“It Depends on the Outcome”: Prisoners, Grievances, and Perceptions of Justice

Abstract

Social scientists have long investigated the social, cultural, and psychological forces that shape perceptions of fairness. A vast literature on procedural justice advances a central finding: the process by which a dispute is played out is central to people's perceptions of fairness and their satisfaction with dispute outcomes. There is, however, one glaring gap in the literature. In this era of mass incarceration, studies of how the incarcerated weigh procedural justice versus substantive justice are rare. This article addresses this gap by drawing on unique quantitative and qualitative data, including face-to-face interviews with a random sample of men incarcerated in three California prisons and official data provided by the California Department of Corrections and Rehabilitation (CDCR). Our mixed-methods analysis reveals that these prisoners privilege the actual outcomes of disputes as their barometer of justice. We argue that the dominance of substantive outcomes in these men's perceptions of fairness and in their dispute satisfaction is grounded in, among other things, the high stakes of the prison context, an argument that is confirmed by our data. These findings do not refute the importance of procedural justice, but show the power of institutional context to structure perceptions of and responses to fairness, one of the most fundamental principles of social life.

A sense of fairness is fundamental to the human condition and plays a significant role in modern social, economic, and political institutions. Many scholars have examined the psychological, social, cultural, and political dimensions of fairness and perceptions of justice. A plethora of empirical findings in the sociolegal field documents the relationship between people's perceptions of fair legal processes and their satisfaction with legal outcomes (Carman 2010; Casper et al. 1988; Hasisi and Weisburd 2011; Hollander-Blumoff and Tyler 2008, 2011; Jackson et al. 2012; Johnson et al. 2014; Lind and Tyler 1988; Thibaut and Walker 1975; Tyler 1984, 1988, 1990, 1994, 2003; Tyler et al. 2014; Vidmar 1990). According to this scholarship, when people perceive a decision-making process to be procedurally fair, they are likely to be satisfied with the outcome even when it does not favor them.

There is, however, a glaring gap in this otherwise extensive and varied literature. In this era of mass incarceration, studies of how the incarcerated weigh procedural and substantive justice are rare. We thus know relatively little about how those who are held in today's mammoth prison system perceive the fairness of disputing procedures inside prison walls and of the criminal justice system in general. This gap in the literature is particularly glaring given that prisons are literally places where “justice” is supposed to be carried out.

This article addresses this gap by drawing on unique data to unpack the complicated nature of perceptions of fairness and justice in the prison context. Our primary data consist of confidential, face-to-face interviews conducted with a random sample of 120 men incarcerated in three California prisons. The grievance process is the administrative means that prisoners in the United States may use to contest the conditions of their confinement. This disputing process thus provides an empirical lens through which to analyze the relative importance of procedural and substantive justice to men in prison.

Our findings are at odds with some of the prevailing procedural justice literature: not only are actual grievance outcomes more important to these prisoners’ satisfaction than their perceptions of a fair process are, but in many cases the former drives the latter. Thus, when a grievance outcome does not go their way, as is usually the case, these prisoners infer that the process was unfair (and vice versa). Indeed, in our interviews with them, these men were often hard-pressed to make any distinction between an unfavorable outcome and an unfair process.

The arguments we advance to explain these deviations of our findings from those that dominate the procedural justice field are contextual in nature. Prison is a hierarchical total institution where the stakes are high, where autonomy for prisoners is deliberately curtailed, and where prisoner appellants rarely prevail. We argue that these high stakes, limited autonomy, and asymmetrical power relations comprise an environment in which the outcome of a prisoner's grievance can sometimes literally mean life or death. In this context, the possibility of procedural fairness is either scoffed at by prisoners or subordinated to the more important goal of, as prisoner James Little told us, “getting what you're supposed to have coming.” This constellation of characteristics is in some ways unique to the prison environment, and we do not argue that our findings are directly generalizable to all settings. Nor do we argue that procedural justice is not important to these prisoners. To the contrary, we are advancing an empirically driven argument for more context-specific research into perceptions of justice—research that may reveal that some aspects of the prison environment are applicable to other settings, including other total institutions (Goffman 1961).

In previous work (Calavita and Jenness 2015), we describe the grievance process in detail and document the “pyramid of disputes” as prisoners name, blame, and make claims about problems in prison, as well as exploring both prisoners’ and staff's framing of grievances (see also Jenness and Calavita 2017). However, that work does not address the critical issue of how these prisoners perceive fairness and justice in the disputing process, specifically how they weigh procedural and substantive justice. Our current focus on prisoners’ perceptions of disputing justice presents a contrast to findings in much of the procedural justice literature and reveals the perils of over-generalization. As such, it underscores the profoundly contextual nature of perceptions of justice.

In the next section, we outline the mechanics of the California prisoner grievance process. Following that, we situate our central analytic concerns in the larger literature on procedural justice. We then provide an overview of our research site, data, and methods of analysis. Next, we turn to our findings and analysis. Throughout our analysis, we highlight the differences between our findings and those that dominate the procedural justice literature, and bring to bear a “situated justice” perspective (Berrey et al. 2012) on those differences. In the conclusion, we make the case for future research on procedural and substantive justice perceptions in various other venues.

Mechanics of the Prisoner Grievance System

A prisoner grievance system¹ was established in California in 1973 at the height of the prisoner rights movement. Prisoner rights advocates and others advocated for an appeals system on the grounds that it would provide a nonviolent avenue for prisoners to contest the conditions of their confinement, and would minimize prisoner riots such as those that occurred at the state prison in Attica, New York in 1971. Officials who put the grievance system in place in California, however, refused the recommendations of a California Senate Commission (Berke and Dillard 1973) to appoint a director of legal affairs to manage the system, appeals officers with legal training, or independent external ombudsmen to provide oversight. In the two decades that followed, the California grievance system—operated then, as now, entirely by the California Department of Corrections and Rehabilitation (CDCR) officials with little legal training—was rarely used by inmates who more often took their complaints directly to the courts.

As the size of the prison population in the United States soared in the 1980s and 1990s, courts were flooded with prisoner lawsuits (Schlanger 2003: 1558). In response, in 1996 Congress passed the Prison Litigation Reform Act (PLRA), the explicit purpose of which was to restrict prisoners’ access to court. Among other provisions, the PLRA mandated that “no action shall be brought with respect to prison conditions … by a prisoner confined in any jail, prison, or other correctional facility until such administrative remedies as are available are exhausted” (42 U.S.C. Section 1997e[a]; emphasis added). This mandate, coupled with a continuing rise in the prison population, resulted in a dramatic increase in the number of prisoner grievances.

