison systems are fortresses of secrecy, closed to impartial observation, and thereby immune to critical analysis frbm anyone not already part of the correctional authority. Second, in any real prison, it is impossible to separate what each individual brings into the prison from what the prison brings out in each person.

We populated our mock prison with a homogeneous group of people who could be considered “normal-average» on the basis of clinical interviews and personality tests. Our participants (10 prisoners and 11 guards) were selected from more than 75 volunteers recruited through ads in the city and campus newspapers. The applicants were mostly college students from all over the United States and Canada who happened to be in the Stanford area during the summer and were attracted by the lure of earning $15 a day for participating in a study of prison life. We selected only those judged to be emotionally stable, physically healthy, mature, law-abiding citizens.

This sample of average, middle-class, Caucasian, college-age males (plus one Oriental student) was arbitrarily divided by the flip of a coin. Half were randomly assigned to play the role of guards, the others of prisoners. There were no measurable differences between the guards and the prisoners at the start of the experiment Although initially warned that as prisoners their privacy and other civil rights would be violated and that they might be subjected to harassment, every subject was completely confident of his ability to endure whatever the prison had to offer for the full two-week experimental period. Each subject unhesitatingly agreed to give his “informed consent” to participate.

The prison was constructed in the basement of Stanford University’s psychology building, which was deserted after the end of the summer-school session. A long corridor was converted into the prison «yard” by partitioning off both ends. Three small laboratory rooms opening onto this corridor were made into cells by installing metal barred doors and replacing existing furniture with cots, three to a cell. Adjacent offices were refurnished as guards’ quarters, interview-testing rooms and bedrooms for the “warden” (Jaffe) and the “superintendent” (Zimbardo). A concealed video camera and hidden microphones recorded much of the activity and conversation of guards and prisoners. The physical environment was one in which prisoners could always be observed by the staff, the only exception being when they were secluded in solitary confinement (a small, dark storage closet, labeled “The Hole”).

Our mock prison represented an attempt to simulate the psychological state of imprisonment in certain ways. We based our experiment on an in- depth analysis of the prison situation, developed after hundreds of hours of discussion with Carlo Prescott (our ex-con consultant), parole officers and correctional personnel, and after reviewing much of the existing literature on prisons and concentration camps.

“Real” prisoners typically report feeling powerless, arbitrarily controlled, dependent, frustrated, hopeless, anonymous, dehumanized and emasculated. It was not possible, pragmatically or ethically, to create such chronic states in volunteer subjects who realize that they are in an experiment for only a short time. Racism, physical brutality, indefinite confinement and enforced homosexuality were not features of our mock prison. But we did try to reproduce those elements of the prison experience that seemed most fundamental.

We promoted anonymity by seeking to minimize each prisoner’s sense of uniqueness and prior identity. The prisoners wore smocks and nylon stocking caps; they had to use their ID numbers; their personal effects were removed and they were housed in barren cells. All of this made them appear similar to each other and indistinguishable to observers. Their smocks, which were like dresses, were wom without underearments. causing the prisoners to be restrained in their physical actions and to move in ways that were more feminine than masculine. The prisoners were forced to obtain permission from the guard for routine and simple activities such as writing letters, smoking a cigarette or even going to the toilet; this elicited from them a childlike dependency.

Their quarters, though clean and neat, were small, stark and without esthetic appeal. The lack of windows resulted in poor air circulation, and persistent odors arose from the unwashed bodies of the prisoners. After 10 P.M. lockup, toilet privileges were denied, so prisoners who had to relieve themselves would have to urinate and defecate in buckets provided by the guards. Sometimes the guards refused permission to have them cleaned out, and this made the prison smell.

Above all, “real” prisons are machines for playing tricks with the human conception of time. In our windowless prison, the prisoners often did not even know whether it was day or night. A few hours after falling asleep, they were roused by shrill whistles for their “count.” The ostensible purpose of the count was to provide a public test of the prisoners’ knowledge of the rules and of their ID numbers. But more important, the count, which occurred at least once on each of the three different guard shifts, provided a regular occasion for the guards to relate to the prisoners. Over the course of the study, the duration of the counts was spontaneously increased by the guards from their initial perfunctory 10 minutes to a seemingly interminable several hours. During these confrontations, guards who were bored could find ways to amuse themselves, ridiculing recalcitrant prisoners, enforcing arbitrary rules and openly exaggerating any dissension among the prisoners.

The guards were also “deindividualized”: They wore identical khaki uniforms and silver reflector sunglasses that made eye contact with them impossible. Their symbols of power were billy clubs, whistles, handcuffs and the keys to the cells and the




“main gate.”; ^though our guards received, no formal training from us jn how to be guards, for the. most part they moved with apparent ease into their roles. The media had already provided them with ample models of prison guards to emulate.

Because we were as interested in the guards’ behavior as in the prisoners’, they were given considerable latitude to improvise and to develop strategies and tactics of prisoner management. Our guards were told that they must maintain ‘‘law and order” in this prison, that they were responsible foF handling any trouble that might break out, and they were cautioned about the seriousness and potential dangers of the situation they were about to enter. Surprisingly, in most prison systems, “real” guards are not given much more psychological preparation or adequate training than this for what is one of the most complex, demanding and dangerous jobs our society has to offer. They are expected to learn how to adjust to their new employment mostly from on-the-job experience, and from contacts with the “old bulls” during a survival-of-the-fittest orientation period. According to an orientation manual for correctional officers at .San Quentin, “the only way you really get to know San Quentin is through experience and time. Some of us take more time and must go through more experiences than others to accomplish this; some really never do get there.”

You cannot be a prisoner if no one «will be your guard, and you cannot be a prison guard if no ope takes you or your prison seriously. Therefore, over time a perverted symbiotic relationship developed. As the guards became more aggressive, prisoners became more passive; assertion by the guards led to dependency in the prisoners; self-aggrandizement was met with self-deprecation, authority with helplessness, and the counterpart of the guards’ sense of mastery and control was the depression and hopelessness witnessed in the prisoners. As these differences in behavior, mood and perception became more evident to all, the need for the now “righteously” power- ful guards to rule the obviously inferior and powerless inmates became a sufficient reason to support almost any further indignity of man against man:

Guard K: “During the inspection, I went to cell 2 to mess up a bed which the prisoner had made and he grabbed me, screaming that he had just made it, and he wasn’t going to let me mess it up. He grabbed my throat, and although he was laughing I was pretty scared. … I lashed out with my stick and hit him in the chin (although not very hard), and when I freed myself I became angry. I wanted to get back in the cell and have a go with him, since he attacked me when I was not ready.”

Guard M: “I was surprised at myself … I made them call each other names and clean the toilets out with their bare hands. I practically considered the prisoners cattle, and I kept thinking: T have to watch out for them in case they try something.’”

Guard A: “I was tired of seeing the prisoners in their rags and smelling the strong odors of their bodies that filled the cells. I watched them tear at each other on orders given by us. They didn’t see it as an experiment. It was real and they were fighting to keep their identity. But we were always there to show them who was boss.”

Power takes as ingratitude the writhing of its victims.

–Rabindranath Tagore, “Stray Birds.”

Because the first day passed without incident, we were surprised and totally unpre- pared for the rebellion that broke out on the morning of the second day. The prisoners removed their stocking caps, ripped off their numbers and barricaded themselves inside the cells by putting their beds against the doors. What should we do? The guards were very much upset because the prisoners also began to taunt and curse them to their faces. When the morning shift of guards came on, they were upset at the night shift who, they felt, must have been too permissive and too lenient. The guards had to handle the rebellion themselves, and what they did was startling to behold.

At first they insisted that reinforcements be called in. The two guards who were waiting on stand-by call at home came in, and the night shift of guards voluntarily remained on duty (without extra pay) to bolster the morning shift. The guards met and decided to treat force with force. They got a fire extinguisher that shot a stream of skin-chilling carbon dioxide and forced the prisoners away from the doors; they broke into each cell, stripped the prisoners Saked, took the beds out, forced the prisoners who were the ringleaders into solitary confinement and generally began to harass and Intimidate the prisoners.

After crushing the riot, the guards decided to head off further unrest by creating a privileged cell for those who were “good prisoners” and then, without explanation, switching some of the troublemakers into it and some

of the good prisoners out mw the other cells. The prisoner ringleaders could not trust these new cellmates because they had not joined in the riot and might even be “snitches.” The prisoners never again acted in unity against the system. One of the leaders of the prisoner revolt later confided:

Voice from the real world

Professor Zimbardo’s research into the psychology of prisons has put him in touch with many real prisoners. Here are excerpts from a convict’s letter.

I was recently released from “solitary confinement” after being held therein for ■37 months. A silent system was imposed upon me and to even “whisper” to a man in the next cell resulted in being beaten by guards, sprayed with chemical Mace, blackjacked, stomped and thrown into a “strip-cell” naked to sleep on a concrete floor without bedding, covering, wash basin, or even a toilet The floor served as toilet and bed, and even there the “silent system” was enforced. To let a “moan” escape your lips because of the pain and discomfort . . . resulted in another beating. I spent not days, but months there during my 37 months in solitary … I have filed every writ possible against the administrative acts of brutality. The state courts have alt denied the petition. Because of my refusal to let “things die down” and “forget” all that happened during my 37 months in solitary … I am the most hated prisoner in ——

Penitentiary, and called a “hard-core incorrigible.”

Professor Zimbardo, maybe I am an incorrigible, but if true, it’s because I would rather die than to accept being treated less than a human being. I have never complained of my prison sentence as being unjustified except through legal means of appeals. I have never put a knife on a guard’s throat and demanded my release. I know that thieves must be punished and I don’t think I will be a thief when I am released. No, I’m not rehabilitated. It’s just that I no longer think of becoming wealthy by stealing. I now only think of “killing.” Killing those who have beaten me and treated me as if I were a dog. I hope and pray for the sake of my own soul and future life of freedom, that I am able to overcome the bitterness and hatred which eats daily at my soul, but I know to overcome it will not be easy.

“If we had gotten together then, I think we could have taken over the place. But when I saw the revolt wasn’t

Professor Zimbardо, ш “superintendent” of the mock prison meets with a group of*)pn*oners” in numbered smocfes.

other as people, they only extended the oppressiveness and reality of their life as prisoners. For the most part, each prisoner observed his fellow prisoners allowing the guards to humiliate them, acting like compliant sheep, carrying out mindless orders with total obedience and even being cursed by fellow prisoners (at a guard’s command). Under such circumstances, how could-a prisoner have respect for his fellows, or any self-respect for what he obviously was becoming in the eyes of all those evaluating him?

working, I decided to toe the line. Everyone settled into the same pattern. From then on, we were really controlled by the guards.”

It was after this episode that the guards really began to demonstrate their inventiveness in the application of arbitrary power. They made the prisoners obey petty, meaningless and often inconsistent rules, forced them to engage in tedious, useless work, such as moving cartons back and forth between closets and picking thorns out of their blankets for hours on end. (The guards had previously dragged the blankets through thorny bushes to create this disagreeable task.) Not only did the prisoners have to sing songs or laugh or refrain from smiling on command; they were also encouraged to curse and vilify each other publicly during some of the counts. They sounded off their numbers endlessly and were repeatedly made to do pushups, on occasion with a guard stepping on them or a prisoner sitting on them.

Slowly the prisoners became resigned to their fate and even behaved in ways that actually helped to justify their dehumanizing treatment at the hands of the guards. . Analysis of the tape-recorded private conversations between prisoners and of remarks made by them to interviewers revealed that fully half could be classified as nonsupportive of other prisoners. More dramatic, 85 per cent of the evaluative statements by prisoners about their fellow prisoners were uncomplimentary and deprecating.

This should be taken in the context of an even more surprising result. What do you imagine the prisoners talked

about when they were alone in their cells with each other, given a temporary respite, from the continual harassment and surveillance by the guards? Girl friends, career plans, hobbies or politics?

No, their concerns were almost exclusively riveted to prison topics. Their monitored conversations revealed that only 10 per cent of the time was devoted to “outside” topics, while 90 per cent of the time they discussed escape plans, the awful food, grievances or ingratiation tactics to use with specific guards in order to get a cigarette, permission to go to the toilet or some other favor. Their obsession with these immediate survival concerns made talk about the past and future an idle luxury.

And this was not a minor point. So long as the prisoners did not get to know each

Life is the art of being well deceived; and in order that the deception may succeed it must be habitual and uninterrupted.

–William Hazlitt, «On Pedantry,” in “The Round Table.”

The combination of realism and symbolism in this experiment had fused to create a vivid illusion of imprisonment. The illusion merged inextricably with reality for at least some of the time for every individual in the situation. It was remarkable how readily we all slipped into our roles, temporarily gave up our identities and allowed these assigned roles and the social forces in the situation to guide, shape and eventually to control our freedom of thought and action.

But precisely where does one’s “identity” end and one’s “role” begin? When the private self and the public role behavior clash, what direction will attempts to impose consistency take? Consider the reactions of the parents, relatives and friends of the prisoners who visited their forlorn sons, brothers and lovers during two scheduled visitors’ hours. They were taught in short order that they were our guests, allowed the privilege of visiting only by complying with the regulations of the institution. They had to register, were made to wait half an hour, were told that only two visitors could see any one prisoner; the total visiting time was cut from an hour to only 10 minutes, they had to be under the surveillance of a guard, and before any parents could enter the visiting area, they had to discuss their son’s case with the warden. Of course they complained about these arbitrary rules, but their conditioned, middle-class reaction was to work within the system to appeal privately to the superintendent to make conditions better for their prisoners.

