The record of the Stanford prison experiment is a collection called the Philip G. Zimbardo papers, call number SC0750, in the Department of Special Collections and University Archives of the Stanford University Libraries. The finding aid gives its dates as 1953 to 2017 and its extent as 256 linear feet in 182 boxes, “donated by Philip G. Zimbardo to Stanford University, Special Collections in multiple accessions from 2011-2017”. The materials, it says, “consist of Dr. Zimbardo’s research and teaching files, professional files and correspondence, audiovisual materials used in the classroom, professional papers and articles, and materials documenting the Stanford Prison Experiment, for which he is perhaps best known.” They are arranged in nine series: audiovisual materials; born-digital materials; professional files; publications and writing; research files; the Stanford Prison Experiment, which is Series 6; teaching files; oversized materials; and restricted materials. The boxes are stored off site and must be asked for forty-eight hours ahead. “Boxes 9 and 9A in Series 6 (Stanford Prison Experiment) are restricted to protect participant privacy.” Files in Series 9 are closed for seventy-five years from their creation. The copyright is Zimbardo’s; permission to examine the papers “is not an authorization to publish.”

A second, smaller deposit is at the Archives of the History of American Psychology at the University of Akron in Ohio, a single box, which holds among other things a guard’s letter to Zimbardo of November 18, 1971, and a prisoner’s retrospective diary of January 1972.

What is online

Stanford digitised the experiment’s core and put it in its Digital Repository as one object, Stanford Prison Experiment, at purl.stanford.edu/vx097ry2810, with Zimbardo, Haney, Banks and Jaffe named as authors and the note that it “was carried out August 15-21, 1971, in the basement of Jordan Hall”. Its use statement asks only that the content “will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals”. Its inventory runs to about 140 files, each a PDF, a scan or a transcript, and the names are the best map of the week that exists.

There are the papers of the design: “Prison Life Study: General Information”, the sheet given to applicants; the consent form; “Application for Institutional Approval of Research Involving Human Subjects”, dated July 31, 1971; “Prisoner Rules Prepared by the Warden and Guards”; “Rules for Prisoners”; “Suggested Daily Schedule”; “Count Tally Sheet”; “Critical Incident Report Sheet”; “Guard Shifts and Guard Shift Reports”; “Stanford Prison Letterhead”; a food service menu; “Media Release Forms”; “Arrest cards”; “Volunteer background information forms”; “Attitude Questionnaire” and “Opinion Survey II” and “Mood Adjective Checklist”; “Final Prison Study Evaluation”; “Final Release Statements”. There are the logs: “Day One notes”, a log for each of days two, three and four, and for days five and six “NBC style logs”, three-column tables of events, prisoners’ reactions and guards’ reactions compiled from the files. There is “Dave Jaffe, Log and Self-Perception”; “David Jaffee, ‘A Simulated Prison’”, the Toyon Hall report; “Jaffe’s speech”; “Parole Board Hearing”; “Zimbardo’s comments and conclusions”; “Overview of method & data results” of August 1971; “Results” of July 31, 1972; “Video Analysis (Haney)” and “Audio Analysis (Banks and Rosenfeld)” of December 23, 1971; “Video notes” for days one to three and tapes four to six; and “Audio tape log”.

And there are the tapes, as transcripts and, for some, as digitised sound. “Audio transcript—tape 2, Orientation of Guards” and “Audio transcript—orientation of guards, 1971”, the Saturday; “tape 3, Guard Final Evaluations”; “tape 6, 416”; “tape 7, Prisoner 8612”; “tape 8”; “tape 10”; “tape 11, 3401”; “tape 12, NBC Interviews”; “tape 1A, 5486”; “tape E”; “tapes 4 and 6”; “Tape A, Interview with 8612”; “Audio transcript—selections”. The catalogue entries for the cassettes themselves give their source identifiers, SC0750_s5_b2_02 for the orientation, SC0750_s5_b2_07 for Korpi’s final reactions, SC0750_s5_b2_21 for the tape with the warden’s talk to a guard, and SC0750_s5_b1_10 for the interviews of September 16, 1972; the film reel of the orientation and arrests is “Prison #20”, SC0750_s5_b6_01, and there are U-matic and one-inch videotapes numbered 1 to 12, including “#12”, the debriefing of August 29, 1971. Then the afterlife: the slide show and its scripts, long and short; “Quiet Rage” in three parts with its transcript; “NBC Chronolog transcript—guard and prisoner”; the International Journal of Criminology and Penology article; “Statement to Judiciary Committee” of October 25, 1971, and the letter from Emanuel Celler of October 18; “Prison Testimony” of September 11, 1973; “Prison study—Dimsdale critique and response” of August and September 1971; the Savin critique of 1973; the Lovibond, Mithiran and Adams paper of 1979; news clippings from 1971 to 1973; two film scripts, one from 2006; and “Letter to editor, Stanford Magazine, re: 40th anniversary article errors” of August 2011.