The states have been given free rein by the PLRA and subsequently by the courts to design grievance systems that in practice are often cumbersome, confusing, and laden with what one judge has called “stalling tactics” (Campbell v. Chaves 2005; see Schlanger 2003; Schlanger and Shay 2008). In 2011, in his report on California's grievance system, California Inspector General Robert Barton identified several areas of serious concern (Barton 2011). A survey of 39 states and the District of Columbia reveals that California is by no means an outlier in this regard (cited in American Civil Liberties Union 2006: 43–44).

In California, inmate grievance procedures are specified in Article 8, Title 15 of the California Code of Regulations. To file a grievance, prisoners are required to complete an Inmate/Parolee Appeal Form—officially numbered and colloquially referred to as a “602.” When we began this research, inmates had to attempt to resolve their grievances informally with the relevant staff as the first step in filing a 602. In some cases—such as an allegation of staff misconduct, contestation by the inmate of a serious disciplinary violation, or an inmate complaint about his/her classification—this informal step could be waived. If the informal effort was unsuccessful, or if it had been bypassed, the 602 was subject to three potential levels of formal review. The informal level of review was eliminated in 2011, but the subsequent three formal levels remain largely unchanged.

At the first formal level, the prisoner's narrative of his/her complaint and proposed remedy is limited to one paragraph on the 602 form, plus a potential additional page, front and back; supporting documentation may be attached. The prisoner forwards this to the Inmate Appeals Coordinator (IAC) at the prison, a position commonly held by a CDCR correctional officer. The IAC is charged with screening each grievance to determine if it is in compliance with CDCR regulations, and returning to the appellant those that do not conform to strict requirements. Among the most important of these requirements is that the 602 be filed within 15 days of the incident being contested. If the reason for the screen-out can be corrected, the appeal can be revised and resubmitted. The IAC then usually forwards the grievance to the official most relevant for the complaint for a response. The response can take one of three forms: granting the requested remedy in full, granting the requested remedy in part, or denying the requested remedy. The official response and justification are entered onto the 602 form beneath the prisoner's narrative. If dissatisfied with this outcome, the prisoner may explain his/her dissatisfaction beneath the official's response and submit the grievance for a second level of review, conducted by the Warden or his/her designee, such as the Deputy Warden. Once again, the grievance may be granted, partially granted, or denied.

If the appeal is not fully granted at the second level, the prisoner may request a third-level review by the Office of Inmate Appeals in Sacramento, California. There, each grievance is read and responded to by an examiner who may or may not do further investigation. Such investigation may consist of telephone calls to officials at the prison where the complaint originated or document retrieval, but rarely involves a trip to the prison. The Chief of Inmate Appeals informs the appellant in writing of the final decision, making it clear that the appellant has exhausted all internal remedies (for more details on the inmate appeal process, see Calavita and Jenness 2015).

The most common complaint is related to medical treatment, followed by contestations of disciplinary actions taken against the prisoner, missing or damaged property, concerns about disability accommodations, complaints against staff, and a host of other issues such as work detail problems, mailroom complaints, living conditions, etc. The vast majority are denied, with only 0.2 percent of the 15,836 grievances that made it to the third level in 2005–2006 fully granted. While there are no official statistics on fully granted grievances at lower levels of review, we estimate that 2.3 percent and 1.9 percent, respectively, were granted at the first and second levels (for further discussion of the frequency and content of complaints as well as estimates of outcomes, see Calavita and Jenness 2015).

Despite this low level of grants and extensive critiques of the PLRA and the internal remedy system it mandated, the prison officials interviewed in this study uniformly lauded the system as fair to prisoners, praised the social control aspects of the grievance system, and embraced the presumed role the grievance system plays in protecting the CDCR against lawsuits (Calavita and Jenness 2015). Hand-in-hand with an often expressed disdain for prisoners, they told us, as one grievance examiner put it, “Inmates can, do, and should have rights.” Perhaps more surprisingly, although the prisoners we interviewed were often cynical about the grievance system, they repeatedly referred to the rights accorded them in the Title 15, and their expectation that if they properly cited rules and had evidence, they could win a grievance. Adolfo Flores told us, “There's certain rules that they go by… and so if the Title 15 says that's what we got coming, then they [grievance responders] can't go against it.” In other words, in spite of the structured inequalities in the grievance system, officials saw it as a fair expression of prisoners’ rights and prisoners clung to an on-again/off-again belief that it could work.

Procedural Justice, Substantive Outcomes, and Legitimacy

Much of the scholarly literature on the concept of procedural justice and its impacts comes from a social psychological perspective. In early work, Thibaut and Walker (1975) and Walker et al. (1979) reported from their studies of various disputing procedures—based primarily on laboratory experiments using student subjects—that people's perceptions of whether a disputing process is fair (“procedural justice”) affects people's satisfaction with the dispute, whether or not the outcome favors them. As Brunton-Smith and McCarthy (2016: 1033) summarized it, this scholarship emphasizes that process is more important than outcome in determining people's satisfaction and the legitimacy they accord decision makers.

The concept of procedural justice and the theoretical framing of its impact have been elaborated and empirically tested for decades. Thibaut and Walker's (1975) initial conceptualization focused on whether people felt a sense of control over the process—for example, through participating in the proceedings—and through the proceedings’ perceived neutrality. Subsequent work added the “normative” (Tyler 1990: 7) dimension, which includes interpersonal treatment and respect. Blader and Tyler (2003) and Tyler et al. (2007) later developed a comprehensive four-component model that includes “the procedural justice values of voice, participation, neutrality, and acknowledging the rights, needs, and concerns of people involved in the decision” (Tyler et al. 2007: 470). Adding nuance to this basic four-component model of procedural justice, Hollander-Blumoff and Tyler (2011: 5) included trustworthiness of the decision maker(s) in the mix. Specific details of the model aside, this research reveals that, to quote Baker et al. (2015: 435), “even when negative outcomes occur the individual may be more … satisfied with the results and the decision makers if the process is viewed as equitable.”

Numerous scholars have found that some version of this concept of procedural justice is positively related to people's satisfaction—regardless of outcomes—in a variety of venues, including the workplace (Tyler and Blader 2000; Tyler et al. 2007), criminal courts (Casper et al. 1988; Landis and Goodstein 1986; Tyler 1988), traffic and misdemeanor court (Tyler 1984), police encounters (Tankebe 2009; Tyler 1990; Tyler et al. 2014), contract disputes (Hollander-Blumoff and Tyler 2008), tort claims (Lind et al. 1990), airport screening (Hasisi and Weisburd 2011), and attitudes toward government (Carman 2010; Tyler 1994). Over the last few years, researchers have argued that this positive relationship between perceptions of procedural justice and satisfaction with outcomes is not unique to the United States (Carman 2010; Hasisi and Weisburd 2011; Jackson et al. 2012; Tankebe 2009).