In less than 36 hours, we were forced to release prisoner 8612 because of extreme depression, disorganized thinking, uncontrollable crying and fits of rage. We did so reluctantly because we believed he was trying to “con” us–it was unimaginable that a volunteer prisoner in a mock prison could legitimately be suffering and disturbed to that extent. But then on each of the next three days another prisoner reacted with similar anxiety symptoms, and we were forced to terminate them, too. In a fifth case, a prisoner was released after developing a psychosomatic rash over his entire body (triggered by rejection of his parole appeal by the mock parole board). These men were simply unable to make an adequate adjustment to prison life. Those who endured the prison experience to the end could be distinguished from those who broke down and were released early in only one dimension–authoritarianism. On a psychological test designed to reveal a person’s authoritarianism, those prisoners who had the highest scores were best able to fuhction in this authoritarian prison environment.

If the authoritarian situation became a serious matter for the prisoners, it became even more serious–and sinister–for the guards. Typically, the guards insulted the prisoners, threatened them, were physically aggressive, used instruments (night sticks, fire extinguishers, etc.) to keep the prisoners in line and referred to them in impersonal, anonymous, deprecating ways; “Hey, you,” or “You [obscenity], 5401, come here.” From the first to the last day, there was a significant increase in the guards’ use of most of these domineering, abusive tactics.


Everyone and everything in the    was defined

by pdwTh To be a guard who jjid. not take advan- tage f^^^his institutionally sanction^ use of power was to appe&^weak,” “out of it,” “wiretfe*ip..by the prisoners,” or simply,ia. deviant from the established norms of appropriate guard behavior. Using Erich Fromm’s definition of sadism, as “the wish for absolute control over another living being,” all of the mock guards at one time or another during this study behaved sadistically toward the prisoners. Many of them reported –in their diaries, on critical- incident report forms and during post-experimental interviews–being delighted in the new-found power and control they exercised and sorry to see it relinquished at the end of the study.

Some of the guards reacted to the situation in the extreme and behaved with great hostility and cruelty in the forms of degradation they invented for the prisoners. But others were kinder; they occasionally did little favors for the prisoners, were reluctant to punish them, and avoided situations where prisoners were being harassed. The torment experienced by one of these good guards is obvious in his perceptive analysis of what it felt like to be responded to as a “guard”: “What made the experience most depressing for me was the fact that we were continually called upon to act in a way that just was contrary to what I really feel inside. I don’t feel like I’m the type of person that would be a guard, just constantly giving out . . . and forcing people to do things, and pushing and lying –it just didn’t seem like me, and to continually keep up and put on a face like that is just really one of the most oppressive things you can do. It’s almost like a prison that you create yourself–you get into it, and it becomes almost the definition you make of yourself, it almost becomes like walls, and you want to break out and you want just to be able to tell everyone that ‘this isn’t really me at all, and I’m not the person that’s confined in there–I’m a person who wants to get but and show you that I am free, and I do have my own will, and I’m not the sadistic type of person that enjoys this kind of thing.’ ”

Still, the behavior of these good guards seemed more motivated by a desire to be liked by everyone in the system than by a concern for the inmates’ welfare. No guard ever intervened in any direct way on behalf of the prisoners, ever interfered with the orders of the crudest guards or ever openly complained about the subhuman quality of life that characterized this prison.

Perhaps the most devastating impact of the more hostile guards was their creation of a capricious, arbitrary environment. Over time the prisoners began to react passively. When our mock prisoners asked questions, they got answers about half the time, • but the rest of the time they were insulted and punished–and it was not possible for them to predict which would be the outcome. As they began to “toe the line,” they stopped resisting, questioning and, indeed, almost ceased responding altogether. There was a general decrease in all categories of response as they learned the safest strategy to use in an unpredictable, threatening environment from which there is no physical escape– do nothing, except what is required. Act not, want not, feel not and you will not get into trouble in prisonlike situations.

And the only way to really make it with the bosses [in Texas prisons] is to withdraw into yourself, both mentally and physically–literally making yourself as small as possible. It’s another way they dehumanize you. They want you to make no waves in prison and they want you to make no waves when you get out.

–Mike Middleton, ex-con. The Christian Science Monitor.

Can it really be, you wonder, that intelligent, educated volunteers could have lost sight of the reality that they were merely acting a part in an elaborate game that would eventually end? There are many indications not only that they did, but that, in addition, so did we and so did other apparently sensible, responsible adults.

Prisoner 819, who had gone into a rage followed by an uncontrollable crying fit, was about to be prematurely released from the prison when a guard lined up the prisoners and had them chant in unison, “819 is a bad prisoner. Because of what 819 did to prison property we all must suffer. 819 is a bad prisoner.” Over and over again. When we realized 819 might be overhearing this, we rushed into the room where 819 was supposed to be resting, only to find him in tears, prepared to go back into the prison because he could not leave as long as the others thought he was a “bad prisoner.” Sick as he felt, he had to prove to them he was not a “bad” prisoner. He had to be persuaded that he was not a prisoner at all, that the others were also just students, that this was just an experiment and not a prison and the prison staff


were only research psychologists. A report from the warden notes, “While I believe that it was necessary for staff [me] to enact the warden role, at least some of the time, I am startled by the ease with which I could turn off my sensitivity and concern for others for ‘a good cause.’ ”

Consider our overreaction to the rumor of a mass escape plot that one of the guards claimed to have overheard. It went as follows:         Prisoner

8612, previously released for emotional disturbance, was only faking. He was going to round up a bunch of his friends, and they would storm the prison right after visiting hours. Instead of collecting data on the pattern of rumor transmission, we made plans to maintain the security of our institution.. After putting a confederate informer into the cell 8612 had occupied to get specific information about the escape plans, the superintendent went back to the Palo Alto Police Department to request transfer of our prisoners to the old city jail. His impassioned plea was only turned down at the last minute when




In this real prison scene, shaved heads reduce each prisoner’s sense of individuality. In the mock prison, inmates wore nylon stocking caps to simulate baldness.


the problem of insurance and city liability for our prisoners was raised by a city official. Angered at this lack of cooperation, the staff formulated another plan. Our jail. was dismantled, the prisoners, chained and blindfolded, were carted off to a remote storage room. When the conspirators arrived, they would be told the study was over, their friends had been sent home, there was nothing left to liberate. After they left, we would redouble the security features of our prison making any future escape attempts futile. We even planned to lure exprisoner 8612 back on some pretext and imprison him again, because he had been released on false pretenses! The rumor turned out to be just that – a full day had passed in which we collected little or no data, worked incredibly hard to tear down and then rebuild our prison. Our reaction, however, was as much one of relief and joy as of exhaustion and frustration.

When a former prison chaplain was invited to talk with the prisoners (the grievance committee had requested church services), he puzzled everyon&rby disparaging each inmate f6r not having taken any constructive action in order to gef released. “Don’t you know you must have a lawyer in order to get bail, or to appeal the charges against you?” Several of them accepted his invitation to contact their parents in order to secure the services of an attorney. The next night one of the parents stopped at the superintendent’s office before visiting time and handed him the name and phone number of her cousin who was a public defender. She said that a priest had called her and suggested the need for a lawyer’s services! We called the lawyer. He came, interviewed the prisoners, discussed sources of bail money and promised to return again after the weekend.

But perhaps the most felling account of the insidious development of this new reality, of the gradual Kafka- esque metamorphosis of good into evil, appears in excerpts from the diary of one of the guards, Guard A:

Prior to start of experiment: “As I am a pacifist and nonaggressive individual I cannot see a time when I might guard and/or maltreat other living things.”

After an orientation meeting: “Buying uniforms at the end of the meeting confirms the gamelike atmosphere of this thing. I doubt whether many of us share the expectations of ‘seriousness’ that the experimenters seem to have.”

First Day: “Feel sure that the prisoners will make fun or my appearance and I evolve my first basic strategy – mainly not to smile at anything they say or do which would be admitting it’l^all only a game. … At cell ‘3 I stop and setting my voice hard and low say to 5486, ‘What are you smiling atr «’Nothing, Mr. Correctional Officer.’ ‘Well, see that you don’t.’ (As I walk off I feel stupid.)”

Second Day: “5704 asked for a cigarette and I ignored him – because I am a non- smoker and could not empathize. . . . Meanwhile since I was feeling empathetic towards 1037, I determined not to talk with him . . . after we had count and lights out [Guard D] and I held a loud conversation about going home to our girl friends and what we were going to do to them.”

Third Day (preparing for the first visitors’ night)’: “After warning the prisoners not to make any complaints unless they wanted the visit terminated fast, we finally brought in the first parents. I made sure I was one of the guards on the yard, because this was my first chance for the type of manipulative power that I really like–being a very noticed figure with almost complete control over what is said or not. While the parents and prisoners sat in chairs, I sat on the end of the table dangling my feet and contradicting anything I felt like. This was the first part of the experiment I was really enjoying. . . . 817 is being obnoxious and bears watching.”

Fourth Day: “. . . The psychologist rebukes me for handcuffing and blindfolding a prisoner before leaving the [counseling] office, and I resentfully reply that it is both necessary security and my business anyway.”

Fifth Day: “I harass ‘Saige’ who continues to stubbornly overrespond to all commands.

pressures of a «prison environment.” If this could happen in so short a time, without the excesses that are possible in real prisons, and if it could happen to the “cream-of-the- crop of American youth,” then one can only shudder to imagine what society is doing both to the actual guards and prisoners who are at this very moment participating in that unnatural “social experiment.” The pathology observed in this study cannot be reasonably attributed to pre-existing personality differences of the subjects, that option being eliminated by our selection procedures and random assignment. Rather, the subjects’ abnormal social and personal reactions are best seen as a product of their transaction with an environment that supported the behavior that would be pathological in other settings, but was “appropriate” in this prison. Had we observed comparable reactions in a real prison, the psychia-

  • undoubtedly would have able to attribute any ner’s behavior to charac- efects or personality maladjustment, while critics of the prison system would have been quick to label the guards as “psychopathici’f-This tendency to locate the source of behavior disorders inside a j particular person or group un- i derestimates the power of situational forces.

! Our colleague, David Rosen- j han, has very convincingly shown that once a sane person (pretending to be insane) gets labeled as insane and committed to a mental hospital, it is the label that is the reality which is treated and not the person. This dehumanizing tendency to respond to other people according to socially determined labels and often arbitrarily assigned roles is also apparent in a recent

We’ve traveled too far, and our momentum has taken over; we move idly towards eternity, without possibility of reprieve pr hope of explanation.

–Tom Stoppard, “Rosencrantz and Guildenstem Are Dead.”

Was it worth all the suffering just to prove what everyone knows–that some people are sadistic, others weak and prisons are not beds of roses? If that is all we demonstrated in this research, then it was certainly not worth the anguish. We believe there are many significant implications to be derived from this experience, only a few of which can be suggested here.

The potential social value of this study derives precisely from the fact that normal, healthy, educated young men could be so radically transformed under the institutional

1 let the food slide down his face. I didn’t believe it was me doing it. I hated myself for making him eat but I hated him more for not eating.” Sixth Day: “The experiment is over. I feel elated but am shocked to find some other guards disappointed somewhat because of the loss of money and some because they are enjoying themselves.”

We were no longer dealing with an intellectual exercise in which a hypothesis was being evaluated in the dispassionate manner dictated by the canons of the scientific method. We were caught up in the passion of the present, the suffering, the need to control people, not variables, the escalation of power and all of the unexpected things that were erupting around and within us. We had to end this experiment. So our planned two-week simulation was aborted after only six (was it only six?) days and nights.

1 have singled him out for special abuse both because he begs for it and because I simply don’t like him. The real trouble starts at dinner. The new prisoner (416) refuses to eat his sausage … we throw him into the Hole ordering him to hold sausages . in each hand. We have a crisis of authority; this rebellious conduct potentially undermines the complete control we have over the others. We decide to play upon prisoner solidarity and tell the new one that all the others will be deprived of visitors if he does not eat his dinner. … I walk by and slam my stick into the Hole door. … I am very angry at this prisoner for causing discomfort and trouble for the others. I decided to force- feed him, but he wouldn’t eat.


“mock hospital” study designed by Norma Jean Orlando to extend the ideas in our research.

Personnel from the staff of Elgin State Hospital in Illinois role-played either mental patients or staff in a weekend simulation on a ward in the hospital. The mock mental patients soon displayed behavior indistinguishable from that we usually associate with the chronic pathological syndromes of actual mental patients: incessant pacing, uncontrollable weeping, depression, hostility, fights, stealing from each other, complaining. Many of the “mock staff” took advantage of their power to act in ways comparable to our mock guards by dehumanizing their powerless victims.

During a series of encounter debriefing sessions immediately after our experiment, we all had an opportunity to vent our strong feelings and tore- fleet upon the moral and ethi- > cal issues each of us faced, and we considered how we might react more morally in future “real-life” analogues to this situation. Year-long follow-ups with our subjects via questionnaires, personal interviews and group reunions indicate that their mental anguish was transient and situation- ally specific, but the self- knowledge gained has persisted.

For the most disturbing implication of our research comes from the parallels between what occurred in that basbment mock prison and daily experiences in our own lives–and we presume yours. The physical institution of prison is but a concrete and steel metaphor for the existence of more pervasive, albeit less obvious, prisons of the mind that all of us daily

create, populate and perpetuate. We speak here of the prfeons of racism, sexism, deS|air, shyness, “neurotic haijg-ups” and the like. The social convention of marriage, as $he example, becomes for many couples a state of imprisonment in which one partner agrees to be prisoner or guard, forcing or allowing the other to play the reciprocal role–invariably without making the contract explicit.

