Book Search:  


Google full text of our books:


Games Prisoners Play:
The Tragicomic Worlds of Polish Prison
Marek M. Kaminski

Book Description | Reviews | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2004, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format


Prison socializes an inmate to behave hyperrationally. It teaches him patience in planning and pursuing his goals, punishes him severely for his mistakes, and rewards him generously for smart action. No wonder that inmates are such ardent optimizers. A clever move can shorten one's sentence, save one from rape or a beating, keep one's spirit high, or increase one's access to resources. There is little space for innocent and spontaneous expressions of emotion when they collide with fundamental interests. Brutal fights, self-injury, and rapes can all be explained as outcomes of carefully calculated actions. Paradoxically, much of the confusion in interpreting prison behavior arises from both a failure to understand the motives of inmates and an unwillingness to admit that outcomes judged as inhuman or bizarre may be consequences of individually rational action.

The main message of the book is that prisoners optimize under the constraints of their harsh life conditions and the local subculture. Their behavior reflects their attempts at optimization. Such a "rational choice" approach helps us to better understand prison behavior.


I beg the reader's forgiveness for a brief personal narrative that explains how I learned this lesson myself, and how I collected the data that support it. This is not an autobiography, but I would not be writing this book had I not experienced the life of a prisoner firsthand.

In 1985 I was a twenty-two-year-old sophomore student of sociology who had switched disciplines, disappointed with abstract concepts after three diligent years of studying math. Poland had just witnessed the glorious rise of the Solidarity movement in 1980 followed by the introduction of martial law under General Jaruzelski in 1981 with the rationale that can be summarized as "I kicked your ass, but the Soviets would shoot it." Dissatisfied with the moral and esthetic poverty of communist way of life, I joined the underground Solidarity resistance network. In 1985, I was running an underground publishing house, STOP that employed about twenty full-time workers and up to 100 moonlighters. Between 1982 and 1989, we published about thirty-five titles of more than 100,000 books combined. We were a part of a decentralized network that included about 100 underground publishing houses, hundreds of periodicals, thousands of trade union organizations with a hierarchically organized leadership structure, a few Nobel prize winners, and even underground theaters, galleries, and video rentals. We called it an "independent society."

Half-revolutionist, half-scholar in the making, I was also looking for a topic for my Masters thesis in social anthropology. With hesitation, I started collecting data on the inner workings of the resistance network. My dilemma was figuring out how to balance facts with fiction. If too accurate, my thesis could easily become a handbook for the communist secret police. After my thesis defense I could also fall under permanent surveillance, effectively preventing me from running my organization. At the very worst, the communist court could use my thesis as evidence and throw me in prison.

On March 12, 1985 my thesis dilemma was solved. During a random stop at a police checkpoint, "Dragon," the driver of our van, was so nervous that the policeman became suspicious. He disregarded Dragon's fake documents and implemented a thorough search of the van, which was filled with illegal Solidarity books. Dragon decided to talk. Within hours, five secret police agents had escorted me to a police station, joking that "you will have to swim, Mr. Marek." In fact, I was "swimming"--police jargon for jail sentence--for five months in the Bialoleka and Rakowiecka jails. On my second day in a police station cell, after overcoming my initial shock and disbelief, I decided that my thesis would be on the subculture of Polish prison.

After just several hours I knew that I was entering a bizarre, terrifying, and incredibly interesting environment. Rapes, knife fights, suicides, brutal sex, blunt talk, and self-injuries appeared to be its chief attributes. Ordinary life was reduced to eating and defecating. It seemed as if Pandora had freed all the imaginable violent human emotions from her box there and let them play without the usual societal constraints.

I decided to make the best of my personal misfortune and use it as a unique opportunity to study this fascinating society-within-society. My goals were clear: I did not want to write nostalgic memoirs or point an accusing finger at the regime that had jailed me. I wanted to conduct an extensive and uncompromising research project, using all of my methodological skills. I expected that this would require developing new research techniques or modifying old ones. I was ready to face the necessary risks. It was up to me whether I mobilized my academic spirit--or gave up and slipped into the monotony of day-to-day prison life. I estimated that I would be in prison for up to three years, enough time for a comprehensive field study. Surprisingly, "researching prison" turned out to be an excellent survival strategy. Mentally, it kept me in good shape in the face of adversity--since adversity facilitated fast learning. My research spared me from the helpless repetitions of the "What-am-I-doing-here?" question that introspective characters like to invoke on life's meanders. It helped me to socialize into my new role as an inmate and, at the same time, maintain a healthy distance from it. If you, my reader, are ever unfortunate enough to be jailed, I highly recommend the strategy of "researching prison."