This formulation of the dominant importance of procedural justice is often coupled with a concern for legitimacy. Specifically, it is argued that people's perceptions of procedural justice are important contributors to their satisfaction with outcomes—independent of the substance of those outcomes—and in turn are a key component in the legitimacy accorded the relevant authorities and institutions (Carman 2010; Hollander-Blumoff and Tyler 2011; Lind and Tyler 1988; Reisig and Mesko 2009; Tankebe 2009; Tyler 1990; Tyler and Blader 2000; Tyler et al. 2007, 2014; Vidmar 1990). In their study of people's attitudes about police stops and police legitimacy, Tyler et al. (2014: 775; emphasis added) report that “perceived procedural justice” was “the most important observed element that respondents reacted to … .”

A number of criticisms have been launched against this body of research and a few issues remain unsettled. Some have argued that findings based primarily on simulations, vignettes, and experiments with college students do not mimic real life (Anderson and Hayden 1980–81; Heinz 1985; but see Vidmar 1990: 881). Others note that procedural justice is not consistently and precisely defined across studies (Bottoms and Tankebe 2012; Johnson et al. 2014).

Heinz (1985) has questioned whether distinctions between and among crucial variables in the procedural justice formulation are empirically valid. She suggested that the line between subjects’ perceptions of procedural justice and whether they believe their outcome is fair may be blurred when the stakes are high. Noting that early studies of the effect of procedural justice were conducted primarily in the laboratory, she reasoned that the stakes in those cases were hypothetical, which may have enabled subjects to make abstract distinctions and to privilege the principle of procedural justice that dominates American jurisprudence. To test this hunch that the importance given procedural justice may depend on how high the stakes are, she conducted a field study of felony plea bargaining, and found that the police and victims in these cases made a distinction between procedural fairness and outcomes, but that for the defendants—with arguably more at stake—“outcome and procedural measures formed a single factor.” Casper et al. (1988) and Napier and Tyler (2008) have responded to findings that the importance of procedural justice perceptions is variable. Based on their study of convicted felons, Casper et al. (1988) admit that the severity of outcome may sometimes temper (but not trump) the impact of procedural justice. Likewise, Tankebe's (2013) work on the legitimacy of policing reveals that outcomes may be related to overall levels of legitimacy, and Brunton-Smith and McCarthy's (2016: 1048; emphases in original) recent work on prisoners’ perceptions of the legitimacy of prisons concludes: “Our data leave open the possibility that prisoners may regard the outcome of the sanctions, rather than the process of delivering the sanctions as illegitimate.”

A few recent sociological studies are relevant to this question. For example, Berrey et al. (2012) develop a “situated justice” approach to procedural and substantive justice, and examine the possibility that those with high-stakes may not make a distinction between them. Drawing from interviews with plaintiffs, lawyers, and defendants’ representatives in employment discrimination cases, they found that participants entertained an abstract notion of fairness, but that what that meant in concrete cases was “relational” (Berrey et al. 2012: 29). That is, participants “only talk about the fairness of the parts of the process that advantage their opposition and disadvantage themselves … . From a situated perspective, what each side wants in a fair legal system is not an unbiased process (as the procedural justice literature suggests) but one that benefits their own side” (Berrey et al. 2012: 4). Anticipating this finding, in their 2003 study of police responses to domestic violence incidents, Hickman and Simpson (2003: 628) also found support for “the outcome hypothesis.” As they conclude, “even while controlling for the effects of other variables associated with procedural justice ratings, the counter-theoretical finding could not be eliminated.”

Partly by way of explaining this discrepancy from the prevailing literature on procedural justice, Berrey et al. (2012) argue that the literature is largely based on individual encounters with the law or hypothetical vignettes and is rarely contextualized within institutions, cultural matrices, or structures of power. Their interviews reveal that in real legal cases in which participants are ensconced in institutional contexts and are advantaged or disadvantaged by resource inequalities and historical experiences with law, the abstract concept of fairness is disrupted. Indeed, their findings show that in these cases participants often do not distinguish between how a decision is arrived at and what that decision is, “complicat[ing] the distinction between process and outcome that is a mainstay of the procedural justice literature” (Berrey et al. 2012: 30).

Confirming the importance of institutional context, historical experience, and power inequalities, Epp et al. (2014: 4), in their book on police stops of African Americans and others, “emphatically depart” from psychological procedural justice theories that place primary importance on officers appearing to act fairly (see, for example, Mazerolle et al. 2012). They argue instead that for African-American motorists, “official politeness could not convert an otherwise offensive police stop into a fair and legitimate one” (Epp et al. 2014: 5; cf. Geller et al. 2014; Tyler et al. 2014). Epp et al.'s study once again illustrates the power of social context in evaluations of fairness and justice and in attributions of legitimacy.

A separate but related strand of research into perceptions of fairness (and their relationship to legitimacy) comes from prison studies. At issue in that literature is the question of what determines perceptions of procedural justice in prison and what role procedural justice and legitimacy play in the maintenance of prison order, compliance, and reoffending (Beijersbergen et al. 2015a, 2015b; Bierie 2013; Bottoms and Tankebe 2012; Brunton-Smith and McCarthy 2016; Carrabine 2005; Jackson et al. 2010, 2012; Reisig and Mesko 2009; Sparks and Bottoms 1995; Sparks et al. 1996; Useem and Kimball 1989). For example, Beijersbergen et al.'s (2015a, 2015b) work reveals the importance of staff characteristics on Dutch prisoners’ perceptions of procedural justice; Bottoms and Tankebe (2012) advance a theoretical understanding of the complexities of legitimacy in the prison context and its relationship to perceived justice; Beijersebergen et al. (2015) examine empirically the relationship among perceptions of procedural justice and recidivism among Dutch prisoners; and most recently, Brunton-Smith and McCarthy (2016) investigate the relationship between perceived procedural justice and legitimacy among prisoners in England and Wales.

Bierie's (2013) work on grievance procedures in federal prisons in the United States focuses on their relationship to violence—specifically whether rates of grievance denials and/or rates of procedural justice, impact prisoner violence rates. Bierie does not measure prisoners’ perceptions of procedural justice; instead, he relies on official data on late responses and screen-outs as indicators of procedural justice. Exploring the intricacies of and conditions for legitimacy and its consequences for prisoner compliance, as these prison studies generally do, is important work. However, these studies do not examine the relative importance of actual outcomes and process in prisoners’ satisfaction with official decisions, and whether those abstract dimensions of justice—so pivotal in the procedural justice literature—are meaningful to prisoners.

We address these issues first through an examination of prisoners’ judgments of how fairly they have been treated by the criminal justice system in general—and what their implicit barometer of fair treatment is in that judgment. Next, we examine the impact of certain external dimensions of procedure (whether there was a hearing, whether the prisoner was interviewed, and whether the prisoner was informed of the outcome) on prisoners’ satisfaction with grievance outcomes. We follow that with an investigation of another, more subjective, dimension of procedure (i.e., prisoners’ satisfaction with how the grievance was managed) and how that correlates with prisoners’ satisfaction with grievance outcomes. Finally, we analyze the impact of actual grievance outcomes (whether or not it was granted) on prisoners’ satisfaction with the outcome. We find that substantive justice—not procedural justice—dominates these prisoners’ perceptions of fairness and drives their satisfaction with outcomes. To a large extent, the outcome defines for them whether or not the process was fair and how satisfied they are with the decision. This finding is inconsistent with the procedural justice literature's argument that people's satisfaction with official decisions depends on their perception of a fair process, independent of outcomes, and it theoretically affirms the importance of institutional context in people's perceptions of justice. We turn now to a discussion of our data and methods.