To what extent do we allow ourselves to become imprisoned by docilely accepting the roles others assign us or, indeed, choose to remain prisoners because being passive and dependent frees us from the need to act and be responsible for our actions? The prison of fear constructed in the delusions of the paranoid is no less confining or less real than the cell that every shy person erects to limit his own freedom in anxious anticipation of being ridiculed and rejected by his guards–often guards of his own making. ■


A mock prisoner enjoys the amenities of a “privilege cell» set up by guards to increase their psychological authority.






The Psychology of Power and Evil:

All Power to the Person? To the Situation? To the System?

Philip G. Zimbardo, Psychology Department, Stanford University1


To understand anti-social behavior by individuals, which includes violence, torture and terrorism, I endorse a greater reliance on situational variables and processes than has been traditional in psychology. The dominant dispositional orientation, embedded in a psychology of individualism, focuses on internal factors that people bring into various situations, such as genetic, personality, character, and pathological risk factors. While this perspective is obviously important to appreciating the integrity of individual functioning, it is vital to add an appreciation of the extent to which human actions may come under situational influences that can be quite powerful. Those influences have not been fully recognized within psychology or society in trying to explain unusual or “evil” behaviors, such as that of the abuses of Iraqi prisoners by United States military police guards at Abu Ghraib Prison. How one understands the root causes of such behaviors then impacts treatment and prevention strategies. This view has both influenced and been informed by a body of social psychological research and theory. The situationist approach is to the dispositional as public health models of disease are to medical models. It follows basic principles of Lewinian theory that propel situational determinants of behavior to a foreground well beyond being merely extenuating background circumstances. Unique to this situationist approach is using experimental laboratory and field research as demonstrations of real world phenomena that other approaches only analyze verbally or rely on archival or correlational data for answers.

The basic paradigm to be presented illustrates the relative ease with which «ordinary,» good men and women are induced into behaving in “evil ways” by turning on or off one or another social situational variable. The plan of this chapter is to outline some of my laboratory and field studies on deindividuation, aggression, vandalism, and the Stanford Prison Experiment, along with a process analysis of Milgram’s obedience studies, and Bandura’s analysis of “moral disengagement.” This body of research demonstrates the under-recognized power of social situations to alter the mental representations and behavior of individuals, groups and nations. I explore briefly extreme instances of “evil” behavior for their dispositional or situational foundations — torturers, death squad violence workers and terrorist suicide-bombers. Finally, we turn to consider the opposite side of the coin, by focusing on the positive virtues of heroism and ways in which society and educational systems can promote pro-social values.

1 . This chapter is a modified version of my PowerPoint presentation for the DHS course, The Psychology of Terrorism, organized by the faculty of the National Center on the Psychology of Terrorism. It relies on my recent chapter, The social psychology of good and evil: Understanding our capacity for Stanfkindness and cruelty. Published in The Social Psychology of Good and Evil. Arthur Miller (Ed.). (pp. 21-50). New York: Guilford.


Evil is intentionally behaving — or causing others to act — in ways that demean, dehumanize, harm, destroy, or kill innocent people. This behaviorally-focused definition makes an agent of agency responsible for purposeful, motivated actions that have a range of negative consequences to other people. It excludes accidental or unintended harmful outcomes, as well as the broader, generic forms of institutional evil, such as poverty, prejudice or destruction of the environment by agents of corporate greed. But it does include corporate responsibility for marketing and selling products with known disease-causing, death-dealing properties, such as cigarette manufacturers, or other drug dealers. It also extends beyond the proximal agent of aggression, as studied in research on interpersonal violence, to encompass those in distal positions of authority whose orders or plans are carried out by functionaries. This is true of military commanders and national leaders, such as Hilter, Stalin, Mao, Pol Pot, Idi Amin, Saddam Hussein, and other tyrants for their complicity in creating political systems of destruction in their own nations and in the world.

The same human mind that creates the most beautiful works of art and extraordinary marvels of technology is equally responsible for the perversion of its own perfection. This most dynamic organ in the universe has been a seemingly endless source for creating ever more vile torture chambers and instruments of horror in earlier centuries, the “bestial machinery” unleashed on Chinese citizens by Japanese soldiers in their rape of Nanking (see Iris Chang, 1997), and the recent demonstration of “creative evil” of the destruction of the World Trade Center by turning commercial airlines into weapons of mass destruction. How can the unimaginable become so readily imagined?

My concern centers around how good, ordinary people can be recruited, induced, seduced into behaving in ways that could be classified as evil. In contrast to the traditional approach of trying to identify «evil people» to account for the evil in our midst, I will focus on trying to outline some of the central conditions that are involved in the transformation of good, or average, people into perpetrators of evil. In the experimental research to be described, “evil” really amounts to the research participant acting in ways that harm others in that same setting.

Locating Evil Within Particular People: The Rush to the Dispositional

«Who is responsible for evil in the world, given that there is an all-powerful, omniscient God who is also all-Good?» That conundrum began the intellectual scaffolding of the Inquisition in the 16th and 17th centuries in Europe. As revealed in Malleus Maleficarum, the handbook of the German Inquisitors from the Roman Catholic Church, the inquiry concluded that the Devil was the source of all evil. However, these theologians argued the Devil works his evil through intermediaries, lesser demons and of course, human witches. So the hunt for evil focused on those marginalized people who looked or acted differently from ordinary people, who might qualify under rigorous examination of conscience, and torture, to expose them as witches, and then put to death. They were mostly women who could readily be exploited without sources of defense, especially when they had resources that could be confiscated. An analysis of this legacy of institutionalized violence against women is detailed by historian Anne Barstow (1994) in Witchcraze. Paradoxically, this early effort of the Inquisition to understand the origins of evil and develop interventions to cope with evil instead created new forms of evil that fulfilled all facets of my definition. But it exemplifies the notion of simplifying the complex problem of widespread evil by identifying individuals who might be the guilty parties, and then making them pay for their evil deeds.

The authoritarian personality syndrome was developed by a team of psychologists (Adorno, Frenkel-Brunswick, Levinson, & Sanford, 1950) after WW 11, trying to make sense of the Holocaust and the broad appeal of national Fascism and Hitler. Their dispositional bias led them to focus on a set of personality factors underlying the fascist mentality. However, what they overlooked were the host of processes operating at political, economic, societal, and historical levels of analysis to influence and direct so many millions of individuals into a constrained behavioral channel of hating Jews and admiring the apparent strength of their dictator.

This tendency to explain observed behavior by reference to dispositions, while ignoring or minimizing the impact of situational variables has been termed the Fundamental Attribution Error (FAE) by my colleague, Lee Ross (1977). We are all subject to this dual bias of overutilizing dispositional analyses and under-utilizing situational explanations when faced with ambiguous causal scenarios we want to understand. We succumb to this effect because so much of our education, social and professional training, and societal agencies are geared toward a focus on individual, dispositional orientations. Dispositional analyses are a central operating feature of cultures that are based on individualistic rather than collectivist values (see Triandis, 1994). Thus, it is individuals who get praise and fame and wealth for achievement and are honored for their uniqueness, but it is also individuals who are blamed for the ills of society. Our legal systems, medical, educational and religious systems all are founded on principles of individualism.

Dispositional analyses of anti-social, or non-normative, behaviors always include strategies for behavior modification to make the deviant individuals fit better by education or therapy, or to exclude them from society by imprisonment, exile or execution. However, locating evil within selected individuals or groups always has the ‘social virtue’ of rendering society or its institutions as blameless. The focus on people as causes for evil then exonerates societal structures and political decision-making for contributing to the more fundamental circumstances that create poverty, marginal existence for some citizens, racism, sexism and elitism.

Most of us take comfort in the illusion that there is an impermeable line separately those bad people from us good people. Its rigid boundaries constrain good from becoming bad, or bad from ever being reversed into fostering good outcomes. That view also means we have little interest in understanding the motivations and circumstances that contributed to how those bad people first came to engage in evil behavior. I find it good to remind myself of the geo-political analysis of the Russian novelist, Alexander Solzhenitsyn, a victim of persecution by the Soviet KGB, that the line between good and evil lies in the center of every human heart.

The Transformation of The Good Dr. Jekyll into the Evil Mr. Hyde

I am sure that most readers were as fascinated as I was with the Robert Louis Stevenson’s tale of the behavioral transformation of good Dr. Jekyll into the murderous Mr. Hyde. That dramatic change required some strange chemical formula. I wondered, along with others, if such a transformation could be accomplished without drugs. Were there other means that people could use to change human behavior is such extreme fashion? I would later discover that social psychology had recipes for such transformations.

It has been my mission as a psychologist to understand better how virtually anyone could be recruited to engage in evil deeds that deprive other human beings of their dignity, humanity and life. So I have always begun my analyses of all sorts of anti-social behavior, even the most horrendous instances of evil, with the question: “What could make me do the same thing?” And further, I wonder what were the set of situational and structural circumstances that empowered others — maybe similar to me — to engage in deeds that they too once thought were alien to their nature. This first led me to set aside any false pride that, “I am not that kind of person,” once I

acknowledged any circumstances under which I might become that kind of person. Then it led me to want to investigate a range of conditions under which ordinary people like me could do things that violated the traditional sense of morality.

I argue that the human mind is so marvelous that it can adapt to virtually any known environmental circumstance in order to survive, to create, and to destroy as necessary. We are not born with tendencies toward good or evil, but with mental templates to do either, more gloriously than ever before, or more devastatingly than ever experienced before–as the World Trade Center disaster of September 11, 2001 revealed. It is only through the recognition that we are all part of the human condition, that humility takes precedence over unfounded pride in acknowledging our vulnerability to situational forces. Although the research I will present next has been fascinated with identifying the variables and processes by which ordinary people can be seduced or initiated into engaging in evil deeds, it is apparent that the time has come to better understand how to enable ordinary people to resist such forces and how to promote pro-social behavior. If we want to develop mechanisms for combating transformations of good people into evil perpetrators, it is essential to learn first the causal mechanisms underlying those behavior changes. We need to discover the range of identifiable variables involved in the complex processes that influence so many of us to do so much bad, to commit so much evil throughout the globe. Space does not allow me to review the many contributions of my colleagues to these issues, thus I recommend their works to concerned readers. Please see the breadth of ideas that have been presented by social psychological colleagues, Baumeister, 1997; Darley, 1992; Staub, 1989, and Waller, 2002, for starters.

Blind Obedience to Authority: The Milgram Investigations

Stanley Milgram (1974) developed an ingenious research procedure to demonstrate the extent to which situational forces could overwhelm individual will to resist. He ‘shocked the world’ with his unexpected finding of extremely high rates of compliance to the demands of an authority figure to continue shocking an innocent victim to the maximum possible level (also see Blass, 2004). He found that about 67% of research participants “went all the way» up to the top shock level of 450 volts in shocking another person that were supposedly helping. Milgram’s study revealed that ordinary American citizens could so easily be led to engage in “electrocuting a nice stranger,” as the Nazis had been led to murder Jews.

After this initial demonstration with Yale College students, Milgram went on to conduct 18 experimental variations on more than a thousand subjects from a variety of backgrounds, ages, both genders and all educational levels. In each of these studies he varied one social psychological variable and observed its impact on the extent of obedience to the unjust authority’s pressure to continue to shock the “learner-victim.” The data told the story of the extreme pliability of human nature: Almost everyone could be totally obedient or almost everyone could resist authority pressures. It all depended on the situational variables he introduced in each study. He was able to demonstrate that compliance rates could soar to 90 percent of people who delivered the maximum 450 volts to the Learner-Victim, or could be reduced to less than 10 percent of total obedience — by introducing one variable into the compliance recipe.

Want maximum obedience? Provide social models of compliance by having participants observe peers behaving obediently. Want people to resist authority pressures? Provide social models of peers who rebelled. Interestingly, almost no one shocked the Learner-Victim when he actually asked to be shocked. They refused authority pressure when the target person acted like a masochist who wanted to be shocked. In each of the other variations on this diverse range of ordinary American citizens from two towns in Connecticut, low, medium, or high levels of

compliant obedience could be readily elicited as if one were simply turning a Human Nature Dial.

What is the expected base rate of such obedience in the Milgram setting according to experts on human nature? When forty psychiatrists were given the basic description of this experiment, their average estimate of the percent of United States citizens who would give the full 450 volts was only one percent! Only sadists would engage in such sadistic behavior, they believed. These experts on human behavior were totally wrong because they ignored the situational determinants of behavior in the procedural description of the experiment. Their training in psychiatry had led them to overly rely on the dispositional perspective that comes from their professional training. This is a strong instance of the operation of the fundamental attribution error in action.

In a sense what was also unique about the Milgram paradigm was its quantification of evil in terms of the shock level each person chose or resisted on the shock generator that allegedly delivered shocks to a mild- mannered confederate who played the role of the pupil or learner while the subject enacted the teacher role. (No one ever actually got shocked, but the participants believed they were actually delivering ever more painful shocks with each increasing shock button).

Ten Steps to Creating Evil Traps for Good People

Let’s outline some of the procedures in this research paradigm that seduced many ordinary citizens to engage in this apparently harmful behavior. In doing so, I want to draw parallels to compliance strategies used by «influence professionals» in real-world settings, such as salespeople, cult recruiters, and our national leaders (see Cialdini, 2001).

