239
Settling for Less? Organizational Determinants of Discrimination-Charge Outcomes
C. Elizabeth Hirsh
Although more than 60,000 workers formally charge their employers with unlawful sex or race employment discrimination annually, fewer than one in five charges results in outcomes favorable to the complainant. Building on sociolegal and organizational theory, this study examines how employing organizations avoid unfavorable discrimination-charge outcomes. Using EEO-1 establishment reports matched to discrimination charge data provided by the Equal Employment Opportunity Commission, I assess the effect of employers’ legal experience, resources, and indicators of legal compliance on the likelihood that complainants receive favorable charge outcomes, benefits, monetary settlements, and policy change mandates. In general, I find that legal experience, establishment size, and indicators of legal compliance insulate employers from unfavorable charge outcomes. However, in situations where employers are willing to settle claims, legally experienced establishments are more likely to pay monetary damages and receive mandates to change their workplace policies.
I
n August 2004, the U.S. Equal Employment Opportunity Commission (EEOC) announced the resolution of a class-wide discrimination lawsuit against Home Depot in which several workers alleged unlawful treatment on the basis of sex, race, and national origin (U.S. Equal Employment Opportunity Commission 2005). The resolution mandated that Home Depot should pay $5.5 million to current and former workers as well as appoint an equal employment opportunity (EEO) coordinator, provide antidiscrimination law training to managers, and remain under EEOC monitoring for more than two years. While this resolution afforded legal redress for those subjected to sex and race discrimination at
Please direct correspondence to Elizabeth Hirsh, Cornell University, 323 Uris Hall, Ithaca, NY 14853; e-mail:
[email protected]. I thank Ron Edwards and Bliss Cartwright at the Equal Employment Opportunity Commission for providing access to their data. I appreciate the helpful comments of Barbara Reskin, Lowell Hargens, Paul Burstein, Becky Pettit, LSR reviewers, and LSR editor Carroll Seron on earlier drafts of this article. A previous version of this article was presented at the annual meetings of the American Sociological Association, Montreal, Canada, August 2006. Research support comes from the National Science Foundation (grant no. SES-0602496).
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
240
Determinants of Discrimination-Charge Outcomes
Home Depot, such far-reaching settlements are the exception rather than the norm for discrimination disputes. Of the 60,000 sex and race discrimination complaints filed annually with the EEOCFthe federal agency that Congress created to receive, investigate, and resolve employment discrimination claimsFfewer than one in five claims results in outcomes favorable to the complainant (U.S. Equal Employment Opportunity Commission 2004). Thus although EEO law enforcement provides redress for a few, most workers who file formal complaints of employment discrimination receive no remedy. This mixed scorecard for antidiscrimination enforcement embodies a long-standing question for law and society scholarship regarding the capacity of the law to provide redress for its intended beneficiaries. Federal EEO laws give workers who perceive unlawful employment discrimination the right to file formal charges of discrimination with the federal EEOC or a local fair employment agency; indeed all workers must file with the EEOC or a local agency before initiating a private lawsuit. However, as EEOC enforcement statistics reveal, employers prevail in the vast majority of discrimination claims. A host of factors are potentially responsible for this employer advantage: workers may bring erroneous complaints, employers may be especially adept at defending claims, or the increasingly subtle nature of discrimination in the post–civil rights era may complicate EEOC investigations. In an effort to understand how workers, employers, and regulatory agents negotiate discrimination disputes, this article explores the conditions that produce (or fail to produce) favorable outcomes for workers who file charges of sex or race discrimination under federal EEO laws. More specifically, I consider how defending organizations influence the charge resolution process to their advantage. A long tradition in law and society research recognizes that legal resolutions do not unfold in a vacuum; rather, structural features of the law and regulated actors shape the legal process and its outcomes. For instance, in his classic article ‘‘Why the ‘Haves’ Come Out Ahead: Speculation on the Limits of Legal Change,’’ Galanter (1974) argues that because legal disputes often occur between parties with unequal resources and experience, the more powerful players can strategically manage their engagement with the legal system to maximize their long-term interests and minimize losses. In addition, new institutional theory in sociology draws attention to the myriad ways in which organizations respond to the law and regulatory efforts. For example, a convincing body of research demonstrates that organizations responded to civil rights law by adopting employment practicesFsuch as EEO offices, affirmative action plans, and due process proceduresFto demonstrate
Hirsh
241
their commitment to the law and the normative ideals that it embodies (Edelman 1990; Dobbin et al. 1993; Sutton & Dobbin 1996). More recently, institutional scholars have noted the capacity of such organizational structures to inform common understandings of what it means to comply with the law (Edelman 2005; Edelman et al. 1999; Nelson & Bridges 1999), such that organizational practices and routines geared toward managing diversity become tantamount to EEO legal compliance. Building on both traditions, this article examines how employers shape discrimination-charge outcomes and the type of benefits that are awarded through charge settlementsFincluding monetary benefits, nonmonetary benefits and mandated workplace policy changes. Drawing on law and society research, I examine the ways in which legal experience, resources, and know-how enable employers to strategically manage their engagement with the charge resolution process. From institutional theory, I consider how organizations emphasize EEO compliance strategies to demonstrate commitment to fair employment practices and antidiscrimination law. My central argument is that legal experience, resources, and compliance strategies enable employers to structure their engagement with the charge resolution process so as to minimize unfavorable charge outcomes. The data for this project come from a national random sample of private work establishments that I matched to discrimination charge data obtained from the EEOC. In addition, I spent several months observing charge processing at an EEOC district office. I incorporate insights from this experience to enrich my account of the EEOC’s charge resolution process and how employers respond to allegations of discrimination. I begin with a discussion of the legal basis for bringing charges of discrimination and the EEOC charge resolution process. Next, I discuss the role organizations play in shaping discriminationcharge outcomes. Here I elucidate theoretical expectations regarding the relationship between legal experience, organizational resources, EEO compliance, and discrimination-charge outcomes. Finally, I present a statistical analysis of the effect of organizational characteristics on the likelihood that complainants receive favorable charge outcomes and various types of reliefFincluding monetary benefits, nonmonetary benefits, and workplace policy changesFamong a sample of sex and race discrimination charges filed with the EEOC between 1991 and 2002. Although the EEO legal framework covers discrimination on the basis of several protected classes, including sex, race, color, national origin, religion, age, and disability status, my analysis focuses only on charges citing race, color, national origin, or sex discrimination.