Data and Methods

Many scholars have lamented the difficulties of accessing prisons as research sites and the subsequent paucity of in-prison research (Goodman 2008; Simon 2000; Wacquant 2002). Indeed, our access was fraught with bureaucratic and legal obstacles that required more than two years of persistence and some measure of good fortune to surmount. Ultimately, we were allowed inside California's prisons to collect original interview data from prisoners and prison staff.² We were also allowed inside the administrative archives where tens of thousands of prisoner grievances and hundreds of prison records are kept, providing us with a rich source of archival data. These data reveal the voices of prisoners and staff, the problematic conditions of prison life, and the administrative work done by the prison officials who respond to prisoner grievances—topics covered in our larger project on the prisoner grievance system in California (Calavita and Jenness 2015). In this article, we focus on the prisoner interviews. The interviews open an empirical window through which to examine issues not yet explored in depth either in our own work or in the prevailing literature: prisoners’ perceptions of fairness in general, and the relative weight of procedural and substantive justice in particular.

At the time that we conducted this study, California administered the largest prison system in the United States, with 33 prisons housing approximately 160,000 prisoners. Soon after we completed data collection, the U.S. Supreme Court ruled that inadequate mental and physical health care in California prisons due to overcrowding violated the Eighth Amendment's ban on cruel and unusual punishment (Brown v. Plata 2011). California is by no means unique in the nature of the problems confronting it. The Commission on Safety and Abuse in America's Prisons (Gibbons and Katzenbach 2006: 390; see also Travis et al. 2014) reported finding “facilities that are crowded to the breaking point, too little medical and mental health care, unnecessary uses of solitary confinement and other forms of segregation,” and a host of other recurring issues. Thus, while the size of California's prison system is unusual, the problems it faces are shared by prison systems across the United States.

The analyses presented in this article draw from our interviews with 40 randomly selected prisoners in each of three men's prisons in California (n = 120). The three prisons from which we recruited prisoners to interview include a minimum, medium, and maximum security prison; they vary by age of facility, distribution of custody levels of the population, population size, number of grievances that reach the third level of review, overcrowding levels, and violence rates. Our diverse sample of prisoners drawn from these three prisons roughly approximates the larger CDCR male population on several important dimensions (see Table 1). Further, the array of complaints expressed in these grievances (relating to medical needs, disciplinary actions, missing or damaged property, disability accommodations, and staff misconduct, and other things) corresponds closely to the types and proportions of complaints recorded in the overall population of thousands of prisoner grievances.

Table 1. A Comparison of Select Characteristics of the Study Sample and the Total Population in CDCR Prisons for Men^a

Source: CDCR's Offender-Based Information System, March 2009.

^a Based on official CDCR records and attendant classification schemes.

^b Figures exclude inmates in fire camps.

^c The total used to compute percentages varies depending on missing data. For example, the percentages for custody level are based on a total of 128,956 prisoners with official classification scores.

^d Sample mean = 38.3 years; Population mean = 37.4 years.

^e Correctional Clinical Case Management System (CCCMS) and Enhanced Out-Patient Program (EOP) inmates, referring to those with chronic or acute mental health disorders.

The authors of this article and three advanced graduate students traveled to each of the three prisons from which random samples of prisoners were drawn, obtained informed consent from prisoners who agreed to participate in the research, and conducted 120 face-to-face interviews. The co-authors of this article conducted the majority of interviews (60 percent), and the graduate students conducted the rest. A Spanish-speaking graduate student interviewed Spanish-speaking prisoners. These confidential interviews were conducted one-on-one in closed rooms, with no officers or anyone else in the room. It is always possible in research that interview respondents may “perform” for an interviewer or be less than candid. We did our best to minimize this ever-present possibility through demeanor and language that signaled acceptance and our trustworthiness (among other things, stressing that we had no connection to or relationship with the CDCR). While performing is inevitably part and parcel of social life, as Goffman (1959) long ago taught us, we repeatedly got the sense that our outsider status enabled our respondents to be candid in their conversations with us. It was not uncommon, for example, for them to share with us their anger, sadness, and remorse, and to tell us about embarrassing moments.³ Indeed, antithetically to what one might expect in the hyper-masculine environment of a men's prison, these men often spoke to us in ways that revealed vulnerability. For this reason, we are confident that our findings reflect these prisoners’ actual perceptions and not a strategically constructed presentation of self—at least no more than that which occurs in any social interaction.

Ninety-three percent of those we recruited to participate in the research agreed to do so, and almost all of the prisoners (91 percent) agreed to have their interviews recorded. The average interview length was slightly over one hour, with some lasting well over two hours. After the interviews were completed, they were transcribed and entered into a data base. We then concatenated the interview data to official data from the CDCR's database on prisoners, including demographics, length of sentence, etc.

During the interview, we asked prisoners a series of open- and closed-ended questions about prison life. For example, we asked them about housing arrangements, daily life, problematic or bothersome conditions, perceptions of and experience with the prisoner grievance process, and how fairly they thought they had been treated. While responding to these questions, prisoners frequently offered unsolicited commentary on an array of issues. Some of their comments elaborated further on how they think about and orient to fairness in the criminal justice system in general and the grievance system in particular. We coded their interviews along a variety of dimensions, including how respondents spoke about satisfaction with the management and outcome of grievances as well as how fairly they thought they had been treated by the criminal justice system in general and in the grievance process in particular. Our team of four trained research assistants achieved over 90 percent intercoder reliability.

If the respondent had filed any grievances, we asked a series of questions about the specific grievances he had filed. An incident form was used to collect systematic information on the 217 grievances these prisoners reported filing. To ensure that a range of types of grievances was included in our incident data, we asked each interviewee to provide details on different types of grievances he may have filed, including: (1) A grievance filing that ended at the informal level; (2) The most recent grievance filed; (3) The second most recent grievance filed; (4) The most important/memorable grievance filed; (5) A grievance that was granted; and (6) A grievance that was filed with other prisoners as a group appeal. After each specific incident was reported, we asked questions that enabled us to capture key elements of procedural justice, including: (1) “How satisfied are you with the way it was managed” and “why”; and (2) “How satisfied are you with the outcome” and “why.” These questions allowed prisoners to provide their assessments of process and outcome by indicating whether they were “very satisfied,” “satisfied,” “neither,” “dissatisfied,” or “very dissatisfied.”