Among the influence principles to be extracted from Milgram’s paradigm for getting ordinary people to do things they originally believe they would not are the following ten:

1)              Offering an Ideology so that a big lie provides justification for any means to be used to achieve the seemingly desirable, essential goal. Presenting an acceptable justification, or rationale, for engaging in the undesirable action, such as wanting to help people improve their memory by judicious use of punishment strategies. In experiments it is known as the “cover story” because it is a cover-up for the procedures that follow which might not make sense on their own. The real world equivalent is known as an “ideology.” Most nations rely on the same ideology of “threats to national security” before going to war or suppressing dissident political opposition. It is a convenient familiar ideological theme that fascist governments and military juntas have used to destroy socialist or communist opposition. When citizens fear that their national security is being threatened they are willing to surrender their basic freedoms when the government offers them that exchange. In the Unites States, the fear of the threat to national security posed by terrorists has led too many citizens to accept torture of prisoners as a necessary tactics for securing information that could prevent further attacks. That reasoning contributed to the background of the abuses by the American guards at Abu Ghraib prison. See the provocative analysis by Susan Fiske and her colleagues on why ordinary people torture enemy prisoners (Fiske, Harris, & Cuddy, 2004).

2)              Arranging some form of contractual obligation, verbal or written, to enact the behavior.

3)              Giving participants meaningful roles to play (teacher, student) that carry with them previously learned positive values and response scripts.

4)              Presenting basic rules to be followed, that seem to make sense prior to their actual use, but then can be arbitrarily used to justify mindless compliance. Make the rules vague and change them as necessary.

5)              Altering the semantics of the act, the actor, and the action, (from hurting victims to helping learners by punishing them)–replace reality with desirable rhetoric.

6)              Creating opportunities for diffusion of responsibility for negative outcomes; others will be responsible, or it won’t be evident that the actor will be held liable.

7)              Starting the path toward the ultimate evil act with a small, insignificant first step (only 15 volts).

8)              Having successively increasing steps on the pathway be gradual, so that they are hardly noticed as being different from one’s most recent prior action. (By increasing each level of aggression in gradual steps of only 30 volts, no new level of harm seemed like a noticeable difference to the Milgram participants.)

9)              Changing the nature of the influence authority from initially “Just” and reasonable to “Unjust” and demanding, even irrational, elicits initial compliance and later confusion, but continued obedience.

10)              Making the «exit costs» high, and making the process of exiting difficult by allowing usual forms of verbal dissent (that make people feel good about themselves), while insisting on behavioral compliance (“I know you are not that kind of person, just keep doing as I tell you.”)

Such procedures are utilized across varied influence situations where those in authority want others to do their bidding, but know that few would engage in the «end game» final solution without first being properly prepared psychologically to do the «unthinkable.»

On Being Anonymous: Deindividuation and Destructiveness

The idea for my doing research that utilized anonymity as an independent variable in the study of aggressive behavior came not from a psychological theory but rather from a novel. William Golding’s (1962) Noble prize-winning novel of the transformation of good British Christian choir boys into murderous little beasts centers on how the change in one’s external physical appearance leads to a change in one’s mental state and behavior. Painting one’s self, changing one’s outward appearance, made it possible for some boys to disinhibit previously restrained impulses to kill a pig for food. Once that alien deed of killing another creature was accomplished, then they could continue on to kill with pleasure, both animals and people alike. Is it psychologically valid that external appearance could impact internal and behavioral processes? That is the question I answered with a set of experiments and field studies on the psychology of deindividuation (Zimbardo, 1970).

The basic procedure involved having young women deliver a series of painful electric shocks to each of two other young women whom they could see and hear in a one-way mirror before them. Half were randomly assigned to a condition of anonymity, or deindividuation, half to one of uniqueness, or individuation. The four college student subjects in each deindividuation group had their appearance concealed by hoods, their names replaced by numbers and treated as a group not as individuals. The comparison group consisted of those in an individuation treatment who wore name tags and made to feel unique. Both were in four-woman groups and asked to make the same responses of shocking each of two woman «victims» over the course of 20 trials. The cover story was that these “victims” were trying to be creative under stress, so the job of our subjects was to stress them by administering painful electric shocks while I, as the experimenter, gave them the creativity test. Unlike the Milgram paradigm, there was no authority insisting on their aggressive behavior because I was in the adjacent room, seen in the two-way observation mirror by the subjects along with each of the two alleged women in the creativity study. The dependent variable was the duration of shock administered, not shock level intensity.

The results were clear: Women in the deindividuation condition delivered twice as much shock to both victims as did the women in the individuated comparison condition. Moreover, they shocked both victims, the one previously rated as pleasant and the other unpleasant victim, more over the course of the 20 trials, while the individuated subjects shocked the pleasant woman less over time than they did the unpleasant one. (Again, no shocks were actually administered, although all participants believed they had delivered shocks to each of the two women, who acted out being hurt by the shocks. One important conclusion flows from this research and its various replications and extensions, some using military personnel from the Belgian army. Anything that makes someone feel anonymous, as if no one knows who they are, reduces a sense of accountability and creates the potential for that person to act in evil ways — if and when the situation gives permission for violence.

Anonymous Children Become Aggressive at Halloween

We know that people also mask themselves for hedonistic pleasures, as at Carnival rituals in many Catholic countries. Children in America and some other countries put on masks and costumes for Halloween parties. My former student, Scott Fraser, (1974) arranged for elementary school children to go to a special, experimental Halloween party given by their teacher. There were many games to play and for each game won, tokens were earned that could be exchanged for gifts at the end of the party. Half the games were non-aggressive in nature, and half were confrontations between two children in order to reach the goal. The experimental design was a within group, A-B-A format; no costumes (A), costumes (B), no costumes (A). Initially while the games were played the teacher said the costumes were on the way so they would start the fun while waiting. Then the costumes arrived and were worn as the games continued, and finally, the costumes were removed to go to other children in other parties, and the games went on for the third phase; each phase for about an hour.

The data are striking testimony to the power of anonymity. Aggression increased significantly as soon as the costumes were worn, more than doubling from the initial base level average. But when the costumes were removed, aggression dropped back well below initial level base rate. Equally interesting was the second result, that aggression cost the children a loss of tokens. Acting in the aggressive games took more time than the non-aggressive games and only one of two contestants could win, so overall it cost money to be aggressive, but that did not matter when the children were costumed and anonymous. The least number of tokens won was during the second, anonymity phase, where aggression was highest. A third important finding was that there was no carry-over of aggressive behavior from the high B phase level to the last A phase level, which was comparable to the initial A phase. The behavior change due to the anonymity had not created a dispositional, internal change, only an outward response change. Change the situation, voila the behavior changes in predictable fashion.

Cultural Wisdom: How to Make Warriors Kill in Battle But Not at Home

Let’s leave the laboratory and fun and games at children’s parties to the real world where these issues of anonymity and violence may take on life and death significance. Some societies go to war without having the young male warriors change their appearance, while others always include ritual transformations of appearance by painting or masking the warriors (as in Lord of the Flies). Does that change in external appearance make a difference in how warring enemies are treated? Harvard anthropologist, John Watson (1974) posed that question after reading my Nebraska Symposium chapter (Zimbardo, 1970). The Human Area Files were his data source to collect two pieces of data on societies that did or did not change appearance of warriors prior to going to war and the extent to which they killed, tortured or mutilated their victims.

The results are striking confirmation of the prediction that anonymity promotes destructive behavior–when permission is also given to behave in aggressive ways that are ordinarily prohibited. Of the 23 societies for which these two data sets were present, the majority (12 of 15, 80 %) of societies in which warriors changed their appearance were those noted as most destructive, while that was true of only one of the eight where the warriors did not change appearance before going to battle. Ninety percent of the time when victims were killed, tortured or mutilated it was by warriors who had first changed their appearance.

Thus, cultural wisdom dictates that a key ingredient in transforming ordinarily nonaggressive young men into warriors who can kill on command is to first change their external appearance. War is about old men persuading young men to harm and kill other young men like themselves in a war. It becomes easier to do so if they first change their appearance, to alter their usual external fa9ade by putting on uniforms, or masks, or painting their faces. With that anonymity in, out goes their usual internal focus of compassion and concern for others. When the war is won, the culture now dictates that their warriors return to their peaceful status — readily accomplished by removing their uniform, taking off the mask, and returning to their former external fa9ade.

Bandura’s Model of Moral Disengagement and Dehumanization

The psychological mechanisms involved in getting good people to do evil are embodied in two theoretical models, the first elaborated by me (1970) and modified by input from subsequent variants on my deindividuation conceptions, notably by Diener (1980). The second is Bandura’s model of moral disengagement (1988) that specifies the conditions under which anyone can be led to act immorally, even those who usually ascribe to high levels of morality.

Bandura’s model outlines how it is possible to morally disengage from destructive conduct by using a set of cognitive mechanisms that alter: a) one’s perception of the reprehensible conduct (engaging in moral justifications, making palliative comparisons, using euphemistic labeling for one’s conduct): b) one’s sense of the detrimental effects of that conduct (minimizing, ignoring, or misconstruing the consequences); c) one’s sense of responsibility for the link between reprehensible conduct and their detrimental effects (displacing or diffusing responsibility), and d) one’s view of the victim (dehumanizing him or her, and attributing the blame for the outcome to the victim).

Bandura and his students (Bandura, Underwood, and Fromson, 1975) designed a powerful experiment that is an elegantly simple demonstration of the power of dehumanizing labels. It reveals how easy it is to induce intelligent college students to accept a dehumanizing label of other people and then to act aggressively based on that stereotyped term. A group of four participants were led to believe they were overhearing the research assistant tell the experimenter that the students from another college were present to start the study in which they were to deliver electric shocks of varying intensity to them (allegedly as part of a group problemsolving study). In one of the three randomly assigned conditions, the subjects overheard the assistant say to the experimenter that the other students seemed «nice.” In a second condition, they heard that the other students seemed like «animals,” while for a third group the assistant did not label the students in the other group of college students.

The dependent variable of shock intensity clearly reflected this situational manipulation. The experimental subjects gave most shock to those labeled in the dehumanizing way as «animals,» and their shock level increased linearly over the ten trials. Those labeled «nice» were given the least shock, while the unlabelled group was in the middle of these two extremes. Thus, a single word — “animals” — was sufficient to incite intelligent college students to treat others so labeled as if they knew them enough that that they deserved to be harmed.

What is also of interest is a close examination of the graphed data shows that on the first trial there is no difference across the three experimental treatments in the level of shock administered, but with each successive opportunity, the shock levels diverge. Those shocking the so-called “animals” shock them more and more over time, a result comparable to the escalating shock level of the deindividuated female students in my earlier study. That rise in aggressive responding over time, with practice, or with experience, illustrates a self-reinforcing effect of aggressive or violent responding — it is increasingly pleasurable. Perhaps the pleasure is not so much in inflicting pain to others as in the sense of power and control one feels in such a situation of dominance.

On the plus side in this study, that arbitrary labeling also resulted in others being treated with greater respect if someone in authority labeled them positively. Compared with the neutral, no information condition, those perceived as “nice” were least harmed. There is an important message here about the power or words, labels, rhetoric, of stereotyped labeling, to be used for good or evil.

Suspension of The Usual Cognitive Controls Guiding Moral Action

What my model adds to the mix of what is needed to get good people to engage in evil deeds is a focus on the role of cognitive controls that usually guide behavior in socially desirable and personally acceptable ways. It can be accomplished by knocking out these control processes, blocking them, minimizing them, or reorienting them. Doing so, suspends conscience, selfawareness, sense of personal responsibility, obligation, commitment, liability, morality and analyses in terms of costs/ benefits of given actions. The two general strategies for accomplishing this objective are: reducing cues of social accountability of the actor (no one knows who I am, nor cares to), and reducing concerns for self-evaluation by the actor. The first cuts out concerns for social evaluation, for social approval, and does so by making the actor feel anonymous. It works when one is functioning in an environment that conveys anonymity and diffuses personal responsibility across others in the situation. The second strategy stops selfmonitoring and consistency monitoring by relying on tactics that alter one’s state of consciousness (through drugs, arousing strong emotions, hyper-intense actions, getting into an expanded present-time orientation where there is no concern for past or future), and by projecting responsibility outward onto others.

My research on deindividuation and that of other social psychologists (see Prentice-Dunn & Rogers, 1983) differs from the paradigm in Milgram’s studies in that there is no authority figure present urging the subject to obey. Rather, the situation is created in such a way that subjects act in accordance to paths made available to them, without thinking through the meaning or consequences of those actions. Their actions are not cognitively guided as they are typically, but directed by the actions of others in proximity to them, or by their strongly aroused emotional states, and by situationally available cues, such as the presence of weapons (see Berkowitz, 1993).

The Evils of Vandalism Spread Through Anonymous Environments

It is possible for certain environments to convey a sense of anonymity on those who live or behave in their midst. Where that happens, the people living there do not have a sense of community. Vandalism and graffiti may be interpreted as an individual’s attempt for public notoriety in a society that deindividuates them, that gives them no legitimate outlets for personal recognition. Vandalism maybe an attempt to have an impact on one’s environment through destruction when doing so constructively does not seem possible.

I did a simple field study to demonstrate the ecological differences between a places where anonymity ruled versus a sense of community dominated the scene. I abandoned used, but good condition cars in the Bronx, New York City and in Palo Alto, California, one block away from New York University and Stanford University, respectively. License plates were removed and hoods raised slightly — to serve as ethological «releaser cues» for the potential vandals’ attack behavior. It worked swiftly in the Bronx, as we watched and filmed from a vantage point across the street. Within 10 minutes of officially beginning this study, the first vandals surfaced. This parade of vandals continued for two days, when there was nothing left of value to strip, then the vandals began destroying the remains. In 48 hours we recorded 23 separate destructive contacts by individual or groups, who either took something from the abandoned vehicle or did something to wreck it. Curiously, only one of these episodes involved adolescents, the rest were by adults, many well dressed and many driving cars, so that they might qualify as at least lower middle-class. Anonymity can make brazen vandals of us all. But what about the fate of the abandoned car in Palo Alto? Our time-lapse film revealed that no one vandalized any part of the car over a 5-day period. When we removed the car, three local residents called the police to say that an abandoned car was being stolen (the local police had been notified of our field study). That is one definition of “community,” where people care about what happens on their turf even to the person or property of strangers. I think they do so based in part on their reciprocal assumption that others in that neighborhood would also care about them.