What was never recorded

The inventory is also a list of what is not there. Thibault Le Texier, who worked through the sixteen boxes at Stanford and the one at Akron in 2017, counted six hours of film and forty-eight hours of audio in all, of which, he wrote, about fifteen hours of audio and the six of film were made during the experiment itself. The experiment, with its orientation day, ran about 150 hours. “Of the 150 hr of the experiment (including the orientation day), less than 15% have been recorded.” Zimbardo has said there were twelve hours of video; Le Texier found six. Nothing was recorded on the third day, when the staff were occupied with the rumour of a break-in and moved the prisoners upstairs. The guard orientation ran, by the warden’s log, from eleven until half past three, and only part of it is on tape. Of the meeting between the warden and the reluctant guard, by contrast, Haslam, Reicher and Van Bavel noted that “a recording of the entire interchange is available (whereas this is true of less than 10% of the entire study)”.

What was recorded was chosen. Craig Haney’s “Video Tape Analysis”, written soon after the experiment, is in Box 10, Folder 8: “as a result of a number of serious distortions in the data, most of which derive from the vicissitudes of initial recording, our statements are rendered much more equivocal. Even if our methods of data reduction, the translation of behavior into numbers, could have maintained perfect fidelity to the original, our video sample was seriously biased at the outset. For the most part, our concerns in filming were cinematic: we recorded primarily the dramatic (e.g., counts) or the unusual (e.g., the priests’ visit). In a sense, these events are unrepresentative by definition. It is the commonplace, the regular, the mundane occurrences which best portray the reality of a total institution and these are the very things we do not have (or have sparsely) recorded.” The “Overview of Method” Zimbardo wrote in August 1971 reports from the video that 88 per cent of “aggressive acts” were guards’ and 91 per cent of “defensive resistance” prisoners’, and that guards became more aggressive over time at a probability of less than one in ten. In his 1973 paper and later books the table of interactions is given without the note that it rests on a few hours of the dramatic.

There is a second absence. The archive holds, in Le Texier’s description, very little about the men themselves: the background forms asked about family, health and criminal history, and the experimenters, he found, “barely questioned them about their religious or political beliefs and activities, their hobbies and sport activities, their cultural background or about their social status.” The participants are coded in his paper, as the ethics of the discipline now require, and the two boxes of their individual files are closed. What can be read is what they wrote for the experimenters: shift reports, questionnaires, diaries kept afterwards at Zimbardo’s request, letters, and the transcripts of what they said into the microphones in the corridor, in the interview room, and a year later.

How to read it

The order of the archive is the order of the experiment’s own filing, and that is the point of it. The orientation transcript sits beside the paper that says the guards received no specific training. The warden’s log, with “Schedule followed pretty carefully” against three days, sits beside the account that says the guards made up the schedule. The press release of August 16, typed before the second night, sits beside the statement that what happened surprised everyone. Korpi’s final reactions of August 19 sit beside the film narration of 1992. None of these documents was hidden; several were on Zimbardo’s own website for years, and the collection has been open to anyone who asked since it was catalogued. Le Texier’s observation in 2019 was that nobody asked: “no psychologist seems to have wanted to know what exactly the archives contained.” His four explanations were a lack of curiosity, respect for a tenured professor at a prestigious university, “possible access restrictions imposed by Zimbardo”, and the labour of archival work, and he said he did not know which was right. Zimbardo’s own view, in 2018, was that “the SPE has been a model of open science long before practices such as public archiving and data sharing were common.” Both statements are true of the same boxes. What the entry takes from them is the one thing nobody disputes: that for forty-seven years the record and the report were in different places, and the report was the one that was read.

Sources

• Philip G. Zimbardo papers (SC0750), Stanford University Libraries, Department of Special Collections and University Archives, finding aid on the Online Archive of California: dates, extent, abstract, arrangement, acquisition and restrictions, quoted above. oac.cdlib.org
• Stanford Prison Experiment, Stanford Digital Repository, druid vx097ry2810: the abstract, the use statement, and the inventory of files read from the object’s public metadata; the catalogue records of the cassettes and tapes with their source identifiers. purl.stanford.edu/vx097ry2810, purl.stanford.edu/dx583pt8943, purl.stanford.edu/fb554xz1315, purl.stanford.edu/wn708sg0050, purl.stanford.edu/rw223jh0253, purl.stanford.edu/pk326gk3338, purl.stanford.edu/dw596kh0653
• Craig Haney, “Video Tape Analysis,” undated, SC0750, Box 10, Folder 8, digitised as SC0750_s5_b10_f08.pdf; Philip G. Zimbardo, “Prison Study August 1971: Overview of Method” and “Overview of Data Results,” digitised as SC0750_s5_b11_f03.pdf. purl.stanford.edu/gc938ns7497
• Thibault Le Texier, “Debunking the Stanford Prison Experiment,” American Psychologist 74, no. 7 (2019), 823–839: the boxes and hours, the 15 per cent, the third day, the twelve hours against six, the personal information, and the four explanations. doi:10.1037/amp0000401
• S. Alexander Haslam, Stephen D. Reicher and Jay J. Van Bavel, “Rethinking the Nature of Cruelty,” American Psychologist 74, no. 7 (2019), 809–822, read in the preprint: “less than 10% of the entire study.” osf.io/b7crx
• Philip Zimbardo, “Philip Zimbardo’s Response to Recent Criticisms of the Stanford Prison Experiment,” June 23, 2018: “a model of open science.” prisonexp.org/response