Following my release, I wrote a couple of term papers, some drafts of which had earlier been smuggled out of jail, my thesis, and a few research articles. However, during all that time I suffered from intellectual discomfort and felt that my grasp of prison life remained inadequate. The available prison literature and inmate memoirs offered fascinating details and stories but were of little help in understanding the general mechanisms. It took me nearly three years to find what I thought was a promising methodological approach--game theory. Trying to model prison interactions as games, I completed my formerly abandoned mathematical studies with a specialization in game theory. I became a game theorist and, from time to time, tried to construct prison games.

I wrote this book to correct what I perceived to be the failure of my earlier research. In effect, this book summarizes my recurring attempts to interpret and understand my prison experience over seventeen years. I believe that the galaxy of random anecdotes that I have collected can be condensed into a coherent system. Game theory seems to be well-suited for capturing the spirit of inmate interactions. Games, decision problems, or just informal descriptions of strategic interactions convey the message that I was struggling to formulate at the time of my thesis. With all its weirdness and inhuman appearance, prison behavior is the product of rational persons who calculate the consequences of their actions and try to maximize their payoffs subject to environmental constraints. The goal of my book is to enhance this message, in addition to providing an ethnographic description of prison codes, argot, and customs.1


The book reconstructs various components of the subculture of grypsmen, the highest inmate caste in Polish prisons, and provides a set of formal and informal models representing strategic interactions that arise in the presence of subcultural and other constraints. The main components of subculture include initiation rituals, various explicitly formulated norms regulating the behavior of grypsmen, secret argot vocabulary and grammar, techniques of exchanging information and goods, prison art and entertainment, and techniques of faking and self-injuring. Within the strategic environment defined by prison constraints and subculture, I focus on specific games, decision situations, and tests that characterize prison life.

Formal models help us to convert the enormous complexity of social interactions into more manageable forms. While the price for modeling is always paid in simplification, a good model may offer surprising benefits. For instance, many cases of prison rape follow an inmate's failure to pass a tricky initiation test that is routinely applied to some rookies. There are immediate policy consequences of such a proposition. Revealing the existence of such a test to new inmates could automatically reduce the number of rapes. As the reader will later learn, an informed inmate passes virtually all tests effortlessly.

Although the number of strategic situations described here is large, I develop only a handful of formal models. There is a good reason for doing this: simplicity. Quite often I abandoned formalism entirely when its use might obscure my main point. A model is useful when it clarifies the structure of interactions and when it has sufficiently wide applicability. In cases of initiation tests, relatively simple models satisfy both criteria. In cases such as the "dirty physiology norms" separating eating from defecating or farting, formalization would rarely enhance the reader's insight. Moreover, it would make the description dramatically boring. An informally stated argument: "Scarcity of space, slow airflow, low food quality introduce strong incentives for coordination on defecation and farting norms" seems to capture the point sufficiently well. Some of the stories that I tell have so many idiosyncratic features that a relevant formal model would represent this particular story only. Building a model to explain such stories would be like hunting for a fly with a revolver. Nevertheless, the story may be interesting and important enough to tell it without formalization. Again, my main goal is not to overformalize what can be said simply.


The "closed" prison environment makes collecting quantitative data extremely difficult, if at all possible. Inmates implement so many methods of lying and misrepresentation that surveys or other techniques are often rendered almost useless. Without hard data, rigorous testing of any empirical hypotheses with appropriate statistical tools is not possible. While this book offers various empirical hypotheses, in addition to case studies and models of interactions, it does not attempt to rigorously test any such hypotheses.

Some of the prison constraints are defined by the physical conditions of the prison environment and the penal system. Others are subcultural and their evolution into the final form is particularly hard to explain. I do not aim to explain the changes of systems of norms and rules and their relation with prison constraints. The rational choice approach does not work very well when dealing with the evolution of complex norms, with multiple iterations, incomplete information, or inadequate beliefs. It works best when there are simple constraints, repeatable and standardized interactions, and full--at least on one side--information.

The reader may also be disappointed in the scarcity of comparison with other prison systems. I included a small number of comparative references but decided not to develop systematic comparisons, as this is the subject for a different work.