I now feel that any environmental, societal conditions that contribute to making some members of society feel that they are anonymous, that no one knows who they are, that no one recognizes their individuality and thus their humanity, makes them potential assassins and vandals, a danger to my person and my property — and yours (Zimbardo, 1976).

Curiously, this little field demonstration which was publicized in Time Magazine (Feb. 28,1969, Diary of a Vandalized Car) was the only empirical research presented in support of a controversial theory about crime, known as “Broken Windows Theory”. Political scientist James Q. Wilson and criminologist George Kelling outlined their novel theory about the twin causes of crime in a popular Atlantic Monthly article (March, 1982 ). Crime is a product of individual criminals and situational conditions of public disorder. When people see abandoned cars in the streets, graffiti everywhere and broken windows not covered, it is a sign that no one really cares about that neighborhood. That perception of public disorder or disarray then lowers inhibitions against further destructive or criminal actions of those who are not ordinarily criminal. Their solution to crime: remove abandoned cars, paint out graffiti and fix broken windows. When that advice was followed in New York City, crime rates dropped significantly the next year. I was pleased that this little study could have such big indirect effects.

The Hostile Imagination Created by Faces of the Enemy

We need to add a few more operational principles to our arsenal of weapons that trigger evil acts among men and women who are ordinarily good people. To do so we need to rise above the research focusing on individual actors and look to nation-states. We can learn about some of these principles by considering how nations prepare their young men to engage in deadly wars and prepare citizens to support the risks of going to war, especially a war of aggression. This difficult transformation is accomplished by a special form of cognitive conditioning. Images of the «Enemy» are created by national propaganda to prepare the minds of soldiers and citizens to hate those who fit the new category of your enemy. This mental conditioning is a soldier’s most potent weapon, without it, he could probably never fire his weapon to kill another young man in the cross-hairs of his gun sight. A fascinating account of how this «hostile imagination» is created in the minds of soldiers and their families is presented in Faces of the Enemy by Sam Keen (1991; 2004), and his companion DVD.

Archetypes of the enemy are created by propaganda fashioned by the governments of most nations against those judged to be the dangerous «them,» «outsiders,» «enemies.» These visual images create a consensual societal paranoia that is focused on the enemy who would do harm to the women, children, homes, and god of the soldier’s nation, way of life, and so forth. Keen’s analysis of this propaganda on a world-wide scale reveals that there are a select number of categories utilized by «homo hostilis» to invent an evil enemy in the minds of good members of righteous tribes. The enemy is: aggressor; faceless; rapist; godless; barbarian; greedy; criminal; torturer; death; a dehumanized animal, or just an abstraction. Finally, there is the enemy as worthy, heroic opponent to be crushed in “mortal combat” — as in the video game of the same name.

Can Ordinary Old Men Become Murderers Overnight?

One of the clearest illustrations of my fundamental theme of how ordinary people can be transformed into engaging in evil deeds that are alien to their past history and to their moral development comes from the analysis of British historian, Christopher Browning. He recounts in Ordinary Men: Reserve Police Battalion 101 and the Final Solution in Poland (1993) that in March, 1942 about 80 percent of all victims of the Holocaust were still alive, but a mere 11 months later about 80 percent were dead. In this short period of time, the Endlosung (Hitler’s ‘Final Solution’) was energized by means of an intense wave of mass mobile murder squads in Poland. This genocide required mobilization of a large-scale killing machine at the same time as able-bodied German soldiers were needed on the collapsing Russian front. Since most Polish Jews lived in small towns and not the large cities, the question that Browning raised about the German high command was «where had they found the manpower during this pivotal year of the war for such an astounding logistical achievement in mass murder?» (p. xvi).

His answer came from archives of Nazi war crimes, in the form of the activities of Reserve Battalion 101, a unit of about 500 men from Hamburg, Germany. They were elderly, family men too old to be drafted into the army, from working-class and lower middle-class backgrounds, with no military police experience, just raw recruits sent to Poland without warning of, or any training in, their secret mission — the total extermination of all Jews living in the remote villages of Poland. In just 4 months they had shot to death at point blank range at least 38,000 Jews and had another 45,000 deported to the concentration camp at Treblinka.

Initially, their commander told them that this was a difficult mission which must be obeyed by the battalion. However, he added that any individual could refuse to execute these men, women and children. Records indicate that at first about half the men refused and let the other police reservists engage in the mass murder. But over time, social modeling processes took their toll, as did any guilt-induced persuasion by those reservists who had been doing the shooting. By the end of their journey up to 90 percent of the men in Battalion 101 were involved in the shootings, even proudly taking photographs of their up-close and personal killing of Jews. Like the photos of the guards at Abu Ghraib prison, these policemen put themselves in their “trophy photos” as proud killers of the Jewish menace.

Browning makes clear that there was no special selection of these men, only that they were as «ordinary» as can be imagined — until they were put into a situation in which they had “official” permission and encouragement to act sadistically and brutishly against those arbitrarily labeled as the “enemy.” He also compares the underlying mechanism operating in that far off land at that distant time to both the psychological processes at work in the Milgram research and the Stanford Prison Experiment.

Educating Hatred and Destructive Imaginations

The second broad class of operational principles by which otherwise good people can be recruited into evil is through education/ socialization processes that are sanctioned by the government in power, enacted within school programs, and supported by parents and teachers. A prime example is the way in which German children in the 1930’s and 40’s were systematically indoctrinated to hate Jews, to make them the all-purpose enemy of the new German nation.

Space limitations do not allow full documentation of this process, but I will include several examples of one way in which governments are responsible for sanctioning evil.

In Germany, as the Nazi party rose to power in 1933, no target of Nazification took higher priority than the re-education of Germany’s youth. Hitler wrote, “I will have no intellectual training. Knowledge is ruin to my young men. A violently active, dominating, brutal youth — that is what I am after.» (The New Order, 1989, pp. 101-2). To teach the youth about geography and race, special primers were created and ordered to be read starting in the first grade of elementary school (see Brooks, 1989). These «hate primers» were brightly colored comic books that contrasted the beautiful blond Aryans with the despicably ugly caricatured Jew. They sold in the hundreds of thousands. One was titled: Trust No Fox in the Green Meadows and No Jew on His Oath. What is most insidious about this kind of hate conditioning is that they were presented as facts to be learned and to be tested upon, or from which to practice new penmanship. In the copy of the «Trust No Fox» text that I reviewed, a series of cartoons illustrates all the ways in which Jews deceive Aryans, get rich and fat from dominating them, are lascivious, mean and without compassion for the plight of the poor and the elderly Aryans.

The final scenarios depict the retribution that Aryan children get first by expelling Jewish teachers and children from their school — so that «proper discipline and order» can now be taught, prohibiting them from community areas, like public parks, then expelling them from Germany. The sign in the cartoon reads ominously, «One-way street.» Indeed, it was a unidirectional street that led eventually to the concentration camps and crematoria that were the center piece of Hitler’s Final Solution for genocide of the Jews. Thus, this institutionalized evil was spread pervasively and insidiously by perverting education away from critical thinking exercises that open student minds to new ideas and toward thinking critically and close-mindedly about those targeted as the enemy of the people. By controlling education and the propaganda media, any national leader can produce the fantastic scenarios depicted in George Orwell’s (1981) frightening novel, 1984.

The institutionalized evil that Orwell vividly portrays in his fictional account of state dominance over individuals goes beyond the novelist’s imagination when its prophetic vision is carried into operational validity by powerful leaders of a cult, or by agencies and departments within the current national administration of the United States. I have outlined the direct parallels between the mind control strategies and tactics Orwell attributes to “The Party” and those that Reverend Jim Jones used in dominating the members of his religious/ political cult, Peoples Temple (Zimbardo, 2005). Jones orchestrated the suicide/ murders of more than 900 American citizens in the jungles of Guyana twenty five years ago, the finale of his grand experiment in institutionalized mind control. I learned from former members of this group that not only did Jones read 1984, he talked about it often and had a song commissioned by the church’s singer entitled “1984 is coming,” that everyone had to sing at some services.

The Stanford Prison Experiment: Institutional and Systemic Power to Corrupt

This research synthesized many of the processes and variables outlined earlier; those of anonymity of place and person that contribute toward creating states of deindividuation, of dehumanization of victims, of giving some actors (guards) permission to control others (prisoners), and placing it all within a unique setting (the prison) that most societies throughout the world acknowledge provides some form of institutionally approved sanctions for evil though the extreme differentials in control and power that prison foster.

In 1971, I and my students (Zimbardo, Haney, Banks, & Jaffe, 1973) designed a dramatic experiment that would extend over a two-week period to provide our research participants with sufficient time for them to become fully engaged in their experimentally assigned roles of either guards or prisoners. Having participants live in that setting day and night, if prisoners, or work there for long 8-hour shifts, if guards, would also allow sufficient time for situational norms to develop and patterns of social interaction to emerge, change and become crystallized. The second feature of this study was to ensure that all research participants would initially be as normal as possible, healthy physically and mentally, and without any history of being involved in drugs or crime or violence.

These preconditions were essential if we were to untangle the situational versus dispositional knot: What the situation elicited from this collection of similar, interchangeable young men versus what was emitted by the research participants based on the unique dispositions they brought into the experiment. The third feature of the study was the absence of any prior training in how to play the randomly assigned roles of prisoner and guard, to leave that up to each subject’s prior societal learning of the meaning of prisons and the behavioral scripts associated with the oppositional roles of prisoner and guard. The fourth feature was to make the experimental setting as close to a functional simulation of the psychology of imprisonment as possible. The details of how we went about creating a mind set comparable to that of real prisoners and guards is given in several of the articles I wrote about the study (see Zimbardo et al., 1973; Zimbardo, 1975).

Central to this mind set were issues of power and powerlessness, dominance and submission, freedom and servitude, control and rebellion, identity and anonymity, coercive rules and restrictive roles. In general, these social psychological constructs had operational reality by putting all subjects in appropriate uniforms, using assorted props (handcuffs, police clubs, whistles, signs on doors and halls), replacing corridor hall doors with prison bars to create prison cells, having no windows or clocks to tell time of day, institutional rules that removed/ substituted individual names with numbers (prisoners) or titles for staff (Mr. Correctional Officer, Warden, Superintendent), and that gave guards control-power over prisoners.

Subjects were recruited from among nearly 100 who answered our advertisements in the local city newspaper. They were given a background evaluation that consisted of a battery of five psychological tests, personal history, and in-depth interviews. The 24 who were evaluated as most normal and healthy in every respect, were randomly assigned half to the role of prisoner and half to be guards. The student-prisoners underwent a realistic surprise arrest by officers from the Palo Alto Police Department, who cooperated with our plan. The arresting officer proceeded with a formal arrest taking the “felons” to the Police Station for booking, after which each prisoner was brought to our prison in the reconstructed basement of our Psychology Department.

The prisoner’s uniform was a smock/ dress with a prison ID number. The guards wore military-style uniforms and silver-reflecting sunglasses to enhance anonymity. At any one time there were 9 prisoners on «the yard,» 3 to a cell, and 3 guards working 8-hour time shifts. Data were collected in terms of systematic video recordings, secret audio recordings of conversations

of prisoners in their cells, interviews and tests at various times during the study, post-experiment reports, and by direct, concealed observations.

For a detailed chronology and fuller account of the behavioral reactions that followed, readers are referred to the above references, and to Zimbardo, Maslach, & Haney (1999), and to our web site: www.prisonexp.org (an Italian translation has recently by made by Piero Bocchiaro).

For current purposes, let me simply assert that the negative situational forces overwhelmed the positive dispositional tendencies. The Evil situation triumphed over the Good people. Our projected 2-week experiment had to be terminated after only 6 days because of the pathology we were witnessing. Pacifist young men were behaving sadistically in their role of guards, inflicting humiliation and pain and suffering on other young men if they had the inferior human status of prisoner. Some guards even reported they were enjoying doing so. Others, who had been intelligent, healthy college students were behaving pathologically, many having «emotional breakdowns,» as in stress disorders, so extreme that five of them had to be terminated within that first week. Their fellow prisoners who adapted better to the situation were those who mindlessly followed orders, became blindly obedient to authority, who allowed the guards to dehumanize and degrade them ever more with each passing day and night. The only personality variable that had any significant predictive value was that of F-scale authoritarianism: the higher the score, the more days the prisoner survived in this totally authoritarian environment.

I terminated the experiment not only because of the escalating level of violence and degradation by the guards against the prisoners that was apparent when viewing the video tapes of their interactions, but also because I was made aware of the personal transformation that I was undergoing personally. I had become a Prison Superintendent, the second role I played in addition to that of Principal Investigator. I began to talk, walk and act like a rigid institutional authority figure more concerned about the security of “my prison” than the needs of the young men entrusted to my care as a psychological researcher. In a sense, I consider that the most profound measure of the power of this situation was the extent to which it transformed me. Finally, we had extended debriefing sessions of guards and prisoners at the end of the study, and for periodic checkups over many years. Fortunately, there were no negative lasting consequences of this powerful experience

The Evil of Inaction

“The only thing necessary for evil to triumph is for good men to do nothing,” is an important message to highlight our next section, it comes from statesman, Edmund Burke.