We also asked prisoners about specific aspects of procedure, variations of which are discussed in the procedural justice literature. Specifically, we inquired whether the grievance was granted an official hearing, whether the prisoner was officially interviewed as part of the grievance process, and whether the interviewee was informed of the outcome of the grievance. In the context of prison, where rights and respect are severely curtailed, these elements of procedure provide at least a minimum level of the “participation,” “voice,” and respect, which are central dimensions of procedural justice in the literature. Definitions and the attendant operationalization of procedural justice have varied across the literature (Bottoms and Tankebe 2012), no doubt in part according to the venue of the particular study and its constraints. Our indicators of procedural justice in prison are not as expansive as sometimes used in the larger literature, and this is in part the result of the constricted environment of prison. In the context of prisoner grievances, these are the most concrete external indicators of procedural justice available.

In order to test whether the severity of potential grievance consequences impacted prisoners’ evaluations of process and outcome, the incident data were coded according to how high the stakes were. We reasoned that medical and disciplinary grievances generally entail particularly high stakes. Because medical grievances relate to physical well-being and sometimes speak to matters of life or death, we coded them “highest stakes.” We also coded disciplinary grievances—in which prisoners contest disciplinary actions taken against them for alleged misconduct—as “highest stakes” because they have the potential to impact negatively what many prisoners told us they value above all: their release dates. In the prison context of generally high stakes, grievances that were not medical or disciplinary were coded as “other.”

Finally, we relied on analytic abduction as a methodological strategy to advance our empirical analysis and contribute to theory development. Unlike grounded theory in the traditional sense (Glaser and Strauss 1967; but see also Charmaz 2006), analytic abduction is purposely informed by existing theory and findings (Timmermans and Tavory 2012). In this case, our analysis is informed by decades of research on procedural justice.

In the next section, we present findings from our qualitative and quantitative data, including thematic findings from our interviews as well as standard descriptive statistics, chi-square tests, and logistic regression analysis. This mixed-methods approach allows us to engage in triangulation to identify, assess, and illuminate the relationship between actual grievance outcomes, prisoners’ satisfaction with the grievance process and outcome, and their views of fairness more generally.

Findings

Prisoners’ Evaluations of Criminal Justice System Fairness

The men we interviewed described a host of conditions in prison—from untreated medical conditions to missing personal property and dilapidated facilities—that they deemed problematic and, in some cases, a violation of their rights. But, despite the dire situation in which they find themselves, these men do not unilaterally report that the criminal justice system has treated them unfairly. When asked, “Looking back over your experiences in life and things you might have done, do you think you've been treated fairly by the criminal justice system?” Over one third (38.7 percent) said they had been treated fairly overall; another 2.5 percent responded to the question with some version of “it depends.”

Orlando Martínez,⁴ a Hispanic prisoner who was 13 years old the first time he was arrested, 14 when he was put in juvenile detention, and in his late twenties when we interviewed him, told us he had been treated more than fairly:

I want to base it on a lot of stuff that I've done, the criminal justice system gave me opportunities … . Like, for example, I got possession of a controlled substance for sale while armed … and instead of sending me to prison right away they give me opportunities, occupation, Prop. 36 [a drug diversion program], drug programs that are actually helping my addiction, and yeah dude, that's how they give me chances and opportunities … . I've had a lot of chances, a lot of chances.

Daniel Sánchez, a prisoner from Mexico also said he had been treated fairly, adding, “Because I know what I did was wrong, and I've been treated good.” When the interviewer asked, “What makes you think you've been treated good,” he replied without hesitation, “Compared to Mexico.” Jorge Bermúdez, a young Hispanic prisoner who told us he could have been sentenced to life in prison, said he had been treated fairly because he had received “just twelve years.” Terrell Davis, a Black prisoner, similarly said he had been treated fairly: “I coulda got a lot more time, but I feel like I was treated fairly. For the crime that I committed, they gave me eight years. I think I was treated fairly.” This theme of having been given a break by the criminal justice system emerged quite often in our conversations with these men.

Many others reported that they had been treated fairly because they were being punished for the crime they committed and that they deserved to be in prison. Mark Crenshaw, a White prisoner on a Sensitive Needs Yard (SNY),⁵ said, “We're in here because we're being punished … this is our punishment … . The minute I came to prison I fully understood that.” Martin Tate, another White prisoner, said with some exasperation, “This is a prison and it's not Disneyland. They're not here to make you comfortable. We did a crime and this is what we get.” Harold Steele, a White Vietnam Veteran diagnosed with post-traumatic stress disorder, in prison for 28 years when we interviewed him, also told us he had been treated fairly: “Yes. I killed two people. I got two life sentences. I don't like it, but I was found guilty, and actually two 15-to-life sentences is better than two 25-to-lifes.” Frank Thurmond, a Black man in for robbery put it this way: “Yeah. I think I've been treated fairly because I've never just been put in prison or in jail for no reason at all … . I've done things to be put in prison.”

In addition to these expressions of “just desserts,” others told us they had committed crimes for which they had not been caught and so things “even out.” When asked if he had been treated fairly, Frank Boutín, a Mexican man in prison 15 years at the time of his interview, replied, “You know what? I gotta say yes. Because I've done a lot of bad things in my life and for the things I haven't been caught for, I believe that it evens out.”

Some of the same themes touched on by those who said they had been treated fairly played out in reverse among the many men (close to 60 percent of our sample) who said the system had not been fair to them. For example, some objected to policies that enhance sentences based on past behavior. Reginald Thompson, in and out of prison since he was 18, told us the system was fair to a point, but, “I still have problems with how they've added all these other laws to enhance a person. Like, for instance, how they take my past, prior convictions, and use them against me … . Convict me on the crime that I come here for. Don't convict me again on my past … . It's not fair at all.” Many others also believed that their sentence was too harsh. Some faulted their public defenders. A rare few made broad references to violations of due process, such as James Little, a Black man in prison for 10 years at the time of his interview: “The Board of Prison Terms is a kangaroo court. They go on hearsay.”

Two points about these comments are striking. First, when discussing whether they had been treated fairly, very few of these men referred to what we might think of as procedural justice. Even though most of them focused on their arrest, prosecution and sentence, they rarely made reference to issues of due process during trial, getting their say before a neutral court, or any of the other dimensions of procedural justice so often cited in the literature as paramount to people's satisfaction. Instead, these men privileged outcomes at almost every turn. Whether talking about being given second chances, getting a break on their sentence, deserving what they got, the “evening out” of justice, or receiving an overly long sentence, it was substantive justice that dominated their thinking.

Second, despite the adversarial environment of prison, a significant minority of these men thought the system had been fair to them. They are by no means unilateral “whiners” or “narcissists” as some corrections staff refer to them (Calavita and Jenness 2015). On the contrary, many seemed determined to be fair in their judgments of the system that had put them behind bars.