242
Determinants of Discrimination-Charge Outcomes
EEO Law and Charge Resolution The most comprehensive EEO legislation is Title VII of the Civil Rights Act of 1964. Title VII bans the use of race, color, national origin, sex, and religion in employment decisions among work establishments with at least 15 employees. Originally, the law was intended to protect workers from race discrimination; however, during congressional debates, opponents of the law added sex as a way to thwart its passage, assuming that if legal protection was extended to white women, the legislation would fail (Graham 1990:134–9). To the surprise of opponents, the law passed even with the inclusion of sex and provides legal recourse for those discriminated against on the basis of sex, race, color, national origin, and religion. In 1978, Congress passed the Pregnancy Discrimination Act, which amended Title VII to include pregnancy discrimination as a form of unlawful sex discrimination. Title VII also created the EEOC to receive, investigate, and conciliate workers’ claims of discrimination. In filing claims with the EEOC, workers can cite discrimination involving a variety of employment issues. First, complainants can bring charges citing the use of sex, race, color, and national origin in the allocation of workplace rewards. Such charges may involve allegations of discrimination in hiring, promotion, termination, pay, or disputes concerning the terms and conditions of employment such as benefits or working hours. Second, workers can bring harassment claims. Harassment involves unwelcome conduct that is sufficiently enduring or severe to render the workplace a hostile environment. Although sexual harassment has been the subject of considerable public attention in recent years, complainants can cite harassment on the basis of any protected status (i.e., race, national origin, age, religion). Finally, complainants can bring charges of retaliation. Retaliation complaints do not involve claims of direct discrimination; rather, complainants allege retaliation if they suspect that their employers retaliated against them after they complained about discrimination against themselves or another worker. While Title VII of the Civil Rights Act of 1964 is the foundation of antidiscrimination law and generates the majority of discrimination claims, the Equal Pay Act (EPA) of 1963 provides additional protection from pay discrimination for women. Under the EPA, women can bring charges against their employer alleging that their pay is not comparable to that of men working in the same capacity. Although women alleging pay discrimination have the option to file under either Title VII or the EPA, most complaints of pay discrimination are processed under Title VII. Because the EPA covers workplaces with at least two employees while the size criterion for Title VII is 15 employees, most EPA claims involve gender pay
Hirsh
243
disparities that occur in very small work establishments that fall under Title VII’s statutory minimum. In addition to filing charges with the EEOC, workers can also bring claims of discrimination to local Federal Employment Protection Agencies (FEPAs), such as local civil rights commissions. However, all complaints that allege violation of federal law are cofiled with the EEOC, and FEPAs often transfer cases to the EEOC for processing. Among the EEOC charges that I analyze, roughly one-fifth were transferred from FEPAs. Title VII also provides access to the federal courts but requires that workers file with the EEOC or a FEPA to receive a ‘‘right-to-sue’’ letter before proceeding to court. Thus although the EEOC constitutes one step in a larger regulatory framework that involves local agencies as well as the federal courts, nearly all employment discrimination claims interact with the EEOC administrative process in some way. Figure 1 provides a schematic of the EEOC’s charge resolution process. When the EEOC receives a complaint of discrimination, the first step is an intake interview with the complainant. The purpose of this interview is twofold: to determine if the grievance falls under federal EEO law, and to assess the prima facie merit of the claim. If the claim is within federal jurisdiction, the EEOC intake officer writes a formal charge of discrimination and serves it on the accused employer.1 Beginning in the late 1990s, the EEOC started a ‘‘Charge Handling Priority System’’ to prioritize cases at intake according to the presumed merit of the case. Under this system, roughly 19 percent of cases are classified ‘‘A,’’ indicating that there is strong prima facie evidence to suggest that unlawful discrimination has occurred; 58 percent of cases are categorized ‘‘B,’’ indicating that prima facie evidence of discrimination is moderate; and the remaining 23 percent of cases are classified ‘‘C,’’ indicating that the prima facie case for discrimination is weak (unpublished analysis of the EEOC charge data; available upon request). In the late 1990s, the EEOC also implemented an Alternative Dispute Resolution (ADR) program that offers some complainants and employers the option to informally mediate the charge before a formal investigation takes place. These mediations are distinct from formal settlements obtained by the EEOC in that a professional mediator rather than an EEOC investigator oversees the resolution. In addition, details of the resolution are not disclosed to EEOC personnel or to the public. Typically, the EEOC will extend 1 The EEOC will write a formal charge for any claim that falls under federal EEO law, even if the initial evidence is weak. Most claims that the EEOC turns away are due to jurisdictional issues. Unfortunately, EEOC data do not allow an accurate count of how many inquiries do not result in formal charges.
244
Determinants of Discrimination-Charge Outcomes
CHARGE FILED
Intake interview and charge classified: A, B, or C MEDIATION (ADR)
Unsuccessful mediation
INVESTIGATION
Settle pre-cause determination*
Successful mediation*
DETERMINATION
Reasonable cause found
No reasonable cause Dismissed: issue right to sue
CONCILIATION
Conciliation fails
Conciliation succeeds*
LITIGATION
Does not recommend litigation Settle pretrial* Dismissed: issue right to sue
Complainant loses or case dismissed
Go to trial
Complainant wins*
Figure 1. The Charge Resolution Process. Note: nOutcome favorable to complainant.
the ADR option to cases with moderate prima facie evidence of discrimination (i.e., those classified as ‘‘B’’ cases). If the parties come to an agreeable resolution in ADR, there is no further investigation and the case is closed. If the dispute is not resolved through ADR, the case is returned to EEOC personnel for investigationFalong with all other cases not deemed appropriate for ADR. During an investigation, EEOC investigators solicit information from both the complainant and the employer. Such information includes a position statement from the employer, which allows
Hirsh
245
employers the opportunity to describe their side of the story. In addition, investigators typically give the employer a ‘‘request for information’’ in which they ask the employer to provide information related to the claim, such as personnel records for the complainant as well as similarly situated employees, pay systems, statements from witnesses, and other relevant evidence. Investigators may also conduct phone interviews with managers, personnel officers, and other witnesses, and in some cases, they may conduct on-site visits to gather relevant evidence. The intensity of investigations varies considerably across cases. At one end of the continuum are complainants who approach the EEOC to receive a default right-to-sue letter, rather than a formal investigation, so that they can proceed to court on their own. According to EEOC policy, complainants can request an immediate right-to-sue letter if they are facing a statute of limitations requiring that they act quickly in court. Under such conditions, the EEOC will waive the investigation and issue a right-to-sue, certifying that the case is sufficiently complex such that the Commission cannot conclude the investigation before the statute of limitations expires. While there are no data on the proportion of claims that result in default right-to-sue letters without an investigation, one lead EEOC investigator described these as infrequent and emphasized that even among those seeking a right-to-sue, most complainants and their attorneys want to know the employer’s position on the claim and will hold out for the EEOC investigative report before proceeding on their own.2 At the other end of the continuum are cases that receive detailed investigations, which might entail phone interviews and on-site visits to solicit statements from coworkers, supervisors, and company leadership. The processing category assigned at intake provides some indication of the depth of investigation. For instance, cases that are categorized as strong cases receive thorough investigations, while the weaker cases receive considerably less attention. Indeed the categorization system was introduced to enable investigators to concentrate their efforts on cases with stronger initial evidence of discrimination. Thus it is plausible to assume that roughly 20 percent of cases are identified as having strong prima facie evidence of discrimination (‘‘A’’ cases) and receive thorough investigations, roughly 60 percent (‘‘B’’ cases) receive moderate investigative attention, and the remaining 20 percent (‘‘C’’ cases) may generate little more than a position statement from the employer. Based on the evidence obtained through investigation, the EEOC then determines the merit of the claim. If investigators 2 This information is based on personal correspondence with a lead investigator at an EEOC district office.