Our usual take on evil focuses on violent, destructive actions, but non-action can also become a form of evil, when helping, dissent and disobedience are called for. Social psychologists heeded the alarm when the infamous Kitty Genovese case made national headlines. As she was being stalked, stabbed and eventually murdered, 39 people in a housing complex heard her screams and did nothing to help. It seemed obvious that this was a prime example of the callousness of New Yorkers, as many media accounts reported. A counter to this dispositional analysis came in the form of a series of classic studies by Bibb Latane and John Darley (1970) on bystander intervention. One key finding was that people are less likely to help when they are in a group, when they perceive others are available who could help, than when those people are alone. The presence of others diffuses the sense of personal responsibility of any individual.

A powerful demonstration of the failure to help strangers in distress was staged by John Darley and Dan Batson (1974). Imagine you are a theology student on your way to deliver the sermon of the Good Samaritan in order to have it videotaped for a psychology study on effective communication. Further imagine that as you are heading from the psychology department to the video taping center, you pass a stranger huddled up in an alley in dire distress. Are there any conditions that you could conceive that would not make you stop to be that Good Samaritan? What about time press? Would it make a difference to you if you were late for your date to give that sermon? I bet you would like to believe it would not make a difference, you would stop and help no matter what the circumstances. Right? Remember you are a theology student, thinking about helping a stranger in distress, which is amply rewarded in the biblical tale.

The researchers randomly assigned students of the Princeton Theological Seminary to three conditions that varied how much time they thought they had between being given their assignment by the researchers and getting to the communication department to tape their Good Samaritan speeches. The conclusion: Don’t be in a victim in distress when people are late and in a hurry, because 90 percent of them are likely to pass you by, giving you no help at all! The more time the seminarians believed they had, the more likely they were to stop and help. So the situational variable of time press accounted for the major variance in helping, without any need to resort to dispositional explanations about theology students being callous or cynical or indifferent, as Kitty Genovese’s non-helpers were assumed to be.

In addition to perpetrators of evil, there are almost always those who know what is going on and do not intervene to help, to challenge the evil, and thereby they enable evil to persist by their inaction when they should have acted. They were the good guards in the Stanford Prison Experiment who did no harm to the prisoners but never once opposed the demeaning deeds of the bad guards. In the Abu Ghraib prison abuse case, it is clear that many people knew of the abuses, even doctors and nurses, but never intervened. (see Zimbardo, 2004).

Torturers and Executioners: Pathological Types or Situational Imperatives?

There is little debate but that the systematic torture by men of their fellow men and women represents one of the darkest sides of human nature. Surely, my colleagues and I reasoned that here was a place where dispositional evil would be manifest among these torturers who did their dirty deeds daily for years in Brazil as policemen sanctioned by the government to get confessions through torturing enemies of the state. We began by focusing on torturers, trying to understand both their psyches and the ways they were shaped by their circumstances, but we had to expand our analytical net to capture their comrades-in-arms who chose or were assigned to another branch of violence work — death squad executioners. They shared a “common enemy” — men, women, and children who, though citizens of their state, even neighbors, were declared by “the authorities” to be threats to the country’s national security. Some had to be eliminated efficiently, while others who might hold secret information had to be made to yield it up and confess to their treason.

In carrying out this mission, these torturers could rely in part on the “creative evil” embodied in torture devices and techniques that had been refined over centuries since the Inquisition by officials of The Church, and later of the National State. But they added a measure of improvisation to accommodate the particular resistances and resiliencies of the enemy standing before them, claiming innocence, refusing to acknowledge their culpability, or not being intimidated. It took time and emerging insights into human weaknesses to be exploited for these torturers to become adept at their craft, in contrast to the task of the death squads, who with hoods for anonymity, good guns, and group support, could dispatch their duty to country swiftly and impersonally. For the torturer, it could never be just business. Torture always involves a personal relationship, essential for understanding what kind of torture to employ, what intensity of torture to use on this person at this time. Wrong kind or too little: No confession. Too much, the victim dies before confessing. In either case, the torturer fails to deliver the goods. Learning to select the right kind and degree of torture that yields up the desired information, makes rewards abound, and praise flow from the superiors.

What kind of men could do such deeds, did they need to rely on sadistic impulses and a history of sociopathic life experiences to rip and tear flesh of fellow beings day in and day out for years on end? Were these violence workers a breed apart from the rest of humanity, bad seeds, bad tree trunks, bad flowers? Or, is it conceivable that they could be programmed to carry out their deplorable deeds by means of some identifiable and replicable training programs?

Could a set of external conditions, situational variables, that contributed to the making of these torturers and killers be identified? If their evil deeds were not traceable to inner defects, but rather attributable to outer forces acting on them — the political, economic, social, historical, and experiential components of their police training — then we might be able to generalize across cultures and settings those principles responsible for this remarkable transformation. Martha Huggins, Mika Haritos-Fatouros and I interviewed several dozen of these violence workers in depth and recently published a summary of our methods and findings (Huggins, Haritos- Fatouros, & Zimbardo, 2002). Mika had done a similar, earlier study of torturers trained by the Greek military junta, and our results were largely congruent with hers (Haritos-Fatouros, 2003).

Sadists are selected out of the training process by trainers because they are not controllable, get off on the pleasure of inflicting pain and thus do not sustain the focus on the goal of confession extraction. From all the evidence we could muster, these violence workers were not unusual or deviant in any way prior to practicing this new role, nor were there any persisting deviant tendencies or pathologies among any of them in the years following their work as torturers and executioners. Their transformation was entirely understandable as a consequence of: the training they were given to play this new role; group camaraderie; acceptance of the national security ideology, and of the belief in socialist-communists as enemies of their state. They were also influenced by being made to feel special, above and better than peers in public service, by the secrecy of their duties, by the constant pressure to produce desired results regardless of fatigue or personal problems. We report many detailed case studies that document the ordinariness of these men engaged in the most heinous of deeds, sanctioned by their government at that time in history, but reproducible at this time in any nation’s obsession with national security and fears of terrorism that permit suspension of basic individual freedoms.

Suicide Bombers: Mindless Fanatics or Mindful Martyrs?

Amazingly, what holds true for these violence workers is comparable to the nature of the transformation of young Palestinians from students to suicide bombers killing Israelis. Recent media accounts converge on the findings from more systematic analyses of the process of becoming a suicidal killer (see Atran, 2003; Bennet, 2003; Hoffman, 2003; Merari, 1990, 2002; Myer, 2003). There have been more than 95 suicide bombings by Palestinians against Israelis since September, 2000. Originally and most frequently the bombers were young men, but recently a half dozen women have joined the ranks of suicidal bombers. What has been declared senseless, mindless murder by those attacked and by outside observers, is anything but that to those intimately involved. It was believed that it was poor, desperate, socially isolated, illiterate young people with no career and no future who adopted this fatalistic role. That stereotype is shattered by the actual portraits of these young men and women, many are students with hopes for a better future, intelligent, attractive, connected with their family and community.

Ariel Merari, an Israeli psychologist, who has studied this phenomenon for many years, outlines the common steps on the path to these explosive deaths. Senior members of an extremist group first identify particular young people who appear to have an intense patriotic fervor based on their declarations at a public rally against Israel, or supporting some Islamic cause or

Palestinian action. These individuals are invited to discuss how serious they are in their love of their country and hatred of Israel. They are then ask to commit to being trained in how to put their curses into action. Those that do, are put into a small group of 3 to 5 similar youth who are at varying stages of progress toward becoming agents of death. They learn the tricks of the trade from elders, bomb making, disguise, selecting and timing targets. Then they make public their private commitment by making a video tape, declaring themselves to be “living martyrs” for Islam, and for the love of Allah. In one hand they hold the Koran, a rifle in the other, their headband declares their new status. This video binds them to the final deed, since it is sent home to the family of the recruit before they execute the final plan. The recruits also realize that they will not only earn a place beside Allah, their relatives will also be entitled to a high place in heaven because of their martyrdom. Then there is a sizable financial incentive that goes to their family as a gift for their sacrifice.

Their photo is emblazoned on posters that will be put on walls everywhere in the community the moment they succeed in their mission — to become inspirational models. To stifle concerns about the pain from wounds inflicted by exploding nails and other bomb parts, they are told that before the first drop of their blood touches the ground they will already be seated at the side of Allah, feeling no pain, and only pleasure. As an ultimate incentive for the young males is the promise of heavenly bliss with scores of virgins in the next life. They become heroes and heroines, modeling self-sacrifice to the next cadre of young suicide bombers.

We can see that this program utilizes a variety of social psychological and motivational principles to assist in turning collective hatred and general frenzy into a dedicated, seriously calculated program of indoctrination and training for individuals to become youthful living martyrs. It is neither mindless nor senseless, only a very different mind set and with different sensibilities than we have been used to witnessing among young adults in our country. A recent television program on female suicide bombers went so far as to describe them more akin to the girl next door then to alien fanatics. That is what is so frightening about the emergence of this new social phenomena, that so many intelligent young people can be persuaded and directed toward envisioning and welcoming their lives ending in a suicidal explosive blast.

To counteract the powerful tactics of these recruiting agents requires providing meaningful life-affirming alternatives to this next generation. It requires new national leadership that explores every negotiating strategy that could lead to peace and not to death. It requires these young people to share their values, their education, their resources, to explore their commonalities not highlight differences. The suicide, the murder, of any young person is a gash in the fabric of the human connection that we elders from every nation must unite to prevent. To encourage the sacrifice of youth for the sake of advancing ideologies of the old might be considered a form of evil from a more cosmic perspective that transcends local politics and expedient strategies.

Summing Up Before Moving On

I will end with some notions about what is involved in reversing the negative processes we have been considering by outlining some ideas about transformations into goodness. Before doing that, I want to briefly report on my role as expert witness for one of the guards involved in abusing prisoners at Abu Ghraib. In October, 2004, I testified via closed circuit television to the military trial judge in Baghdad in defense of Sgt. Ivan Frederick. I will outline some of the key issues that I raised and what I learned about situational power from that trial and from my access to the many reports of the military investigations and from my personal analysis of this young soldier.

It is a truism in psychology that personality and situations interact to generate behavior, as do cultural and societal influences. However, I have tried to show in my research over the past 30 years that situations exert more power over human actions than has been generally acknowledged by most psychologists nor recognized by the general public. However, this situationist approach continues to be dominated by the traditional dispositional perspective fueled by reliance on the individualist orientation central in Anglo-American psychology, and in our institutions of medicine, education, psychiatry, law and religion. Acknowledging the power of situational forces does not excuse the behaviors channeled by their operation. In many circles, any attempts at situational analyses are dismissed as nothing more than “excusiology.” People are assumed to always be in control of their behavior, to act from free will, and thus be personally responsible for any and all of their actions. Unless insane, individuals who do wrong should know that they are doing wrong and be punished accordingly. The situation is taken to be nothing more than a set of minimally relevant extrinsic circumstances.

The importance of adopting the situational perspective provides a knowledge base to shift attention away from simplistic «blaming of the victim,» and ineffective individualistic treatments designed to change the evil doer, toward more profound attempts to discover causal networks that should be modified if that behavior is to be prevented, circumvented or stopped. Sensitivity to situational determinants of behavior, also guides risk alerts for avoiding or changing prospective situations of vulnerability (see Richard, Bond, & Stokes-Zoota, 2003, for a quantitative summary of the effects of 100 years of social psychological research).

Several related dimensions come to the fore from the ideas outlined here. First, we should be aware that a range of apparently simple situational factors can function to impact our behavior more compellingly than seems possible. The research outlined here, along with others of my colleagues presented in this volume, points up the influential force of: role playing, rules, presence of others, emergent group norms, group identity, uniforms, anonymity, social modeling, authority presence, symbols of power, time pressures, semantic framing, stereotypical images and labels, among others.

Second, the situationist approach redefines heroism. When the majority of ordinary people can be overcome by such pressures toward compliance and conformity, the minority who resist should be considered heroic. Acknowledging the special nature of this resistance means we should learn from their example by studying how they have been able to rise above such compelling pressures. That suggestion is coupled with another that encourages the development of an essential but ignored domain of psychology — heroes and heroism.

Third, the situationist approach should, in my view, encourage us all to share a profound sense of personal humility when trying to understand “unthinkable,” “unimaginable,” “senseless” acts of evil. Instead of immediately embracing the high moral ground that distances us good folks from those bad ones, and gives short shrift to analyses of causal factors in that situation, the situational approach gives all others the benefit of “attributional charity» in knowing that any deed, for good or evil, that any human being has ever done, you and I could also do — given the same situational forces. If so, it becomes imperative to constrain our immediate moral outrage that seeks vengeance against wrong doers; instead to uncover the causal factors that could have led them in that aberrant direction.

The obvious current instantiation of these principles is the rush to the “evil” disposition to characterize terrorists and suicide bombers instead of working to understand the nature of the psychological, economic and political conditions that foster such generalized hatred of an enemy nation, including our own, that young people are willing to sacrifice their lives and murder other human beings. The “war on terrorism” can never be won solely by plans to find and destroy terrorists, since any individual, anywhere, at any time, can become an active terrorist. It is only by understanding the situational determinants of terrorism that programs can be developed to win the hearts and minds of potential terrorists away from destruction and toward creation. Not a simple task, but an essential one that requires implementation of social psychological perspectives and methods in a comprehensive, long-term plan of attitude, value and behavior change.