The prison subculture is immensely difficult to penetrate. Inmates carefully protect information, because they know that the frivolous disclosure of a secret may prolong an inmate's prison sentence, jeopardize his parole, lower his status among comrades, cut off access to resources, or reveal that sickness is simulated. Inmates develop ingenious methods of cheating on one another, on guards, physicians, or psychologists. Techniques of deciphering other inmates are applied in order to identify squealers. A sociologist using a questionnaire in Polish prisons is usually confused with a prison psychologist. Survey answers commonly reflect an inmate's perception of his own self-interest against a person who is perceived as a part of the prison administration. A typical inmate spends a lot of time with his cellmates working out answers to anticipated psychologists' questions that would work best for his case. Such an environment "defends itself against research."2

The main broadly defined source of data was, naturally, my own experience as an observing participant (OP). I define this particular research role, in contrast to participant observation, with two conditions: (a) OP enters a community through a similar social process as its other members and is subject to similar rules; (b) OP undertakes field research as if he or she was a researcher. An ideal OP lives through his/her social role, impassively registers randomly generated personal experience, and applies available data gathering techniques.3

Epistemology of Participation versus Observation

A participant or a participating observer may gather useful data when more formalized methods of data collection are not available or provide unreliable output. A participant perceives his world differently than a participating observer perceives the domain of his study. Differences in beliefs, access to information, and attitudes of these two related roles lead to role-specific epistemological deformations. Such typical deformations are briefly characterized below.

A participant is personally interested in his story. He avoids topics that are inconvenient for him and "forgets" embarrassing facts. A political prisoner emphasizes his own heroism against an unjust regime. A criminal prisoner claims innocence against an unjust court. Both of them believe, after Solzhenitsyn, Bukovsky, and others, that "only a prisoner will understand another prisoner."4 In other words, a typical inmate hardly considers his prison experience to be intersubjectively communicable. He rarely applies any standardized techniques of data gathering. Instead, he focuses on anecdotes and interprets events through his own experience.

A participant observer lacks the sense of real-life pressure participants experience. He is not as affected emotionally by the events as a participant. He lacks experiences that can stimulate one's understanding of insiders' problems. In prison, such experience includes the stress of being arrested, interrogated, or transferred to another prison. He may be unaware that inmates use incredibly ingenious techniques to decipher squealers and that such techniques are applied routinely to newbies. Inmates may check his background, his papers and timing of various events, his contacts in his previous prisons and in the "freedom world," and where he lived and worked.5 They monitor his in-cell and out-of-cell activities. Most likely, he will be deciphered in a matter of minutes in a new cell. There is an interesting correlation here: one can learn most from those inmates who are most likely to decipher him. Despite all of my precautions, I was "deciphered" twice by my cellmates as a "sociologist who takes notes and does research in prison." In one case, a beating followed. All that occurred despite the fact that I was a true inmate, that my research was only a by-product of my role, and that I knew both the argot and prison norms well.

Sources of Data

My data sources can be sorted into a few categories: (i) Living through various inmate roles; (ii) Informal evening tea chats; (iii) Secret code training of grypsmen candidates; (iv) Informal conversations with inmates, typically face-to-face; (v) Prison artifacts such as pictures, songs, letters, and hand-made products; (vi) The memoirs and written relations of political and criminal prisoners and conversations with former political prisoners; (vii) Underground Solidarity research reports on prisons and uncensored Warsaw University working papers and officially released statistical data.

The data were collected over five months of imprisonment in thirteen cells of two jails, including three police station and court cells. I met about 190 inmates and developed some form of close relationship with about 140 inmates (see table I.1). In references to the sources of my data that appear in the text, I provide the cell number where the relevant data were collected.

I recorded data using various means. Note-taking in prison is extremely difficult. Prisoners prohibit and punish cellmates for writing down any account of cell activity. Guards try to seize the notes on the inmate's way out and during routine searches of the cell. Writing standardized regular daily reports was impossible. Initially, I secretly recorded my observations and argot vocabulary on scraps of paper, often while all of the other inmates went for a walk, trying to hide the notes before their curious eyes. A few notes were destroyed, but later reconstructed. Once I received a beating; however, in most cells after a few days my writing was tolerated and confused with "studying." I took some notes in English, a truly foreign language for most inmates. The "English lessons" that I offered voluntarily to inmates also served as a cover-up. No paper document was ever seized by prison personnel.

Copies of the notes were prepared before expected family visits, my principal smuggling channel. Next, these media were placed in a specially prepared shoe or in underwear, taken to the visiting room, and smuggled out of prison for safe storage. As a backup, I mailed numerous letters to my family with descriptions of those aspects of prison life that could make it through the prison censorship. The first set of notes was expedited as a secret message with the help of a fellow inmate as an apology for an unjustified beating. A few notes were smuggled by a helpful guard who was recruited by my sister.