246
Determinants of Discrimination-Charge Outcomes
determine that there is not reasonable cause to believe that unlawful discrimination did occur, the case is dismissed and the complainant then receives the right-to-sue letter. If investigators determine that there is reasonable cause to believe that the employer violated antidiscrimination law, investigators move to conciliate the case. According to EEOC aggregate charge data, investigators find reasonable cause of discrimination in only 5 to 10 percent of all cases.3 In particularly egregious instances of discrimination, a case may be moved directly to litigation after the determination of reasonable cause. But typically, conciliation follows a determination of reasonable cause. About one-third of cases that receive reasonable cause determinations are successfully conciliated, with both parties agreeing to a settlement. In the remaining two-thirds of cases, conciliation fails and the investigator must decide to either close the case and issue a right-to-sue letter or recommend litigation and transfer the case to the EEOC legal unit. Less than 1 percent of all claims filed with the EEOC reach the legal unit. Among those that do, the decision to litigate is at the discretion of EEOC attorneys. If the legal unit decides to litigate, the EEOC becomes the plaintiff and takes control of the case. Attorneys on both sides begin the discovery of information process. At the end of discovery, either party can initiate a request for summary judgment, which enables the courts to resolve cases where the facts indicate that one party is entitled to judgment. This is more commonly used as a defense tactic by employers. If an employer brings a summary judgment request, the judge examines the strength of the EEOC’s case and determines whether the facts of the case warrant a trial. If the judge grants an employer’s request for summary judgment, the ruling indicates that the EEOC does not have sufficient evidence to continue with the lawsuit and the case is dismissedFan agreeable outcome for the employer. If a defendant’s request for summary judgment is denied, the case proceeds to court. Given that the EEOC litigates few cases, federal regulation of race and sex discrimination is best characterized as an administrative rather than a formal legal process. Formal legal actors, such as attorneys and the courts, are involved in a minority of cases; these include the few hundred that the EEOC litigates each year as well as an estimated 10,000 sex or race discrimination cases that individuals may litigate privately after exhausting the EEOC process.4 3 While cases categorized as having strong prima facie evidence at intake (i.e., ‘‘A’’ cases) are more likely to receive reasonable-cause determinations, reasonable-cause decisions do not necessarily correspond to intake categorizations. 4 While specific data on the number of sex and race discrimination lawsuits filed annually by private parties in federal district court are unavailable, data from the Administrative Office of the U.S. Courts suggest that in the late 1990s and early 2000s, roughly
Hirsh
247
The remaining 50,000 annual claims are resolved in the context of the EEOC administrative process. It is important to recognize that a case can settle at any point during this process if both parties agree to the terms of a settlement. Although the EEOC manages the resolution process and defines specific criteria necessary for settlement, the agency does not unilaterally issue resolutions and does not represent the interests of the complainant.5 Rather, in the majority of cases, the actors involved in the dispute determine when and how resolutions occur. In some circumstances, parties reach a settlement during the initial investigation, while in other cases, parties settle after the EEOC makes a formal decision regarding the merit of the case. As this discussion demonstrates, charge resolution is a complex process in which social actors encounter a formal administrative system but remain very much involved in the evolution of the outcome. Because Title VII did not explicitly define what constitutes discrimination (see Blumrosen 1993; Graham 1990), the law itself provides little guidance for EEOC personnel to define and identify discrimination. As a result, EEOC investigators are left with the onerous task of distinguishing lawful from unlawful behavior. While this is a difficulty faced by all regulatory agents, for EEO law enforcement, the situation is complicated by the fact that most discrimination in the post–civil rights era is subtle, concealed in structures and practices that are not blatantly discriminatory (Bisom-Rapp 1999; Sturm 2001). This diffuse nature of contemporary discrimination, coupled with the ambiguity of EEO law generally, forces investigators to make decisions regarding the merit of cases without clear legal instruction or empirical evidence. Under these conditions, the parties involved in the case have considerable opportunity to manipulate the administrative process and its outcomes. In addition, because the parties involved can settle the claim at any time, charge outcomes are not isolated events at which point the complainant either prevails or loses. Outcomes emerge out of bargaining, negotiation, and compromiseFblurring the lines between winning and losing. For instance, a complainant may receive some compensation but never get a definitive statement (i.e., a reasonable-cause finding) that the alleged discrimination occurred. Alternatively, the complainant may receive a favorable decision with respect to the merit of the claim but walk away with no monetary settlement. For example, in a racial harassment lawsuit filed 20,000 employment discrimination cases were filed in district court annually (Nielson & Nelson 2005). Assuming that private lawsuits are similar to EEOC charges in terms of bases of discrimination, about 10,000 lawsuits would involve race or sex discrimination. 5
The only exception is if the case is litigated.
248
Determinants of Discrimination-Charge Outcomes
against North Carolina’s transportation department in 2005, while a jury ruled that workplace conditions did constitute a racially hostile environment, they did not award the plaintiffs the monetary damages they had sought. Because the jury validated their claims, plaintiffs considered this decision a victory. However, the employer could also view the decision favorably, as the employing organization escaped a potentially costly payout (Lillard 2005). As this example illustrates, winning and losing can take on different meanings depending on the actors’ immediate and long-term incentives for engaging with the law (Burstein & Monaghan 1986; Galanter 1974).