Prisoners’ Evaluations of the Prisoner Grievance System

Almost three-quarters of the men in our study (74.2 percent) told us they had filed at least one grievance while in a California prison. More than three-quarters of these had filed more than once, some had filed dozens of claims, and two said they had filed “hundreds” of times. Further, a majority of men in every demographic and racial group had filed a grievance. When asked why they filed grievances, the vast majority of prisoners indicated some version of “to get what I've got coming,” often citing the Title 15 to reference what they have been told to expect. When we asked specifically about other potential reasons for filing, very few responded that they had filed, for example, “to get to court” (7 percent), “to get back at an officer” (5.7 percent), and/or “to get something off your chest” (10 percent). In response to these specific questions and in unsolicited comments, the prisoners expressed understanding that winning a grievance is a long shot, but, to quote one prisoner, “It's all we got.”

To begin our analysis of the impact of procedure on satisfaction with outcome, we coded each grievance for three specific elements of procedure: whether there was an official hearing, whether the prisoner was interviewed as part of the grievance process, and whether the prisoner was informed of the grievance outcome. Taking these dimensions of grievance processing as at least minimal indications of participation, voice, and respect for prisoners who file grievances, we see considerable variation across grievances. In Table 2, zero indicates the absence of all three of these elements and three indicates the presence of all three. Surprisingly, the relative presence of these key elements of procedural justice is not significantly associated with satisfaction with outcomes (Table 2).

Table 2. Frequency of Satisfaction with Outcome and Number of Procedural Justice Elements (n = 164)

Note: Pearson chi-square = 6.0, p = 0.109. Percentages reflect row percentages.

In contrast, whether the grievance was granted (i.e., the actual outcome) is statistically significantly associated with prisoners’ satisfaction with grievance outcomes (χ ² (1) = 56.2, p ≤ 0.001). In other words, outcomes matter significantly to these prisoners’ satisfaction, while specific aspects of procedural justice do not.

We also asked prisoners whether they were satisfied with the procedures followed in the grievance process (“How satisfied are you with how the grievance was managed?”). We found that this more subjective measure of procedural justice (satisfaction with grievance management) is statistically significantly associated with satisfaction with the outcome (χ ² (1) = 76.3, p ≤ 0.001). At first glance, this finding appears consistent with the literature's emphasis on the importance of procedural justice in people's satisfaction with outcomes. A closer look, however, reveals a more complicated reality: actual outcomes (whether the grievance was granted or not) are significantly associated with people's satisfaction with how the grievance was managed (χ ² (1) = 12.5, p ≤ 0.001), and actual outcomes are also statistically associated with satisfaction with outcome, as detailed above. Thus, the relationship between satisfaction with procedure and satisfaction with outcome may be driven by the fact that both are so strongly associated with actual outcomes.

To explore this possibility, we examined separately grievances that our respondents told us had been granted and those that had not been granted. In granted cases, the modal pattern is satisfaction with both outcome and management (69.2 percent) (χ ² (1) = 4.74, p = 0.029). Conversely, in cases that were not granted, the modal pattern is dissatisfaction with both outcome and management (60 percent) (χ ² (1) = 63.76, p = 0.000). Most strikingly, and most inconsistent with procedural justice theory, in the 135 cases that were not granted prisoners who were satisfied with the management of their grievance only reported satisfaction with the outcome in 25.9 percent of them. In other words, in direct opposition to the procedural justice theory that “Procedural justice drives the satisfaction people have with their outcomes” ( Hollander-Blumoff and Tyler 2011: 3), when the grievance was denied even those who were satisfied that it had been managed adequately were not likely to be satisfied with the outcome. What these data show is that when the grievance is granted, the appellant is likely to be satisfied with both the outcome and the procedures through which it was managed; when the grievance is not granted, the appellant is likely to be dissatisfied with both outcome and management.⁶

These statistical findings are consistent with findings from our qualitative interview data. For example, James Little, whose long and thoughtful interview provides many rich insights, was direct and to the point. When asked how satisfied he was overall with the grievance process, he replied without hesitation, “Very dissatisfied.” When the interviewer followed up with, “And why would you say that?” James answered with a tinge of impatience at her apparent obtuseness, “Because it doesn't work.” Understanding that these men are well aware of others’ grievance outcomes and their perceptions of them,⁷ the interviewer followed up again with, “Do you think most other inmates are very satisfied, satisfied, neither, or dissatisfied or very dissatisfied,” he proffered, “Neither satisfied nor dissatisfied, is what I think. It's either going to work or not. It depends on the outcome.”

Outcomes were so important to our respondents that they often obscured the procedural dimensions of grievances, making distinctions between them virtually meaningless. This was brought home to us in a particularly powerful way when we followed the course of our interview schedule moving from questions about how a grievance evolved and how it turned out to questions about how it had been managed. As we dutifully proceeded through the required questions, time after time respondents looked at us quizzically as if we had not been paying attention when we asked about their satisfaction with the management of a grievance after we had asked about the outcome. Sometimes patiently, sometimes with muted impatience, they schooled us. Harold Steele, the Vietnam War veteran with post-traumatic stress disorder, reminded us that his medical grievance had been rejected and that he was subsequently angry about how it had been managed and how unfair it was: “Dissatisfied all the way around,” he said. Steve López, in prison for six years when we interviewed him, similarly reported that his grievance was denied and that he was therefore “very unsatisfied” with the outcome. When asked how satisfied he was with how it had been managed, he laughed and said, as if it were patently obvious, “The same way.” James Little also described his perception that process and outcome were inextricably linked: “It should be about justice,” he said. “Getting it done fairly and getting what you're supposed to have coming.”

These consistent findings across our quantitative and qualitative data elaborate the procedural justice scholarship cited earlier. That literature has generally reported that people's evaluations of procedural fairness predominate in their degree of satisfaction with legal/institutional decision making and are largely independent of whether a decision is favorable to them. As Tyler et al. (2007: 468-69) state:

The procedural justice argument is based upon the belief that people's procedural justice judgments are distinct from their instrumental concerns. That is, their … judgments about the fairness of the organization's procedures are not related to judgments about the outcomes that they receive from their organization.

In contrast, for these prisoners, not only does the actual outcome impact their perception of process but the former drives the latter to such an extent that there is little meaningful distinction between them. As we detail elsewhere (Calavita and Jenness 2018), these findings hold even among racial groups that in other contexts are reported to have radically divergent attitudes about fairness in the criminal justice system.