Understanding What Went Wrong in Abu Ghraib Prison

The abuses by American prison guards against Iraqi prisoners in that prison horrified the sensibility of people around the world, in part because it was the first time in history that such abuses were detailed in graphic photographic images. How could these men and women do such terrible things to helpless prisoners? They were condemned by the military leadership as “morally corrupt,” and by the press as a few “bad apples.” They were made to appear as exceptions to the rule of American soldiers being good young men and women proudly serving their country to preserve freedoms and advance the cause of democracy. Their images of wanton abuse humiliated the U. S. Military and was a blow to the image of the Bush administration. The initial focus of the government “to get to the bottom” of this mess clearly followed the dispositional orientation to blame pathological behavior of this kind on those with sadistic personalities and other personal pathologies.

I became an expert witness for one of those army reserve guards who was the sergeant in charge of the night shift where all the mayhem occurred. I did so in part to understand in depth the nature of that situation and the human nature of this young man, Sgt. Ivan “Chip” Frederick from reviews of all available investigations and personal contact and assessments of this soldier.

I testified on his behalf from a remotely televised setup at the U. S. Navy Base in Naples, Italy to the ongoing trial in Baghdad (which I refused to go to for fear of my safety). I will briefly outline what I learned about that person, that situation, and that system, and describe the sentence of the Military Judge.

Everything I could learn about the Abu Ghraib prison, Tier 1-A, the “soft torture” interrogation center of that prison revealed to me that virtually all of the social psychological processes operating in the Stanford Prison Experiment were at work on the night shift in that prison. In fact, one of the independent investigations (headed by James Shlesinger) specifically details the parallels between the two prisons, my mock prison and that all too real horror prison.

In addition, the guards were not trained soldiers but Army Reservists forced into this job, with no mission-specific training for such a difficult role, no supervision ever by superiors and no personal accountability. There was an emergent norm operating of support & reward for prisoner humiliation and encouragement of physical abuse to prepare the prisoners for interrogation by “softening them up.” This implicit norm was advanced by the civilian contract interrogators, the Military Police, the CIA, and on up the entire chain of military and administration command.

I asked the Judge to consider the evidence in each of these three domains prior to delivering his verdict and sentencing of this soldier:

Dispositional: Evidence for any personal pathologies, sadistic tendencies that he brought into that situation. Also evidence of any positive traits, values and personal background.

Situational: Evidence of the terrible working conditions on that night shift in that prison and in particular the nature of the situation faced by this soldier.

Systemic: Evidence of the broader conditions that spawned and sustained that situation, focusing on the nature of the leadership and the objectives of that interrogation center.

With regard to the disposition of this so-called “bad apple,” I was able to report that this soldier was totally and unequivocally “normal” on all measures that had been administered by an Army clinical psychologist (and independently validated by a civilian expert in assessment). .

There was no evidence of any psychopathology, no sadistic tendencies, The only negatives were: obsessive about orderliness, neatness, discipline, personal appearance— all found missing in action in the filth, chaos and daily disarray at Abu Ghraib! I spent a full day with Chip and his wife, conducted an in-depth four-hour interview, which led me to conclude that:

This army reservist is an All-American young man, almost stereotypically so. He is very Patriotic, the son of a West Virginia coal-miner, attends Baptist Church services regularly, hunts, fishes, plays softball, has many close friends, in a strong marriage to Martha, an African- American woman, is a loved step father to her two daughters, and was in good physical and mental status when he first arrived at this prison. He had been a good guard in low-security, small town civilian prison with 100 inmates. Chip had been in the Army Reserve for many years without any negative incidents on his record. He was a model soldier-army reservist, proud to serve in Iraq on first duty before assignment to Abu Ghraib prison, worked with children in a small village and was starting to learn Arabic to better communicate with Iraqis.

With regard to the situational conditions, the behavioral context, I was able to describe an impossible set of working conditions that bordered on the inhumane–for both guards and prisoners. First it was evident that directly comparable processes were operating in that prison as were observed in the Stanford Prison Experiment, such as — deindividuation, dehumanization, moral disengagement, social modeling, conformity pressures, anonymity of place, passive observing bystanders, power differentials, use of enforced nakedness, sexually humiliating tactics, and most of the other contextual variables that were part of SPE. Additionally, the worst abuses in both settings took place on the night shift. The working conditions faced by Chip Frederick were inhumane, consisting of 12- hr night shifts (4PM-4AM, 7 days a week, for 40 days with not a day off, then 14 days after one day off. If that incredible work demand was not bad enough, the level of exhaustion and stress was exacerbated by the chaotic conditions, unsanitary and filthy surroundings that made it smell like a putrid sewer all the time, with limited water for showering, and frequent electrical blackouts that created dangerous opportunities for prisoner attacks. This young man with no mission-specific training was put in charge of more than 300 prisoners initially but that number soon swelled to more than 1,000, along with being in charge of 12 army reserve guards, 60 Iraqi police, who often smuggled contraband to the inmates. He rarely left the prison, when off duty he slept in a cell in a different part of prison, missed breakfasts, stopped exercising, or socializing. Tier 1-A became his total reference setting.

This would qualify him for total job burnout (Maslach, 1982). If all that was not bad enough as situational forces that distorted usual judgments and decision making, there was the intense daily fear that Chip and the other guards felt because this prison was under frequent insurgency attacks, with 5 U.S. soldiers and 20 prisoners killed, and many others wounded by almost daily shelling during the time that Chip was in that job. Finally, we have to factor in his feelings of revenge against seven prisoners who had rioted in another part of the prison and were sent to Tier 1-A for “safe keeping,” and also revenge against four other Iraqi prisoners who had raped a boy prisoner. Frederick had complained to a superior officer about such dangers of housing adolescents and adults together (as well as mentally disturbed prisoners and those with Tuberculosis and other contagious diseases in the general population) but was reprimanded for not realizing it was a war and emergency measures had to be taken. Two other situational contributors to the abuse were the presence of a dominating, charismatic group leader who initiated some of the abuses and encouraged other guards to join in, and the presence of digital cameras that made it easy to document and perhaps even to facilitate the domination of these guards over their prisoners in these “trophy photos.”

With regard to systemic influences that created this bad barrel into which the army thrust Chip Frederick and the others, I turn to the summaries of all five military investigations that are available. They all point blame at “failures of leadership, lack of leadership, indifferent leadership, conflicting leadership demands. These independent investigations highlight the total absence of accountability, and the lack of supervision or oversight. The Superintendent of the prison never visited this part of her prison complex because she was told by her senior officers not to do so. That meant everyone knew there was no top-down surveillance. These reports point out the fact that none of these guards had received any mission- specific training for this demanding job with inmates who were so culturally different from them and who did not speak their language. Guards. These reports continue to document the lack of vital resources on that prison tier. There was not any medical or mental health program for these 10,000 prisoners.

A critical systemic consideration is that Tier-1A was created for Interrogation of detainees assumed to have vital information, about terrorist groups or insurgent. These interrogations relied on a variety of ”soft torture” tactics by civilian interrogators and others. These guards were encouraged to stress and abuse detainees, and were reinforced for breaking them down to prepare them to make confessions.

A few quotes from one of the reports (by General Faye) are pertinent to understanding the systemic influences operating in that prison situation and on those prison guards.

“By not communicating standards, policies, and plans to soldiers, these leaders conveyed a tacit approval of abusive behaviors toward prisoners.

“There is both institutional and personal responsibility at higher levels.”

“For a period of 7 months, Military Intelligence personnel allegedly requested, encouraged, condoned or solicited Military Police personnel [the Army Reserve guards] to abuse detainees, and/ or participated in detainee abuse, and/or violated established interrogation procedures and applicable laws…”

“Abuses would not have occurred had [military] doctrine been followed and mission training conducted.”

“The Environment created at Abu Ghraib contributed to the occurrence of such abuse and the fact that it remained undiscovered by higher authorities for a long period of time.”

For a period of 7 months, “Military Intelligence personnel allegedly requested, encouraged, condoned or solicited Military Police personnel [the Army Reserve guards] to abuse detainees, and/ or participated in detainee abuse, and/or violated established interrogation procedures and applicable laws…”

“Abuses would not have occurred had [military] doctrine been followed and mission training conducted.”

“The Environment created at Abu Ghraib contributed to the occurrence of such abuse and the fact that it remained undiscovered by higher authorities for a long period of time.”

The military judge took none of these arguments into account when he issued his sentence, none mitigated Frederick’s receiving the maximum penalty. The judge asserted that he was personally responsible for the abuses, he should have known better, he knew it was wrong, so, he had free will to do what was right and he did the morally wrong action, as part of a conspiracy with other guards. Thus, he sentenced this “good apple” to be imprisoned for 8 (eight) years, to be dishonorably discharged, to have his rank lowered to private, to deprive him of 22 years of hi army reserve retirement funds, and he was sent to Kuwait in solitary confinement (until he was ready to testify against other guards whose trials would be held in the United States).

For me, this verdict represents the triumph of a mindless dispositional view of such unusual behavior. It totally places full blame on the person, ignores the host of situational variables that contributed to the abusive behavior and absolves the corrupt, irresponsible military and political system that created that situation in its rush to war against terrorism. It takes us back to the Inquisition. The issue was never the guilt of this young soldier since he pleaded guilty as charged. The issue was whether the Court would acknowledge this entire set of circumstances that so obviously transformed some one who entered that situation as a model soldier and a good citizen and soon was transformed into becoming a perpetrator of notorious evil. Our legal system does not have a mechanism for dealing with the challenges posed by psychological analyses of situations and systems (Ross &Shestowsky, 2003).

Promoting Civic Virtue, Moral Engagement and Human Goodness

In this final section we turn to consider briefly the enormous challenge facing the world today to promote civic virtue and resistance against situational temptations to engage in the kinds of evil behavior discussed in this chapter. There is no simple solution; were there one it would have been enacted long before by those far wiser than I. My goal is to outline some speculations about what might be done at individual, situational and systems levels to combat the seductive influences on people to transgress against others and violate fundamental moral principles. My analysis will continue to be from a broad social psychological perspective (but also see the important ideas advanced by Seligman (2002) about the role of positive psychology and Shermer’s views, 2004).

At the individual level, let us first image the reverse of the Milgram experiment in which the objective was to create a setting in which people would comply with ever increasing demands to do good, to gradually behave in more altruistic ways, to slowly but surely move further in agreeing to ever more positive, pro-social actions. Instead of the paradigm arranged to facilitate the slow decent into evil, we substitute a paradigm for the slow ascent into goodness.

As a thought experiment for you, the reader, how could you design a setting where that was possible? As a starter, perhaps imagine for any participant in our escalation into goodness experiment that we arrange a hierarchy of experiences, actions that range from only slightly more positive than he or she is used to doing to ever more extreme “good” actions, all the way up to those that are hardly imaginable as personally possible. It might be a time-based dimension for those who do not do good deeds because they don’t have the time to spare. The first “button” on the Goodness Generator might be to spend 10 minutes writing a thank you note to a friend. Next level might be 20 minutes giving advice to a troubled child, Increasing the pressure in the new paradigm might then be that the participant agrees to give 30 minutes to prepare a meal for a needy person. Then the altruism scales upward to spending an hour doing another good deed, baby sitting for a few hours to allow a single parent a break in routine, working an evening in a “soup kitchen” to help feed those in need of food, giving a day to take a group of orphaned children to the zoo, committing to giving some precious time every week to some other good cause, and so forth. If the gradual escalation principle can work to get good people to do evil deeds as Milgram has shown, can we reverse the process using a similar paradigm to get ordinary people to do increasingly good things? Ideally, our experiment in social goodness would end when the person was doing something that he or she could never have imagined doing previously, acting in such an extremely altruistic, beneficent manner that had been alien to one’s self concept. Obviously, we want to extend this concept to practical ways of subtly leading people down the path of taking ever increasingly good actions that help others and enhance their society. The goodness track could also have to do with contributions to making the environment more sustainable. It might go from minimal conservation activities to ever more substantial ones, giving money, time, and personal involvement in “green” causes. I invite readers to expand on

this notion for a host of domains where society would benefit as more citizens “went all the way.”

If we consider some of the social psychological principles that fostered the evils I have noted earlier, then again as with the Goodness Generator example, let’s use variants of those principles to get people to accentuate the positives and to eliminate the negatives in their lives. Here are my 11-steps toward promoting civic virtue that are in opposition to the 10-steps toward evil that I outlined from extending the Milgram paradigm to our lives.

Encouraging admission of one’s mistakes, accepting errors in judgments, being willing to say that you were wrong. Openly doing so reduces the need to justify the mistakes, to continue the wrong or immoral action. It undercuts the motivation to reduce dissonance by asserting or believing in the public commitment when it was a bad decision.

Encouraging “Mindfulness” (Langer, 1989) in which people are reminded in a variety of ways not to live their lives on automatic pilot, but to take a moment to reflect on the immediate situation, to think before acting, to not go mindlessly into situations where angels and sensible people fear to tread.

Promoting a sense of personal responsibility and accountability for all of one’s actions, making people aware that conditions of diffused responsibility merely disguise their own individual role in the outcomes of their actions.

Discouraging even the smallest transgressions, cheating, gossiping, lying, teasing and bullying. They provide the first steps toward escalating downwards to ever worsening behaviors.

Learning to distinguish between Just Authority, to whom respect and even obedience may be appropriate, and Unjust Authority (as in the Milgram study), to whom disrespect and disobedience are necessary to oppose and change that tyrant.

Supporting critical thinking from the earliest times in a child’s life and maintaining it throughout life. Asking for evidence to support assertions, demanding that ideologies be sufficiently elaborated to separate rhetoric from reality-based conclusions, to independently determine whether specific means ever justify vague and harmful ends.