The Organizational Construction of Charge Resolution Taken together, the ambiguous nature of EEO law and the EEOC administrative process as well as the emergent nature of charge resolutions provide considerable latitude for the actors involved with charge resolutions to strategically structure their interaction with the law. As Edelman et al. note, ‘‘[t]he more ambiguous and politically contested the law, the more open it is to social construction’’ (1999:407). Thus the structure of the charge resolution process allows the actors involved to participate in how charges are resolved and with what outcomes. In the discussion that follows, I suggest three mechanisms by which employing organizations can influence the charge resolution process and its outcomes. Legal Experience
First, employing organizations bring prior legal experience to the charge resolution process and can exploit this experience in dealing with discrimination disputes. As Galanter (1974) observes, actors who repeatedly interact with the lawFrepeat playersF enjoy the advantages afforded by legal experience in comparison to actors who utilize the law on rare occasionsFthe one-shotters. Formally, the legal system is neutral with respect to these types of parties, but informally repeat players enjoy certain advantages. Repeat players are larger units with greater experience and resources. Having done it before, repeat players enjoy specialized knowledge and institutional memory, which allows them to strategically interact with the legal process. On the other hand, oneshotters are smaller, inexperienced actors, usually individuals for whom the stakes of any single case are high. An extensive literature affirms Galanter’s thesis. For example, Wheeler and his colleagues (1987:428) found employers to have an edge in disputes with employees in a sample of state Supreme Court cases; Albiston (2003)
Hirsh
249
uncovers a similar employer advantage among published cases involving the Family Medical Leave Act. In the context of discrimination-charge adjudication, employers are much more likely than the average employee to have prior experience with the legal system in general and the charge resolution process in particular. Drawing on this experience, employers can tailor their response to claims depending on the nature of the dispute, the strength of available evidence, and how the adjudication process unfolds. This may entail settling particularly strong cases early in the administrative process to avoid a disruptive investigation or subsequent legal battle. An EEOC settlement is often preferable to fighting an ongoing dispute that may consume organizational resources and tarnish the company’s public record. Agreeing to settle may also give employers more leverage in determining the terms of the settlement. For instance, complainants and EEOC officials may view an organization that is willing to settle as cooperative and thus soften their demands in negotiations. Alternatively, legal experience may afford employers the prudence to stall the administrative process to their advantage, especially in situations where the evidence is inconclusive. By delaying the production of relevant documents and delaying discussions of settlement, employers can exhaust complainants, persuading them to settle for less. Complainants, on the other hand, typify the one-shotters. For most complainants, the filing of a discrimination claim is a rare and reluctant encounter with the legal process. As studies of claiming behavior suggest, aggrieved workers hesitate to raise claims because doing so reifies their status as victims, jeopardizes their jobs, and in the end is not worth the legal battle (Albiston 2005; Bumiller 1988; Marshall 2005). When workers do take legal action, to the extent that they endure the psychic costs of initiating a dispute and mobilizing the regulatory apparatus, the stakes are high. Organizational Resources
In addition to the legal experience edge, employers enjoy greater resources and exclusive access to key evidence relevant to workplace discrimination claims, as compared to the typical worker. While discrimination charges are filed against individual work establishments, many establishments are subsidiaries of larger firms and thus can draw on the resources of their parent organization when faced with a charge. In addition, as Bisom-Rapp (1999:990) notes, all documents relevant to liability are the property of the establishment. Aside from an initial statement from the complainant, the EEOC obtains most evidence related to the claim from the employer. This includes personnel files, pay records, and
250
Determinants of Discrimination-Charge Outcomes
statements from supervisors and other employees who have information pertinent to the claim. Establishments familiar with the charge resolution process can strategically manage how they disclose such evidence to EEOC investigators. Indeed, in recent years organizations have increasingly established internal legal or human resource units whose primary responsibilities include dealing with employee disputes (see Edelman & Suchman 1999; Heinz et al. 1998; Westin & Feliu 1988). The presence of such specialized legal units and actors allows organizations to draw on pre-established routines to manage the uncertainty associated with accusations of discrimination and ensures calculated responses to claims. Demonstrating Compliance
Finally, organizations may shape individual charge resolutions by demonstrating compliance with EEO law more generally. First, employers may demonstrate compliance with EEO law by drawing attention to workplace structures and procedures that are consistent with fair employment practice. An extensive literature documents the extent to which work organizations have adopted EEO policies and practices, such as formal grievance procedures, an EEO office, affirmative action plans, and internal labor markets in the wake of antidiscrimination legislation (Edelman 1990, 1992; Dobbin et al. 1993; Sutton & Dobbin 1996). In resolving discrimination charges, employers may emphasize these fair employment structures to underscore the fact that the company takes EEO legislation seriously and to signal to EEOC administrators as well as the aggrieved worker that the issue raised in the claim is an aberration rather than a systematic workplace issue. For instance, as Schultz (1990) reports, in Title VII cases challenging sex segregation, employers referenced affirmative action plans, training programs, and sex-specific recruitment efforts as evidence of their efforts to comply with EEO law. In theory, the presence of EEO-related policies should minimize discrimination in the workplace; however, research evidence is mixed regarding whether and which policies do. For instance, Kalev et al. (2006:603–4) find that the presence of affirmative action plans, diversity committees, and diversity staff are related to increases in the representation of women and blacks among managers, yet diversity training programs and diversity evaluations are largely ineffective for increasing managerial diversity. Edelman and Petterson (1999:121) conclude that EEO structures, including affirmative action plans, have had little impact on substantive gains for women and minorities, and Hirsh and Kornrich (2008) report that establishments subject to affirmative action requirements are just as likely to receive discrimination charges as establishments
Hirsh
251
without such requirements. Thus although EEO structures do not guarantee sex- and race-neutral practices, to the extent that such structures are consistent with the normative aim of antidiscrimination law, they may provide leverage in resolving discrimination claims by signaling an organizational commitment to equal opportunity. Second, employers may demonstrate legal compliance by making reference to empirical indicators of gender and race equality, such as the presence of women or minorities in the workplace or among high-level positions. For example, in a charge alleging national origin discrimination that I investigated while observing charge processing at a district EEOC office, an employer provided a list of all employees’ national origins and job titles along with a letter emphasizing the extent to which persons of diverse national origins were represented throughout the occupational hierarchy. Indeed, the EEOC takes such information into consideration when determining the merit of workers’ claims, as occupational integration, especially across hierarchical positions, is suggestive of sexand race-neutral employment practices. Such evidence of a diverse workforce may cast doubt on complainants’ claims of discrimination and employers’ discriminatory intent. In short, emphasizing complianceFby citing the presence of employment structures geared toward managing diversity as well as demonstrating evidence of a diverse workforceFcan provide a powerful means by which organizations mediate the charge resolution process and minimize the penalties associated with noncompliance.
Hypotheses While organizations will generally have more legal experience and resources than the typical worker, not all work organizations are equally equipped to manage the charge resolution process. Establishments will vary in their experience with the law and the EEO administrative process. For instance, as I observed at the EEOC district office, employers’ formal responses to discrimination charges can vary from a handwritten note provided by a small employer to professionally prepared legal documents submitted by in-house counsel. Such variation in employers’ responses will in turn produce variation in the extent to which employers can leverage legal experience, organizational resources, and EEO compliance indicators to resolve discrimination charges with minimal costs. This discussion suggests four hypotheses regarding the influence of organizational characteristics on charge outcomes and settlements.
252
Determinants of Discrimination-Charge Outcomes
Hypothesis 1: Establishments with previous experience with the charge resolution process should be less likely to experience unfavorable charge outcomes. Among cases that settle, establishments with prior experience with charge resolutions should be better equipped to manage the strategic nature of negotiation and thus less likely to incur extensive damages. Hypothesis 2: Drawing on their resources, larger establishments and those that are subsidiaries of larger organizations should be less likely to experience complainant-favorable charge outcomes and, among cases that settle, extensive damages. Hypothesis 3: Establishments that are subject to affirmative action requirements should be more effective in demonstrating compliance with EEO law and thus less likely to experience unfavorable charge outcomes and, among cases that settle, extensive damages. Hypothesis 4: Establishments that are sex- and race-integrated should be more effective in demonstrating compliance with EEO law and thus less likely to experience unfavorable charge outcomes and, among charges that settle, extensive damages.