Another useful technique was borrowed from Solzhenitsyn.6 Every evening, before falling asleep, I repeated all of the new words, rules, customs, jokes, little games, or self-injury techniques that I had learned during that day. An extra benefit of this routine was a fast socialization to the inmate life.

My prison experience was so intense that after my release it allowed me to reconstruct many of the crucial interactions and facts with great precision. For more than one year after release, all of my night dreams revolved around prison facts, events, and people. Over the period of 2.5 years, I wrote down all memories that were not recorded previously and assembled them into a small archive including an argot dictionary and catalogs of initiation tests and techniques of self-injury. I completed five term papers, about ten short prison stories, three interviews and roundtables for the underground Solidarity press, and a Master of Arts thesis in sociology.

The four principal sources of data are discussed below in more detail.


I went through the social roles of rookie (twice), humiliated rookie, potential sucker, aproposman, grypsman,7 self-injury expert, faker, and tough political prisoner. Among the major inmate roles that I did not experience were fag, squealer, corridorman, elder, fuss-master, cat, and jumper.

In addition to all of the routine cell activities, the catalog of social situations I experienced includes interrogation, family meetings, transport, conversations with all ranks of prison personnel, conflict over the status of a political prisoner with guards and warden, punitive reports, help from a guard and physicians, the company of recidivists and juveniles, tea infusion drug effects, losing all belongings in poker, thievery of my tea and other goods, bridge and chess marathons, a beating, an involuntary haircut, a squealer's intrigue, a fag's threat, masturbation, extensive cell trade, intercell communication, smuggling goods and secret documents out and in as well as all major illegal activities, escape planning, armed escort to a freedom hospital, training to become a professional thief, light self-injury, and the successful management of my own faking game. I was subject to fag-making, baptism, and almost all of the other tests described in the book.

Major situations that I did not experience include hardbed solitary confinement, tattooing, beating during interrogation, contact with a lawyer, homosexual intercourse, group masturbation, a love affair, a sucker or fag's attack, and a hunger protest.

An activist attitude helps enormously with collecting data. I tried to assist with all of the meaningful cell activities, such as tattooing, the production of prison artifacts, playing chess and bridge. When possible, I tried to initiate such activities. A risky idea was the experimental breaking of various subcultural and administrative rules. Such experiments included refusing to enter the walkplace from the prison backyard, shouting anticommunist slogans, drawing anticommunist symbols in the walkplaces, and so on. The punishment was so severe that I quickly abandoned this learning channel.

Evening Tea Chats and Informal Conversations

Sykes regards "the relatively "unstructured" talks with the captors and their captives [as] the most useful [source of data] by far, despite the dangers introduced by a lack of standardization and undoubted biases of selection."8 I concur and consider informal talks and chats as a data source on par with the "research-through-the-role" and secret training.

A great source for learning prison customs and argot were about fifty evening tea chats on prison subculture that I used to initiate and that usually engaged the entire cell. As an incentive for inmates to join a tea chat I provided free czajura, which is a tea infusion that was illegal in Polish jails in 1985. Czajura works as a soft drug, stimulates memory, and sparks long conversations. It worked as an invaluable research device, creating incentives for in-depth conversations on prison matters. I traded most of my prison account money, personal belongings and, especially, Amnesty International's humanitarian packages for tea packets.

Face-to-face conversations were especially helpful in uncovering inmate self-injury and faking goals and plans. My credentials as a political prisoner, basic medical knowledge learned from my parents, who are physicians, and, last but not least, a willingness to listen made me a desirable confidant. In almost all cases, inmates revealed their secrets only after one-week- or two-week-long interactions.

A simple principle of "path-independence" that I applied in all chats was to free my mind from any assumptions that could make my learning process dependent on previously heard stories or interpretations. It helped to compare versions of customs reported by various sources without overweighting the first source. I tried to limit my chatting contribution to questions, motivating signs of appreciation, and declarations of surprise.

Secret Code Training

The best source of most secret norms and argot rules is a grypsmen night training, described in detail in chapter 3. Its goal is to transmit efficiently all of the secret knowledge to grypsmen candidates. In striking contrast to inmate interviews, instructors have incentives to convey concisely the letter and spirit of the code rather than feed the listener with fairy tales. They even punish slow learners! Learning through lectures by prison sages, eager to share all their wisdom with a curious student, is the researcher's dream. The only problem is that night training sessions steal one's sleeping time. After a few half-night long iterations a researcher may be ready to trade all of his data for a few hours of good sleep.

