• Share:

In his new book, Incarceration Games: A History of Role-Play in Psychology, Prisons, and Performance, Stephen Scott-Bottoms offers a fresh perspective on some of America’s most notorious social psychology experiments. Incarceration Games is now available in hardcover, paperback, and ebook. There will be a virtual book launch on Thursday, May 23rd. Register at Eventbrite. 

“People are fascinated with the Stanford prison study,” Philip Zimbardo tells me reflectively, “but in reality, in the real world, almost no-one gives a shit about prisons.” He is standing in the doorway of his San Francisco home, bidding me farewell, and this thought has just hit him: “Prisons in and of themselves are not a topic that anybody is curious about, but they are interested in the drama of the prison study. For me, it’s a curious dichotomy.”

It’s February 2020, just before Covid imprisoned us all, and I’ve just been interviewing this famous psychologist for a book I’m researching. It has become apparent, during this conversation, that Zimbardo is still hurt and upset by some very personal accusations made against him in the previous couple of years. The writers Ben Blum and Thibault LeTexier have both claimed publicly that Zimbardo and his co-researchers told flat-out lies about the outcomes of their notorious, 1971 prison experiment at Stanford University. These revelations, which have prompted a predictable Twitter-storm of righteous indignation, are dependent on LeTexier’s rather cursory reading of archival materials relating to the experiment. Yet as Zimbardo quite reasonably points out, he had made those very materials available for public consultation in the interests of transparency. For him, at least, there were no lies to conceal. Moreover, he notes, Blum and LeTexier had cried “gotcha” without making any attempt to relate their critiques back to the subject of prisons. The veracity or otherwise of the prison experiment itself had apparently become more newsworthy than the bleak realities it had set out to dramatize. Hence Zimbardo’s parting observation to me.

His reference to “the drama of the prison study” was also, no doubt, informed by his awareness that I am a drama professor—not a psychologist. My own interest in the Stanford Prison Experiment arises from the fact that this was essentially an exercise in improvised role-play. A volunteer cast of student-age young men were arbitrarily allocated to roles as prisoners and guards in a mocked-up prison setting, and asked to play those parts for a watchful audience of psychologists. What happened next has become the material of popular myth, as the guards allegedly became more “brutal” toward the prisoners than anyone had anticipated. There were, however, no independent observers present to witness these events. The only accounts we have of this intimate drama have been provided by the psychologists who staged it. From the perspective of a theatre historian, there are obvious problems with this—whether or not anybody lied. It’s like having a playwright’s description of their own play, but no access to the play itself. So I’ve been scouring the archives, and interviewing surviving participants, in the hope of reconstructing these events from a more objective point of view. Somewhat remarkably, nobody has ever tried to do this before. 

My hope, in conducting this re-examination, is that a fresh account might help to resolve the “curious dichotomy” that Zimbardo points out. For if, as he suggests, people remain fascinated by the “drama” of the Stanford Prison Experiment, that is because he has always told the story of this study in bold, primary colours—using pleasingly direct language. Take Zimbardo’s book-length account, The Lucifer Effect (2007), with its knowingly populist subtitle: How Good People Turn Evil. This is not exactly scientific language, but it neatly captures his core theoretical claim about the study: under certain situational pressures, he argues, even the best of us may end up acting abominably. And maybe that’s true. Yet Zimbardo’s evidence for this claim rests on a specific experimental situation—an attempt to simulate the power dynamics of a prison. This is the bit that tends to get fudged or ignored in pop-culture retellings of the story, because prisons—as Zimbardo suggests—are not something that most of us want to think about for long. Real prisons are more mundane, more complicated, and more oppressive than any psychology experiment could hope to approximate. So instead, oftentimes, the Stanford study is presented as proving some generalized point about “human nature” (whatever that is).

This is, however, to miss the point. The allegedly abusive behavior of the Stanford guards has always been presented by the researchers as a consequence of a specific, situational phenomenon—namely, the structural power imbalance between guards and prisoners that exists in any prison setting. Zimbardo’s study has often resonated with prison professionals for precisely this reason—because they recognize the problem identified. Institutional conditions can, indeed, result in staff adopting a casually dehumanizing attitude toward the inmates in their charge—unless care is taken to ensure the contrary. Crucially, though, such care is often taken. Not all prisons are the same.

As criminologist Alison Liebling once put it to me, every prison is its own social experiment. The specifics of a building’s size and architecture, of its population’s daily routines, of its governor’s disciplinary philosophy—all these variables and more will impact on the conduct and wellbeing of inmates and staff. Liebling’s own work, as Director of the Prisons Research Centre at Cambridge University, has demonstrated this point convincingly. Her MQPL questionnaire (Measuring the Quality of Prison Life) has been used extensively, in a wide range of institutions, to gather comparative data on the day-to-day experiences of those who live and work in them. This has enabled Liebling and her colleagues to establish clear correlations between—for example—the perceived harshness of prison regimes and the rate of suicide among prisoners.  In crude terms, you are demonstrably more likely to kill yourself if you are held in an institution that exhibits poor moral performance. So Zimbardo was onto something when he highlighted situational conditions over individual disposition. Rather than simply containing “bad apples”, some prisons are indeed “bad barrels”.

But if this is the case, an obvious question arises. What were the specific situational conditions that resulted in such negative outcomes for the Stanford Prison Experiment? What exactly made this simulation such a bad barrel for those subjected to it? Zimbardo and his colleagues have never been very clear on the details here—apparently because their orientation as psychologists was towards extrapolating generalizing claims about psychological phenomena, rather than honing in on specifics. So perhaps this is where a theatre historian like myself might have a role to play. In theatre studies, the critic’s orientation is usually toward the specific. Why, we might ask, was this production of Hamlet received so differently by audiences than that one? How exactly did actors, scripts, sets, costumes, etc., function together to create meaning?  Or, more pertinently here, why is it that one iteration of an improvised role-playing game might result in very different outcomes from another? As any drama teacher knows, improv scenarios can spin off in all sorts of directions, as the various personalities in the room react spontaneously to the developing situation—and to each other. The basic rules of the game may stay the same, but the players’ game tactics will always be different—which is, of course, why good games are worth playing repeatedly. As for bad games, which turn out to be ethically flawed, we might just need a historian to piece together the details of what went wrong.

It’s April 2024. Covid is an uncomfortable memory. Sitting in front of me, on my desk, is an advance copy of my new book, Incarceration Games: A History of Role-Play in Psychology, Prisons and Performance. As the subtitle suggests, I’ve worked hard to resolve Zimbardo’s curious dichotomy. The Stanford experiment sits at the heart of the book, but my story also tracks backwards as far as the 1930s, and forwards as far as the 2000s, to examine other experiments—and other role-playing games—in which the coercion, manipulation, or even rehabilitation of prisoners was an overt or covert concern. The jacket blurbs tell me that “Incarceration Games moves well beyond ‘gotcha’ sensibility … providing a refreshing, novel perspective” (Michael Pettit); and that “by bringing to bear methods familiar to theatre historians,” I have written “a fascinating, readable, rewarding book” (Mike Sell). I will certainly take the compliments. It remains to be seen, though, whether these tales of situational experimentation will inspire readers to give a shit about prisons.