Administrative Factors The accused organization is not the only social institution that can affect the interpretation of EEO violations and the enforcement of law. The regulatory agency responsible for administering the law, in this case the EEOC, can also affect the application of the law and its outcomes. A recent study by Lancaster et al. (2006) thoroughly documents how administrative properties of the EEOC affect charge outcomes net of the legal characteristics of the claim. While the primary purpose of the current study is to assess the influence of employing organizations on charge outcomes, following Lancaster and colleagues, I take into account several administrative factors that may influence charge resolution. First, different kinds of violations may be treated differently during the investigative and administrative process. The EEOC’s charge summary statistics indicate that sex charges are more likely to result in complainant-favorable outcomes as compared to charges involving race discrimination (U.S. Equal Employment Opportunity Commission 2004). Thus I expect charges alleging sex discrimination to obtain more favorable outcomes as compared to those citing race or national origin discrimination. In addition, the employment issue raised in the claim may affect the available evidence, the parties’ stake in resolving the claim, and the final outcome. For instance, cases alleging discrimination in hiring may be more difficult for complainants to win because job applicants have little access to information regarding the workplace and its personnel practices as compared to cases involving promotion or
Hirsh
253
termination, where employees presumably have some knowledge of their current employer’s practices. Second, the scope of the potential infraction may affect its outcome. While all claims filed with the EEOC pertain to a single worker, occasionally the EEOC will group individual claims into class action cases. This grouping may occur at the point of filing if several workers approach the EEOC together, or the EEOC may later designate a group of individually filed claims as a class after identifying a systematic grievance against a single employer. Either way, to the extent that class charges indicate that the alleged pattern of discrimination is systematic and not the result of an isolated misunderstanding, charges involved in class action cases may be more likely to result in favorable outcomes for complainants. Indeed, in analyzing the outcome of sex discrimination lawsuits, Burstein (1989:658) found that plaintiffs who were a part of larger class action suits were more likely to win than individual plaintiffs. Thus to the extent that class filings lend evidentiary support to complainants, class action claims should be more likely to produce complainant-favorable outcomes. Third, the application of the law may vary across temporal and geographical regulatory environments. The EEOC’s annual budget determines case loads and the amount of time investigators can dedicate to each case. Moreover, variation in administrative agendas and investigative techniques across EEOC districts may produce variation in charge outcomes. Indeed, the Lancaster et al. study (2006) found that the state in which the charge was filed accounted for considerable variation in charge outcomes. Given these potential effects of administrative context, I include controls for the basis of the charge (sex or race), the complainant’s race, the employment issue involved in the claim (i.e., hiring, promotion, termination, pay, harassment, or other issues), class action charges, the EEOC’s annual budget, and the EEOC office where the charge was filed in all analyses of charge outcomes.
Summary In sum, building on law-and-society as well as institutional perspectives, I suggest three ways by which organizations can shape the charge resolution. First, organizations can draw on their legal experience to negotiate charge outcomes that are consistent with their long-term interests. Second, organizations can shape the charge resolution process by leveraging organizational resources and controlling evidence relevant to the claim. Third, organizations can emphasize indicators of EEO compliance to cast doubt on complainants’ allegations of EEO violations. Next, I test these
254
Determinants of Discrimination-Charge Outcomes
hypotheses by analyzing the outcomes of roughly 11,000 race and sex discrimination charges filed against mid- to large-sized work establishments from 1991 to 2002.
Data and Methods To monitor organizations’ compliance with federal antidiscrimination laws, the EEOC requires firms to submit EEO-1 forms documenting the gender and racial composition of their workforce. All private work establishments with at least 100 employees and all federal contractors with at least 50 employees are required to file annual EEO-1 reports. These reports cover more than 40 percent of persons employed in the private sector nationally (Robinson et al. 2005:16). The EEOC also collects data describing the basis and outcome of each discrimination charge it receives as well as charges filed with FEPAs.6 These data include characteristics of the charge itself and all actions taken by the EEOC while resolving the charge, including the final resolution and any benefits awarded. To construct a dataset for this analysis, I drew a national random sample of 2,166 work establishments, with probability for inclusion proportional to establishment size, from the EEO-1 national database for 2002. This constitutes a roughly 1 percent sample of all work establishments that filed EEO-1 reports in 2002. I then extracted retrospective EEO-1 reports dating back to 1990 for each establishment included in the 2002 sample. Finally, I matched the EEO-1 sample to all charges of alleged sex or race/ national origin discrimination filed against each establishment from 1990 to 2002. During this period, workers filed 11,471 charges of sex or race/ national origin employment discrimination against the 2,166 establishments included in my sample. Of the 11,471 charges, I dropped 167 charges that were not resolved by 2002, and 46 charges that were missing data on key variables.7 In addition, since I am interested in how organizations’ prior experience with the charge resolution process affects charge outcomes, I did not analyze charges filed in 1990 because I lacked information on charge filings prior to 1990. Thus I dropped the 448 charge filings from 1990. Analyses are based on the 10,810 charges filed against 6 Every charge that alleges violation of federal EEO law is sourced in the EEOC’s charge database regardless of whether the charge was received and processed by the EEOC or a FEPA. 7 The majority of claims that were not resolved by 2002 were complaints filed in 2001 or 2002.
Hirsh
255
establishments included in my sample from 1991 to 2002 for which there are complete data. Dependent Variables
Because charge resolutions can take many forms, I employ four dependent variables to capture various types of charge outcomes and settlements. First, I investigate the determinants of charges that result in any outcome favorable to the complainant. Given the complex nature of the charge resolution process and the myriad ways in which resolutions can occur, delineating complainantfavorable and unfavorable outcomes is difficult. I identify charge outcomes as favorable to complainants if they result in benefits for the complainant and/or successful conciliations.8 Specifically, I employ a binary variable coded 1 if (1) the case is settled with benefits Fincluding monetary benefits or a positive change in employment conditions, such as reinstatement, promotion, or workplace policy changes; (2) the case is withdrawn by the complainant after benefits are awarded;9 or (3) the case is successfully conciliated. I code this variable 0 otherwise. In addition to the final outcome of the case, I am also interested in the type of relief complainants obtain through settlements. To identify the conditions under which complainants receive benefits, my second dependent variable is a binary measure of whether a resolution results in any nonmonetary benefit for the complainant. Nonmonetary benefits include reinstatement, a promotion, a neutral reference, or a favorable change in employment terms, such as a desirable work schedule. Third, to isolate the financial penalties associated with EEO violations, my third dependent variable is a measure of whether a charge resolution provides any monetary relief to the complainant. Again, I use a binary variable to identify cases that involve either direct (i.e., cash) or indirect (i.e., back pay) monetary benefits for the complainant. Finally, some charge resolutions benefit not only the aggrieved party but the entire workforce by mandating that the accused employer change workplace policies in the direction sought by EEO law. For 8
My coding of complainant-favorable outcomes is consistent with the EEOC’s definition of outcomes favorable to the complainant, with one important exception. The EEOC considers unsuccessfully conciliated charges as favorable to the complainant because all conciliations entail a meritorious finding with regard to complainants’ allegations (U.S. Equal Employment Opportunity Commission 2007). However because unsuccessful conciliations imply that the complainant did not receive an agreeable resolution, I do not code these as favorable to the complainant. 9 This may occur when an employer offers the complainant benefits shortly after receiving the charge and the complainant then withdraws the charge. For instance, if a complainant files a charge of discrimination in promotion and the employer then promotes the worker, the complainant may withdraw the charge.