Rewarding social modeling of moral behavior, elevating for societal recognition those who do the right thing, with rewards for “whistle blowers,” such as the U.S. army reservist, Joe Darby, who exposed the abuses at Abu Ghraib, and those who expose wrong doing in government and corporation, and by the Mafia.

Respecting human diversity, appreciating human variability and the differences among people as a fundamental way to reduce our in-group biases that lead to derogating others, prejudice and the evils of discrimination.

Changing social conditions that make people feel anonymous, instead supporting conditions that make people feel special, so that they have a sense of personal value and self worth.

Becoming aware of when conformity to the group norm is counter-productive and should not be followed, when independence should take precedence and be adopted regardless of social rejection by that group.

Never allowing one’s self to sacrifice personal freedoms for the promise of security, it is always a bad deal because the sacrifices are real and immediate and the security is a distant illusion. This is as true in marital arrangements as it is in being a good citizen in a nation where the leader promises to make everyone safer against a current threat by giving up some of their personal freedoms so that the leader can have more power. That bad bargain usually translates to more power Over Them, as well as over the enemy. It is the first step in creating fascist leaders even in democratic societies, as Erich Fromm (1941) reminded us about Hitler, but is as true today in many nations.

I hope that my future research and social-political actions as a citizen-scientist will be more directed toward understanding how to promote goodness in the world than to demonstrate how easy it is to seduce good people to become perpetrators of evil.


Interview with Carlo Prescott

Prison Consultant and Head of the Adult Parole Board, Stanford Prison Experiment

In a recorded telephone interview conducted in July of 2018, Carlo Prescott was asked about an April 28, 2005, op-ed essay, entitled «The lie of the Stanford Prison Experiment,» that was submitted under his name to The Stanford Daily student newspaper. Recent critics of the study have treated the 2005 op-ed as authentic, even though the essay was written in a voice very different from Mr. Prescott’s, and Mr. Prescott himself has denied having anything to do with it (for further information on the op-ed and the motivation of its author, see Professor Zimbardo’s Response to Recent Criticisms!. To set the record straight, Professor Zimbardo’s personal assistant, Taylor Langely, contacted Mr. Prescott by telephone and, with his permission, recorded a brief statement for public distribution.

Langley: So, critics of the Stanford Prison Experiment are claiming it was all a lie. These critics argue that Carlo Prescott, a paid prison consultant during the study, wrote a Stanford Daily article claiming the experiment was flawed and dishonest.

We are here today on July 11, 2018, talking to Mr. Prescott himself to get the truth. Um, so, I just want to ask you one question, and the first one is, did you write or personally collaborate on the Stanford Daily campus newspaper article?

Prescott: To my knowledge, I’ve never endorsed nor do I intend to ever have involved myself in something adverse, uh, adverse and derogatory to the Stanford Prison Experiment. As to my comments: absurd. About Philip Zimbardo? Totally and absolutely absurd. I don’t know why it isn’t–well, maybe I’ll get a chance to tell you what I think later.

Langley: Yeah, thank you. Um, and the second question is: Are the negative allegations made in your name true?

Prescott: No. Positively, absolutely, without any hint.

Langley: Okay. Um, so, why didn’t you request a retraction from the Stanford Daily newspaper?

Prescott: Basically, because I don’t get a daily delivery. [Laughter]

Langley: [Laughter] Yeah.

Prescott: [Laughter] You sound like somebody in Washington calling about my opinion about this Mexico thing. Ah, because I don’t get the Stanford paper, and because it never come to my attention. For some reason the enmity that was shown towards the experiment and towards in particular my friend and brother, Philip Zimbardo, are absurd. But understand clearly, the answer is emphatically, «No.»

July 11, 2018

Note: To hear or download an mp3 audio recording of this interview, please visit: http://www.prisonexp.org/links/#responses

To read a signed statement in which Carlo Prescott confirms that he did not write the Stanford Daily op-ed, and requests that it be retracted, please visit:



Statement by Gordon H. Bower, Ph.D.

Professor Emeritus, Stanford University Psychology Department Recipient, National Medal of Science

I write in support of my professional colleague, Dr. Phillip Zimbardo, one of whose controversial experiments (the Prison Experiment) has recently been questioned by critics. Many of these criticisms are not new, and Dr. Zimbardo has published rebuttals of them as they have appeared over the years. While I cannot vouch for every statement that Dr. Zimbardo has written or said about his experiment (see his rebuttals), I can vouch for the scientific honesty, soundness, and integrity of the man himself. I have known Zimbardo for 60 years and have found him to be a straightforward, honest thinker. He has a flair for communicating in a dramatic fashion; that is part of his attraction for listeners and readers and partly why he is a marvelous teacher. But to my knowledge, he does not knowingly distort the truth of the basic message he is communicating.

The basic message of the Prison Experiment was that people’s behavior is profoundly influenced by the immediate social environment suggesting how they should enact the roles they have adopted, and that there is a very fine line separating behavior of a «pretender» to a role versus a committed «enactor» of the role. I believe that basic fact–the «relative power of the social situation»–has been replicated many times in different experiments. Some of the current critiques also remind me of experiments indicating that people’s recollection of their past actions and beliefs can be influenced by their desire to make them unwittingly consistent with their current evaluation of those past events. In any event, I remain committed to my belief in the scientific integrity and soundness of Dr. Zimbardo’s career in social psychology, and I feel that these attacks upon him and his science are unwarranted.

July 4, 2018


Statement by Larry James, Ph.D., ABPP

Colonel (Retired), U.S. Army Professor and Director, Continuing Education Program School of Professional Psychology, Wright State University

In April of 2004, the United States learned of the abuses at Abu Ghraib Prison in Iraq. I was selected by the U.S. Army to be a part of the team dispatched to this awful place to appropriately train the staff to interview rather than harshly interrogate prisoners, and to develop a policy that would end the abuses.

As with most of us who have taken an introductory psychology course, I had read about the Stanford Prison Experiment, but I had never met Dr. Zimbardo or had the opportunity to discuss with him exactly what happened, what went wrong in his famous study, and how the lessons learned could help our nation end prison abuses at Abu Ghraib.

In early May of 2004, Dr. Zimbardo was scheduled to give a lecture at the University of Hawaii in Honolulu, and I happened to be stationed in Hawaii as an Army Psychologist. Out of the blue, I received a telephone call from Dr. Zimbardo because he had learned that I would soon be visiting Abu Ghraib to help end the abuses. He then offered to meet and make suggestions on what to do and not do.

Our meeting in an Italian coffee shop ended up lasting two hours. Dr. Zimbardo was incredibly generous with his time and expertise. Truly, our nation owes him a great debt of gratitude because he helped me craft an 11-step plan at that meeting to end the abuses at Abu Ghraib, and indeed, to shape policies outlawing abusive practices at military detention facilities around the world.

In my work with Dr. Zimbardo, his character and integrity have been without question. Most of all, I will always think fondly of his warmth, generosity, research expertise, and willingness to help me and others during a very dark time in our nation’s history.

July 17, 2018


Statement by Ken Musen

Writer, Producer and Director of Quiet Rage: The Stanford Prison Study

I first met Doug Korpi [former Prisoner 8612 in the Stanford Prison Experiment] in 1988.1 had invited him to lunch in San Francisco as part of my research for a documentary project on Professor Philip Zimbardo’s renowned 1971 psychology experiment. I discussed with Doug his potential involvement in the documentary as both a past participant in the Stanford Prison Study and as a current forensic psychologist in the San Francisco prison system. At that lunch, we discussed in great detail his perspective of the guard/inmate relationship that he has observed as a prison psychologist. Ultimately, the title for the documentary,

«Quiet Rage,» was born from this discussion–a term he coined describing the bent-up aggression inherent in the guard/prisoner relationship. At that meeting, Dr. Korpi became both a collaborator and a willing interview subject for my documentary project.

Doug suggested that we film in his prison in the Bay Area where he worked. As a filmmaker, I could not have hoped for a better setting than a real prison cell in a working prison. My interview questions were honest, sincere and open-ended. We started with a chronology of the events that led up to his decision to participate in the study. He recounted his entire involvement starting with his «arrest.» I remember he was very impressed with the level of realism Zimbardo created including having his fingerprints stamped and mug shot taken at the Palo Alto Police Station.

The interview became increasingly more intense as Doug recounted his 36-hour involvement and his eventual mental collapse. He remembered «never feeling that out of control» before in his life. He remarked that his decision to choose forensic psychology as a career was somewhat determined by his experiences in the study. «I wanted to better understand… how I could be so out of control.» None of his remarks were coerced or manipulated.

In my preparation for his interview, I had listened to all the recordings Dr. Zimbardo and his graduate students recorded during the six days the study ran. The audio tapes fully document Korpi’s collapse. It was genuine. Ethical or not, 21 year-old Doug Korpi was told by the prison staff that he could not be released from Zimbardo’s jail. At that moment, Zimbardo’s prison became «a real prison.» This fact threw him into a frenzy that lasted over 3 minutes and ended in an expletive-laden breakdown. As a student of psychology, filmmaking and acting, I can without hesitation state that Korpi was not merely acting but reacting to his situation as he grew to understand it.

My relationship with Dr. Korpi continued throughout the production and postproduction of the documentary. Doug wanted to see the rough cut of the project, so we sent him the cut. I believe that he was embarrassed by the section in the video that recounted his breakdown where a portion of his screaming is played over an image of his face. It was one of the most disturbing sections in the documentary. He asked if I would lessen the amount of time spent on that painful moment, and based on our relationship, I agreed, cutting the section in half. Later, when the documentary was completed, Doug showed other forensic psychologists and colleagues the program believing he had participated in a worthwhile documentary that explores important psychology themes, including how the power of the situation can affect human behavior and choices.

Based on my experiences working with Dr. Korpi on “Quiet Rage: The Stanford Prison Study,” it is simply not true that he was unaffected by his time spent in the Stanford Prison or that he was only play-acting to get an early release to study for a test. From my time researching the subject matter and hours spent interviewing Doug Korpi for the project, I believe his emotional breakdown was a genuine reaction to his perceived environment.

June 25, 2018

Note: To see a brief excerpt from Quiet Rage in which Dr. Korpi discusses the breakdown, please visit:


[1] Correspondence should be addressed to Philiip Zimbardo, Stanford University, Stanford, CA 94305, USA (e-mail: zim@stanford.edu).

[2] Wells, M. (Jan. 24, 2002). BBC halts ‘prison experiment’. The Guardian Unlimited, on-line.

[3] Koppel, G., & Mirsky, N. (Series producer and Executive producer, respectively.). (2002 May, 14,15, 20, 21). The Experiment. London: BBC.

[4] Derbyshire, D. (May 3, 2002). When they played guards and prisoners in the US, it got nasty. In Britain, they became friends. The Daily Telegraph, p. 3.

[5] Alleged Random Assignment. The prisoners’ backgrounds were very different from those of the guards: ex-crack addict, martial arts expert, security expert, former army officer and the only black participant. They had more ‘street smarts’ to begin, and used assertive Machiavellian tactics to counter the guards’ authority. They seemed chosen from the hundreds of applicants by BBC ‘central casting’ to be more like what the public imagines convicts look like. One of the most distinguishing features of the prisoner-guard difference is the flamboyant tattoos prominently displayed on some prisoners’ arms — that not coincidentally were the target of many BBC camera close- ups. A central guard figure was a millionaire hi-tech executive who was always aware of how his behaviour would be viewed in the future when the study was shown publicly on the BBC. Accordingly, he enacted the ‘good guard’ role, never being dominant or abusive, even being conciliatory and open to negotiating any and all prisoner demands. Alone with fellow guards, he reminds them of what happened in Nazi Germany when people got too much into their roles. Later on, while trying to defuse tensions, he says to all the participants, ‘There should be two winners, both groups of guards and prisoners, and at the end (of the experiment) we all go to the pub and have a drink together’. He does not want to be guard, proposes a revolution against the experimenters by establishing equality between the two groups. His future orientation mediates against ever getting enmeshed in the immediacy of the present moment. In sum, the alleged random assignment is a very strange case of random

Prisoner 922, Ian Burnett quoted in Nikki Murfitt, Jailed! Evening Standard, (2 May 2002), p. 30.

Matt Wells (10 April, 2002). BBC2 delays ‘unfair’ prison experiment. The Guardian, National News, p. 8.

[8] Correspondence should be addressed to Stephen Reicher, School of Psychology, University of St. Andrews, St. Andrews, Fife, KYI6 9JU, UK (e-mail: sdr@st-andrews.ac.uk).

[9]Correspondence should be addressed to Alex Haslam, School of Psychology, University of Exeter, Exeter, Devon, EX4 4QG, UK (e-mail: A.Haslam@exeter.ac.uk).

[10] This procedure — which ensures equivalence on theoretically relevant variables — was considered superior to ‘pure’ random assignment (i.e. one which does not take account of any individual differences, as in the SPE), as the study’s relatively small sample size means that random assignment has the potential to produce large differences between groups (due to the law of large numbers; e.g. Haslam & McGarty, 2003, pp. 180—183).

[11] In this instance (and with depression data below), analysis of polynomial contrasts was based on a random subset of five phases because the number of levels of the within-subjects variable (study phase) exceeded the number of subjects.


[13] Some commentators have argued that this analysis is inconsistent with evidence arising from the torture of inmates by US troops at Abu Ghraib prison in 2004. However, as we have argued elsewhere (Reicher & Haslam, 2004), it appears that these acts were performed for the benefit of an in-group audience that (like the experimenters in the SPE) was assumed by the perpetrators to approve of the action.