Situated Justice in a Carceral Context

As we have seen, the men in our study made little distinction in their perceptions of grievance process and outcomes, with positive outcomes generally defining for appellants both fair process and their satisfaction. A number of factors might explain this departure from the findings reported in much of the procedural justice literature. A few hints are scattered throughout that literature. Brockner and Wiesenfeld (1996) analyzed 45 studies examining the relationship between perceptions of procedural justice and outcome and concluded that “the combination of low procedural fairness and low outcome favorability engenders particularly negative reactions”—reactions that disturb any distinction between them beyond the purely conceptual. Early on, Tyler (1990: 107) also touched on the debilitating consequences of consistently negative outcomes, proffering, “If a ‘fair’ procedure continually delivers unfavorable outcomes, its fairness may ultimately come under scrutiny.” Or, as Epp et al. (2014: 6) paraphrase Tyler, “[A] procedure that consistently produces unfair outcomes will eventually be viewed as unfair itself.”

Few decision-making processes generate as high a proportion of unfavorable outcomes as those of the prison grievance system. As we mentioned above and discussed in more detail in previous work (Calavita and Jenness 2015), only 0.2 percent of prisoner grievances were granted in California at the final level of review in 2005–2006. At the first and second levels, the estimated chances of success were somewhat better—at 2.3 percent and 1.9 percent, respectively. However fair or unfair its procedures are—and there would appear to be a built-in bias in this process in which the CDCR is simultaneously defendant, judge, and jury—the consistently high denial rate arguably dominates these prisoners’ belief that the process is unfair.

In addition to consistently unfavorable outcomes in the prison context, the stakes for these men are often extremely high. A few previous studies address the degree to which the risks associated with a decision affect whether perceptions of procedural justice have an independent impact on people's satisfaction with that decision. As discussed above, Heinz (1985: 15) interrogated “the empirical reliability of the distinction between the concepts of procedural and distributive justice.” Examining data from hundreds of interviews with police officers, victims, and defendants whose felony cases were part of a field experiment that focused on plea bargaining reform, Heinz found that for the officers and the victims, perceptions of process and outcome were distinct. But, for defendants, “their ratings of the ways in which their case was reviewed by the judge and their ratings of the satisfaction with the disposition formed a single factor, explaining 62 percent of the variance. The most important variables in the factor were the outcome variables” (Heinz 1985: 22). In other words, Heinz (1985: 22) found that whether people evaluated process independently from outcome varied by how high the stakes were for them, and that “the defendants—who had most at risk—combined procedural and outcome perceptions.” A few scholars have countered with findings that evaluations of process are important independent of outcome even for those facing high stakes (e.g., Casper et al. 1988; Tyler 1990).

The stakes are particularly high in the prison context, and this may make all the difference. Prison is a total institution (Goffman 1961) where all of one's daily activities and routines are overseen by authorities, where liberty and agency are stripped, and where the state becomes inmates’ “landlord, employer, tailor, neighbor, and banker” (Preiser v. Rodriguez 1973). In one of the most famous accounts of prison life, Dostoevsky (1861−2/1965: 16) declared that prison “sucks the living sap out of a man.” Bonta and Gendreau (1990: 347) summarize contemporary descriptions of prisons as “barren landscapes devoid of even the most basic elements of humanity.”

In this environment, all kinds of deprivations are fraught with meaning. Missing family photos, confiscation of magazines deemed contraband, broken radios, opened mail, and cold meals not only diminish prisoners’ few remaining pleasures but also are daily reminders of one's powerlessness and infringe on one's sense of dignity. Prisoner Michael Johnson told us poignantly, “What's small to one man might be great to another. This guy [an officer] goes home every day to his wife, to his mistress, to his boyfriend, to whatever. So, what he might think be small might be major to a guy who's bein’ told when to eat, when to go to sleep, when to boo-boo, when not to boo-boo … .” In this restrictive environment in which deprivation defines daily life, it should not be surprising that prisoners’ satisfaction with the grievance process depends on what, if anything, they are able to eke out of it.

While all of these deprivations carry added weight in the prison context, two types of grievance are particularly high-stakes: medical and disciplinary. The year before we began our work to gain access for this study, U.S. District Court Judge Thelton Henderson put the California prison health care system in receivership, having found that its conditions violated the Eighth Amendment's prohibition of cruel and unusual punishment. Five years later, the U.S. Supreme Court also concluded that mental and physical health care in California prisons violated the Eighth Amendment (Brown v. Plata 2011). A wide range of experts and independent review panels reported that in fiscal year 2003–2004 one prisoner a week died in California due to inadequate care (cited in Brown v. Plata 2011).

The larger study from which we draw this article included a content analysis of a random sample of all the prison grievances that made it to the third level of review (Calavita and Jenness 2015). A few illustrations from that grievance sample underscore the high stakes involved in medical grievances. In one case, a prisoner who had lumps on his shoulder and thumb requested that the lumps be removed. In a harrowing account, the prisoner wrote that he had had surgery in prison and “a large lump formed on my right shoulder after [doctor's name] torn the screw loose that [other doctor's name] had installed to repair the joint he fractured when his digital saw slipped during the surgery.” He said that the doctor told him he would remove the lump on his thumb. “Since then,” he wrote,

I have made repeated requests to have these lumps removed … . It is now over 5 years since I began requesting to have these lumps in my right thumb removed. I can't bend my thumb all the way to grip anything with it. It is very painful. The large lump on my right shoulder … has gotten larger and causes me severe pain.

In another case, a prisoner reported experiencing chronic trouble breathing and pain in his lungs. He wrote in his grievance that when he finally saw a doctor, “He recommended pepto-bismal and said they'd call me back in the morning. They never called me back and I continued to have severe pains in my lung and I was getting a fever almost every day.” Thereafter, another doctor diagnosed him with pneumonia, and this time the appellant wrote:

They gave me anti-biotics for a week and shots, but still I was sick … . Approximately 5 months later and still I have pain in my lungs … . They've taken way too long to follow up on the Valley Fever test that came up positive back in [date] and I feel my lungs is just getting more and more damaged.

Another prisoner wrote in his grievance that he had fallen and suffered severe injuries. In his grievance, he said that two scheduled surgeries had been canceled without explanation. His appeal was granted at the first level, and he eventually received surgery. He subsequently requested follow-up attention in a second-level appeal, saying he was in pain and pleading, “I respectfully ask for help.” He was denied, and after one more round of appeals the official response read, “The appellant has deceased rendering the issues under appeal moot.”

These medical grievances bring home the high stakes, where appellants’ survival, bodily integrity, and/or pain management depend on a favorable outcome. In this context, being treated with respect and impartiality by medical personnel is no doubt important, but receiving sometimes life-or-death medical treatment is understandably the ultimate measure of success.