256
Determinants of Discrimination-Charge Outcomes
Table 1. Descriptive Statistics Measure Dependent Variables Complainant-favorable outcome (among all cases, N 5 10,810) Nonmonetary benefits (among cases settled, N 5 1,535) Monetary benefits (among cases settled, N 5 1,535) Policy change (among cases settled, N 5 1,535) Independent Variables Number prior charges Establishment size (log) Parent size (log) Federal contractor Proportion female employees Sex segregation, D Proportion minority employees Race segregation, D Hire Promote Terms and conditions Pay Harassment Other issue Terminate (referent) Sex charge Race charge (referent) Black Asian Other nonwhite White (referent) Class action EEOC budget EEOC district officen Years observed since 1990 N (total cases)
Mean 0.14 0.45 0.66 0.07
Std. Dev. 0.35 0.50 0.47 0.26
29.69 58.52 7.44 1.34 9.88 1.80 0.61 0.49 0.45 0.24 0.33 0.16 0.27 0.19 0.26 0.13 0.06 0.24 0.18 0.38 0.02 0.13 0.07 0.26 0.29 0.46 0.14 0.35 0.49 0.50 0.39 0.49 0.73 0.45 0.57 0.50 0.03 0.18 0.13 0.33 0.23 0.41 0.04 0.20 2,805.73 173.31 F F 6.72 3.41 10,810
n The confidentiality agreement with the EEOC disallows presenting descriptive statistics for the EEOC district office.
instance, resolutions can mandate that employers institute affirmative action plans, hire an EEO coordinator, formally post job openings, or revamp pay systems. Thus my fourth dependent variable is a binary measure coded 1 if the charge resolution involves a mandated change in workplace policy for the accused employer and coded 0 otherwise. Table 1 provides descriptive statistics for all dependent and independent variables. Independent Variables
Theoretically, I am interested in the extent to which employing organizations leverage their legal experience and resources to influence charge outcomes. To assess the effect of employers’ experience with the charge resolution process, I employ a measure of the number of charges filed against the establishment in previous years. I calculate this measure by summing the number of charges filed against each establishment since 1990. This measure includes all previous charges filed against each establishment contained in
Hirsh
257
the EEOC database since 1990. However, because the data are longitudinal, the number of years observed since 1990 will vary depending on when the charge was filed. For instance, for charges filed in 1991, data on prior charges will cover one year, while for charges filed in 2002, data on prior charges will cover 12 years. To account for such variation, I include a control variable indicating the number of years observed since 1990.10 Next, I employ two measures to assess organizational resources. To the extent that larger establishments possess greater resources than smaller establishments, my first measure is the total number of employees (full- and part-time) at each establishment. And because most establishments in my sample (90 percent) are subsidiaries of larger organizations and can presumably draw on the resources of their parent organization when faced with legal claims, I also include a measure of the total number of workers employed by the parent organization.11 In addition to legal experience and size, evidence of compliance with EEO law may affect the likelihood that establishments receive unfavorable charge outcomes. To investigate the effect of EEO-related policies on charge outcomes, I use a binary variable to identify establishments that hold federal contracts and are thus subject to affirmative action requirements enforced by the Office of Federal Contract Compliance (OFCCP).12 To assess the influence of substantive indicators of EEO compliance, I employ four measures of the extent to which women and racial minorities are represented in the workplace and across job levels. To measure the presence of women and minorities in the workplace, I include measures of the proportion of female employees in the establishment and the proportion of employees who are members of racial minority groups, including African Americans, Asian Americans, and Hispanics. Next, to assess the extent to which these groups are represented across job levels, I include measures of sex- and raceoccupational segregation. The index of dissimilarity (D) measures the extent of sex- and race-occupational segregation in the workplace (Massey & Denton 1988:284) by summarizing how similarly 10 To examine whether the left-censored nature of this measure may be picking up on linear effects in the data, I also estimated models using a measure of experience with the charge resolution process that included only charges filed in the previous year. Results (available upon request) were similar to those presented here. 11 In regression models, I employ a log transformation of establishment size and parent size. 12 While this measure identifies establishments that presumably have affirmative action programs, I lack information on additional EEO structures, such as the presence of diversity staff, offices, and training programs. To more fully address the impact of EEO structures on discrimination-charge outcomes, I would ideally have information on these additional structures.
258
Determinants of Discrimination-Charge Outcomes
two groups are distributed across occupational categories within each establishment. For sex segregation, I calculate the dissimilarity between female and male workers across nine broad occupational categories, and for race segregation, I calculate the dissimilarity index between whites and nonwhites.13 Finally, I include several control variables to take into account administrative factors and characteristics of charges that might influence outcomes. First, because EEOC enforcement statistics suggest that charge processing and resolutions differ for sex and race charges, I include a binary variable identifying charges citing sex discrimination. Second, to control for the race of complainants, I include dummy variables identifying charges filed by blacks, Asians, and other nonwhites, using charges filed by whites as the referent category.14 Third, since standards of proof and available evidence differ for charges involving various employment issues, I include a series of dummy variables that identify charges citing discrimination in hiring, promotion, pay, terms and conditions of employment, harassment, and other issues; charges involving termination are the referent category. Fourth, to the extent that class action claims imply evidence of systematic discrimination, I include a dummy variable identifying charges that the EEOC has designated as part of a class action case. Fifth, because these data are longitudinal and the EEOC’s budget varies from year to year, I include a measure of the EEOC’s annual budget (in millions of dollars) in the year the charge was filed. Finally, to account for variation in administrative processes across EEOC local offices, I include a set of 18 dummy variables identifying the EEOC district office where the charge was originally filed.15 However, due to the confidential nature of these data, I do not report model estimates for the EEOC district office measures.16
13 The nine broad occupational categories include officials and managers, professionals, technicians, sales workers, office and clerical, craft workers, operatives, laborers, and service workers. 14
While the sex of the complainant may also be important for charge outcomes, because charges alleging sex discrimination are predominantly filed by women, including a measure of the sex of the complainant resulted in collinearity. Thus I omit complainants’ sex from the analysis. 15 Due to model convergence problems, I do not include the EEOC district office dummy variables in models predicting the likelihood of a policy change mandate. 16 Another administrative characteristic that may affect the processing of claims is how the charge is categorized at intake (A, B, or C), according to the EEOC’s Charge Handling Priority system. Because this system was not fully implemented until 1997, and my charge data cover 1990 to 2002, the majority (79 percent) of charges in my dataset lack information on categorization. Thus I cannot incorporate categorization information into the analyses.