I participated in about 10-15 hours of secret training in cell 6, about 5 hours in cell 8, and about 30-40 hours in cell 13 (see table I.1).


This book can be read by someone who is not interested in game theory or by a game theorist willing to enhance his or her lecture with examples outside of economics or political science. Of course, I wholeheartedly encourage every reader to give game theory a chance. Appendix 1, which briefly discusses some ideas behind games and decisions, may be helpful to set the basic concepts. However, the appendix is obviously not a substitute for a good basic course in game theory. No recommendation is too strong for undertaking this intellectual experience if one is interested in studying social phenomena. Appendix 1 also outlines a formal justification for some of the nonstandard models used in the book. Most importantly, such models include games with no common knowledge, where two players have a different understanding of the game's rules.

All substantive chapters except for the overview in chapter 2, The Constraints of Prison Life, are motivated by a specific class of situations that are associated with various informational characteristics. In Entry (chaper 1) an inmate faces a deceptively simple decision problem that provides a good introduction to his future dilemmas. In his first contact with dwellers of a new cell, he must declare his caste membership. The crux of the situation in chapter 3, Becoming a Grypsman, is the informational asymmetry between old inmates and a rookie. This asymmetry is exploited by the former to test the rookie. In chapters 4 and 5, Prison Code of Behavior and Argot, respectively, various norms and language conventions with varying degrees of secrecy are described. An inmate who passes the initiation tests learns these norms and conventions in a systematic way during night sessions of "prison university." In Everyday Life (chapter 6) an inmate uses skills that are more esoteric and more difficult to teach. He plays against his peers subject to subcultural constraints and tries to entertain himself. When he looks for Sex, Flirtation, Love (chapter 7), his preferences may fluctuate over time chaotically and he may pre-commit himself to temporal celibacy. He displays great ingenuity to satisfy his erotic needs. Chapter 8, Strategic Ailment, is an inmate's action of last resort. He usually makes relevant decisions alone and chooses suffering when no other way of advancing his important goals seems feasible. In Exit (chapter 9) he must quickly deal with the surprise of unexpected release and try to avoid the last hot farewell. The Postscriptum comments on the variants and evolution of the grypsmen subculture.

Numerous conventions appear in the text. Often, I refer to myself as "Student," which was my prison nickname. This convention emphasizes my intention to use my own personal experience as a valuable source of unique data rather than a starting point for reflection or existential speculation. Original material is usually presented with references to "cell n," where n may be an integer between 1 and 13 and refers to one of the cells listed in table 1.1. Universal or secret argot names introduced for the first time, or reintroduced after several pages, are italicized. When the reader needs to look up a particular word quickly, the Glossary of Essential Argot offers key definitions. Argot terms that are self-explanatory are italicized but do not appear in the glossary. Almost all inmates in this book are males. Females, who are usually incarcerated in separate facilities, comprise only a tiny proportion of the prison population in Poland. Their codes of behavior appear to be less complex than those of men.

* * *

A note to the reader: Once I was transferred to an Emergency Room hospital cell. One of my new cellmates was recovering from an operation that removed four pounds of iron from his stomach. He died later after swallowing bedsprings again. Another cellmate was cut off a suicide rope just before he could die. The third, seriously sick with acute pancreatitis, was "doomed to go," according to a "well-informed" guard. All those poor creatures tried to impress the doctors with their symptoms, strategically cried from pain in the night, or pretended the lack of appetite while secretly borrowing food from me.

On Sunday, the day that Polish radio broadcast Catholic mass, I asked the guard to lend us his radio. Listening to mass in prison comforts even the worst sinners and most nonreligious souls. Usually, one may listen to it on Sunday morning in regular cells, but no speaker was installed in our ER. The good-hearted guard, hesitant, took a long look at the four barely alive skeletons, sighed, and said "ok." He disconnected his large box and brought it to our cell for two hours. I installed it and asked my buddies: Do you really want to listen to the mass? "Whatever," they said. They did not care. Then I quickly tuned it to find "Radio Free Europe," the anticommunist radio station funded by Americans. My cellmates, all in plaster and bandages, were delighted. They laughed so hard that I became anxious about their sick stomachs and post-surgery stitches. For two hours, we were laughing and listening to the news from the free world.

Throughout this book, people suffer, die, fight, cheat each other, engage in brutal sex, and make hell out of other mortals' lives. And they laugh. My inclusion of their humor does not make the book less serious. The laughter does not nullify inmates' suffering. The laughter does not redeem their guilt. It only makes their lives more bearable.

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
Princeton Legacy Library
Exam/Desk Copy
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share