Moving beyond medical appeals, disciplinary grievances have their own type of high stakes. Receiving a disciplinary infraction puts a prisoner at risk of losing his ticket to freedom—his release date—and throughout our interviews prisoners told us time and time again that they covet their release date, sometimes above all else. Peter Owen, a young White man in a minimum security prison, told us in an interview how critical his release date is and how far prisoners go to avoid jeopardizing their release dates. He said, “Everyone here has a release date, so police officers think that it's okay to use obscene gestures, put us on the spot in front of other inmates, and they can get away with it here because they know most people won't do anything because … they don't want to do extra time.” Tyrone Jones, a Black prisoner in a medium security prison for more than 14 years when we interviewed him, said (emphasis in the original),

COs [correctional officers] get to mess with you, you know what I'm sayin’? They get to pick at you and this is people's lives versus a game to a CO. It has nothin’ to do with their life. But, it has somethin’ to do with our lives. Because they have the power to take away time. And, that's something God gave us, is time.

James Little also described the lengths he goes to in order to avoid anything that would prolong his prison stay. Speaking of guards who “test” inmates and “push” them, he told us,

So you can't let them see you sweat … . You think ‘I'm really trying to go home, but this dude [CO] is just pushing, pushing, pushing,’ and they do that, and they know they can get away with that. And that really tears you up because you get to thinking about, ‘Okay, I want to see my granddaughter who I haven't seen’ … . Oh, man, they can get under your skin.

If we are right about the high stakes of the prison environment leading to a greater emphasis on favorable outcomes, then we can hypothesize differences across grievances according to how high the stakes are. In other words, if it is the unusually high stakes of the prison environment in general that produce a greater emphasis on outcomes than what is usually found in procedural justice studies, we should expect to see denied medical and disciplinary grievances eliciting the most overall dissatisfaction of all types of prisoner grievances, and the most emphasis on favorable outcomes.

To test this hypothesis, we coded the medical and disciplinary grievances reported to us by our interviewees as “highest stakes” and other types of grievances they told us about as “other.” As expected, for highest-stakes cases, the statistically significant correlation between outcome and satisfaction with outcome was stronger than for other cases (0.746, compared to 0.524). In other words, while the association between outcome and satisfaction with outcome is strong and statistically significant for all cases, it is particularly strong for the highest-stakes cases.

If the disproportionately high stakes of these particular grievances enhance the importance of a favorable outcome, it stands to reason that the generally high stakes of prison may explain why prisoners do not evaluate procedural justice and outcomes independently. Add to this context of deprivation and high stakes the fact that outcomes are very rarely favorable to them, and it is easy to see why, as James Little put it, they are dissatisfied “because it doesn't work.”

Another way to capture the processes delineated here is by presenting a series of logistic regression models that speak to the relationships among key variables and the relative impact of each variable on overall satisfaction with the outcome of a grievance (Table 3).⁸ The simple bivariate relationship presented in Model 1 suggests an association between satisfaction with the procedural management of a grievance and satisfaction with the outcome (b = 3.21, p < 0.001). However, Model 2 reveals that the actual outcome of the grievance produces a considerably larger coefficient (b = 4.32, p < 0.001). Finally, Model 3 adds “stakes” into the equation and confirms a statistically significant negative relationship between stakes and overall satisfaction with grievance outcome (b = −3.15, p < 0.001).⁹ In other words, the important relationship between outcome and satisfaction is further magnified when the stakes associated with the grievance are highest.¹⁰ In conjunction with the finding presented earlier that the presence of three external indicia of procedural justice is not statistically related to prisoners’ satisfaction with grievance outcomes, and with our in-depth qualitative data, these findings pose a significant challenge to any theory that overlooks the social context in predicting the power of procedural justice to shape disputants’ satisfaction.

Table 3. Logistic Regression Models Predicting Satisfaction with Grievance Outcome

Note: Because these are correlated data, we obtain robust standard errors.

^** p < 0.01; ^***p < 0.001 (two-tailed).

Discussion

This study elaborates some aspects of the procedural justice scholarship. Drawing on unique quantitative and qualitative data collected inside prison walls, we found that prisoners privilege actual outcomes in gauging their satisfaction, and that the dominance of procedural justice assessments is thus more variable than usually recognized. Consistent with Heinz's (1985) procedural justice study of felony defendants in plea bargaining cases, the prisoners in our study were above all concerned with how their grievance turned out—specifically, whether they were able to extract remedies from this institution that otherwise is responsible for so much daily deprivation. So dominant is this substantive dimension to their satisfaction that procedural dimensions are largely subordinate to it, and indeed are defined by it. That said, procedural justice does matter to these prisoners.

We argue here that the precedence of outcomes in these men's dispute satisfaction is grounded in the high stakes of the prison context. It may also be the result of a grievance system that routinely delivers negative outcomes in a context of extreme institutional asymmetry and lack of trust. Beyond these specific findings, what we see more broadly is the power of institutional context to shape attitudes about fairness and justice. Prison may be a microcosm of the larger society as the New York State Special Commission on Attica (1972: 82) said long ago, but with its intense state control, disempowerment, and heightened stakes, it is an institution that exposes in high relief the importance of structural context in people's perceptions of justice.

To the extent that it is the high stakes of prison that affect these men's privileging of outcomes, substantive justice may also be dominant in other settings with high stakes. Heinz's (1985) study of felony plea bargaining cases found that for defendants (but not for the police or victims) outcomes were so important that the boundaries between substantive justice and procedural justice were blurred. Similarly, in their work on employment discrimination cases, Berrey et al. (2012) report that when people's jobs were on the line, their determination of fairness relied on whether the outcome was favorable to them. It is hoped that future research will continue to pursue the role that high stakes play in people's perceptions of justice in a variety of settings.

The concept of distributive justice—that is, how we perceive we are treated relative to others in similar circumstances—might also be brought into this conversation. The concept of fairness with which we began this article has a long history in philosophy and social science, and the notion of distributive justice is critical to our understanding of fair play. How we are treated in comparison to others is a central human concern and one that might fruitfully be reintegrated into, indeed center staged, in future procedural justice studies.

Future research is also needed on perceptions of justice in a variety of institutional contexts, including importantly women's prisons. For decades, studies have demonstrated the specific characteristics of women's prisons (Kruttschnitt and Gartner 2004; Owen 1998), and in more recent work Jill McCorkel (2013) emphasizes the role of “the gender regime” in the structure and functioning of modern prisons. Given this, it would be useful to examine whether perceptions of justice and an emphasis on outcomes differs in prisons for women.

Across the gender divide, prison is a total institution with a highly asymmetrical power structure and extreme limits on prisoner agency, and is unique in many ways. With more than 2.2 million people incarcerated in prisons and jails in the United States, it is of critical importance to understand how those who inhabit these massive institutions perceive justice and how the environment shapes those perceptions (Kirk 2016). Furthermore, while prison is its own kind of “peculiar institution,” it arguably shares at least some features with other institutional contexts. The military, mental hospitals, and boarding schools—while no doubt differing among themselves and with the prison environment in terms of how much autonomy is allowed and how much distrust there is—are also total institutions with hierarchical command structures and limited autonomy. It is hoped that future research will explore the ways that institutional contexts such as these affect perceptions of justice, just as social context affects so much else of social life.