Hirsh
259
Models I employ logistic regression to estimate the likelihood that complainants receive (1) a favorable outcome, (2) nonmonetary benefits, (3) monetary benefits, and (4) mandated changes in workplace policy. Because the data include multiple charges for a given establishment, I include a random intercept term, u0j, to account for this nested structure. This random intercept term takes into account the dependence among charges filed against the same establishment by allowing the intercept of each establishment to vary. The models take this general form: pij ¼ g0j þ gp0 Xpij þ g0q Zqj þ u0j þ eij where p 5 the outcome of interest (complainant-favorable outcome, nonmonetary benefits, monetary benefits, policy change) for charge i filed against establishment j, i 5 charge, j 5 establishment, X 5 a vector of p covariates at the charge level, Z 5 a vector of q covariates at the establishment level, u0j 5 a random intercept term for establishment j, and eij 5 a residual error term. To examine the extent to which organizations influence both charge outcomes and the terms of settlement, I estimate two sets of models. First, to identify the determinants of charge outcomes, I estimate the likelihood that the complainant receives a favorable charge outcome using data on all 10,810 charges included in my sample. Results from these analyses will identify the determinants of complainant-favorable charge outcomes. Next, to examine how employers negotiate various forms of settlements, I estimate the likelihood that the complainant obtains (1) a nonmonetary settlement, (2) a monetary settlement, and (3) a change in workplace policy, among only cases that settle in some way. Results from these analyses will reveal the extent to which establishments draw on their organizational resources to structure the penalties associated with settlements.17
Results Charge Bases, Issues, and Outcomes
Before presenting the determinants of charge outcomes, I first provide a descriptive overview of the types of charges included in my sample and their outcomes. Figure 2 provides a breakdown of 17 In additional analyses, I also tried modeling charge outcomes using a multinomial logit model predicting five different types of charge outcomes. Results (available upon request) were consistent with those presented here. For ease of interpretation, I present results generated from logistic regression models.
260
Determinants of Discrimination-Charge Outcomes
70 EPA
60
Title VII
Percent
50 40 30 20 10 0 Sex
Race
National Origin
Multiple
Charge Basis
Figure 2. Breakdown of Charges by Bases, N 5 10,810.
charges by the basis of the claim; charges in my sample can cite discrimination on the basis of sex, race, national, origin, and multiple bases, such as sex and race. As seen here, roughly 60 percent of charges cite discrimination on the basis of race, about 40 percent of charges involve sex discrimination, and about 13 percent involve discrimination on the basis of national origin. Complainants can also file charges citing multiple bases of discrimination; for instance, black women often bring charges on the basis of sex and race. About 12 percent of charges involve multiple bases of discrimination, with the majority citing sex and race discrimination. Figure 3 provides a breakdown of the employment issues cited in charges filed against the establishments in my sample, separately for race and sex charges. Here and throughout the analysis, for ease of presentation I combine charges citing discrimination on the basis of race, color, and national origin into a single category denoted as ‘‘race’’ charges. As seen in Figure 3, termination is the most commonly cited issue among the charges in my sample; nearly half of all sex charges and race charges involve disputes over termination. Harassment and promotion are the next most commonly cited issues in discrimination claims; however, the relative frequency of each varies for sex and race charges. For sex charges, complainants cite harassment in about 43 percent of all cases but cite promotion in only 15 percent of cases. For race discrimination, both harassment and promotion each constitute roughly 20 percent of all race claims. Thus while harassment is a commonly cited issue in both sex and race claims, sex charges more frequently involve harassment. Less than 10 percent of claims involve disputes
261
Hirsh
60 Race Sex 50
Percent
40
30
20
10
0 Hire
Promote
Terminate Harassment
Pay
Terms
Other
Charge Issue
Figure 3. Breakdown of Charge Issues by Sex and Race, N 5 10,810.
over pay, roughly 5 percent involve disputes over hiring, and about 10 percent involve disputes regarding other employment issues, such as benefits, working hours, or retaliation.18 With regard to charge outcomes, EEOC enforcement statistics suggest that fewer than one in five complaints filed with the agency result in outcomes favorable to the complainant (U.S. Equal Employment Opportunity Commission 2004). As displayed in Figure 4, my data are consistent with this estimate. Of the 10,810 complaints, only 16 percent result in favorable outcomes for charging parties. The share of complainants receiving benefits is similar; about 15 percent of complainants receive some kind of benefits, with roughly 10 percent receiving monetary benefits and 6 percent receiving nonmonetary benefits. Resolutions that mandate changes in workplace policy for employers are especially rare; policy change mandates occur in only 1 percent of all cases. Figure 5 provides a breakdown of the type of relief complainants receive in cases where the parties settle the dispute in some manner. Among cases that settle, complainants receive some kind 18 Because my sample is restricted to charges that allege sex or race/national origin discrimination, charges involving retaliation are only included in the sample if they also allege sex or race discrimination.
262
Determinants of Discrimination-Charge Outcomes
50 Race Sex 40
Percent
30
20
10
0 Complainant favorable
Any benefits
Monetary benefits
Non-monetary Policy change benefits
Outcome
Figure 4. Breakdown of Charge Outcomes by Sex and Race, N 5 10,810.
of benefit 90 percent of the time. About two-thirds of settlements include monetary penalties, with the typical monetary settlement totaling just over $20,000. In addition, almost half of settlements include nonmonetary benefits, while only 5 to 10 percent of settlements require employers to change workplace policies. Thus although settlements are uncommon, they typically involve some kind of monetary or nonmonetary relief for complainants yet rarely require a policy change for employers. Next, I examine how employers’ legal experience, resources, and demonstrations of EEO compliance affect the likelihood of these outcomes. The Determinants of Charge Outcomes
Table 2 presents logistic regression estimates of the effect of establishments’ legal experience, size, EEO compliance, and charge characteristics on whether the resolution is favorable to the complainant. The results identify an advantage for establishments with prior legal experience in resolving discrimination charges. As seen in Table 2, the results support Hypothesis 1, that establishments with prior experience with the charge resolution process should be less likely to receive complainant-favorable outcomes. With every prior charge of discrimination, the odds that the complainant will
Hirsh
263
100 90
Race
80
Sex
Percent
70 60 50 40 30 20 10 0 Any Benefits
Nonmonetary benefits
Monetary benefits
Policy change
Outcome
Figure 5. Breakdown of Benefits Among Cases that Settle by Sex and Race, N 5 1,535.
receive a favorable outcome decrease by about 1 percent.19 Thus establishments that have previous experience with discrimination charge processing maintain an edge over less-experienced employers. I find mixed support for Hypothesis 2, regarding the influence of establishments’ size and organizational resources. While larger establishments are less likely to experience complainant-favorable outcomes, the size of the parent organization is not a significant predictor of charge outcomes. This suggests that the immediate resources that are available to establishments due to their size provide an advantage in resolving charges, but employers do not exploit the resources of their larger parent organizations. The results provide support for Hypothesis 3, regarding the effect of affirmative action requirements on charge outcomes. I hypothesized that the presence of affirmative action requirements would protect establishments against unfavorable charge outcomes by demonstrating commitment to equal employment opportunity law. Indeed, the results in Column 1 of Table 2 indicate that for establishments with federal contracts, thus subject to affirmative action requirements, the odds of observing a complainantfavorable outcome decrease by roughly 16 percent, as compared 19 I used the following formula to calculate the percentage change in the dependent variable associated with a unit change, d, in an independent variable, k (see Long 1997:225): 100 [exp(Bk d) 1].
264
Determinants of Discrimination-Charge Outcomes
Table 2. Logistic Regression Estimates of the Effects of Employer and Administrative Characteristics on Complainant-Favorable Charge Outcomes Complainant-Favorable Employer characteristics Prior charges Establishment size (log) Parent size (log) Federal contractor Proportion female Sex segregation, D Proportion minority Race segregation, D Administrative characteristics Hire Promote Terms and conditions Pay Harassment Other issue Sex Black Asian Other nonwhite Class action EEOC budget a EEOC district office Years observed since 1990 Intercept t20j s2ij Deviance N
b
SE
0.004nn 0.059n 0.025 0.176n 0.107 0.200 0.140 0.606n
0.001 0.030 0.023 0.082 0.175 0.251 0.221 0.291
0.205 0.140w 0.423n 0.044 0.025 0.316nn 0.197nn 0.064 0.017 0.015 1.033nn 0.000 F 0.014 1.871 0.358nn
0.131 0.078 0.202 0.113 0.069 0.086 0.069 0.079 0.172 0.096 0.140 0.000 F 0.016 0.977 0.052
0.921nn
0.013 7,689.8 10,810
w po0.10; npo0.05; nnpo0.01 (two-tailed). Controls for EEOC district office are included in the model but are not presented due to confidentiality concerns. a
to results for noncontractors. Thus affirmative action requirements shield employers from unfavorable outcomes in the charge resolution process. Despite the significant effect of affirmative action requirements, I find little support for Hypothesis 4, regarding the insulating capacity of substantive indicators of compliance, including sex and race composition and occupational integration. Among the compositional measures, only race segregation is a significant predictor of charge outcomes; however, the effect of race segregation is in the opposite direction than expected. The more racially segregated the establishment, the less likely complainants are to experience favorable outcomes. Considering the influence of administrative factors on charge outcomes, results suggest that characteristics of the charge and variation in administrative context are important predictors of charge outcomes. Charges alleging sex discrimination are about 22 percent more likely to result in complainant-favorable outcomes as
Hirsh
265
compared to charges alleging race discrimination. Turning to the employment issue involved in the claim, the results presented in Table 2 suggest that claims involving promotion, terms and conditions of employment, and other employment issues are more likely to result in complainant-favorable outcomes than disputes over termination. As predicted, complainants are especially likely to prevail if they are part of a larger class action claim; being part of a class action charge increases the odds of a complainant-favorable outcome by 180 percent, as compared to unaffiliated charges. With regard to administrative context, results indicate that the EEOC’s annual budget is not a significant predictor of observing a complainant-favorable outcome. Due to the sensitive nature of EEOC district office–level data, my agreement with the EEOC precludes me from reporting or interpreting model estimates for the EEOC district office dummy variables in Table 2; however, results suggest some variation in the odds of observing a complainant-favorable outcome by district office. Finally, the variance of the random intercept term, t20j , is statistically significant, indicating significant variation in the intercepts across establishments. The Penalties for Noncompliance
While the previous results reveal the extent to which organizations mediate complainant-favorable outcomes in general, I am also interested in how employers negotiate specific types of settlements. Thus in the next set of models, I estimate the likelihood of observing nonmonetary benefits, monetary benefits, and mandates for policy change among cases in which employers settle the dispute in some way. The first column of Table 3 presents logistic regression estimates of the effects of legal experience, establishment size, EEO compliance, and charge characteristics on the likelihood that the charge results in nonmonetary relief for complainants, among charges that settle. Similar to the model predicting complainant-favorable outcomes among all charges, the number of prior charges has a significant negative effect on the odds that complainants receive nonmonetary benefits; for every one additional charge, the odds of receiving nonmonetary benefits decreases by about 1 percent. However, unlike the results for all complainant-favorable outcomes, neither employers’ resources nor indicators of EEO compliance are significantly related to complainants’ receipt of nonmonetary benefits. Among administrative characteristics, complainants’ race is significantly related to the odds of receiving nonmonetary benefits; blacks, Asians, and other nonwhite complainants are significantly more likely to receive nonmonetary benefits through settlement as compared to their white counterparts.
266
Determinants of Discrimination-Charge Outcomes
Table 3. Logistic Regression Estimates of the Effects of Employer and Administrative Characteristics on Receiving Nonmonetary Benefits, Monetary Benefits, and Policy Change Mandates, Among Charges that Settle Nonmonetary benefits Employer characteristics Prior charges Establishment size (log) Parent size (log) Federal contractor Proportion female Sex segregation, D Proportion minority Race segregation, D Administrative characteristics Hire Promote Terms and conditions Pay Harassment Other issue Sex Black Asian Other nonwhite Class action EEOC budget a EEOC district office Years observed since 1990 Intercept t20j s2ij Deviance N
Monetary benefits
Policy change
b
SE
b
SE
b
SE
0.005n 0.030 0.011 0.125 0.293 0.100 0.472 0.013
0.002 0.055 0.044 0.151 0.317 0.488 0.403 0.562
0.011nn 0.051 0.002 0.213 0.757n 0.300 0.289 0.402
0.003 0.061 0.048 0.166 0.353 0.533 0.448 0.614
0.028nn 0.167 0.058 0.378 w 1.829 w 2.377 1.773 2.992
0.004 0.142 0.128 0.390 0.938 1.264 1.247 1.592
0.279 0.159 0.410 0.246 0.140 0.167 0.142 0.163 0.342 0.187 0.388 0.001
0.198 w 0.319 w 0.945 0.286 0.216 0.499nn 0.074 0.092 0.311 0.293 0.278 0.001
0.273 0.170 0.548 0.265 0.148 0.174 0.149 0.169 0.362 0.195 0.408 0.001
2.217nn 1.645nn w 1.331 0.095 1.190nn 2.188nn 0.296 0.730n 0.313 1.437nn 2.283nn 0.002
0.446 0.275 0.701 0.483 0.272 0.383 0.300 0.329 0.582 0.406 0.597 0.001
w
0.531 w 0.313 0.033 0.323 0.135 0.127 0.117 0.462nn 0.707n 0.425n 0.579 0.000
0.048 0.032 3.317 2.044 nn 0.444 0.126 0.893nn 0.037 1,686.3 1,535
0.049 0.034 3.523 2.187 nn 0.842 0.171 0.812nn 0.034 1,410.5 1,535
w
0.125 0.344 7.029nn 0.310nn 206.2 1,535
0.064 3.478 0.844 0.011
w po0.10; npo0.05; nnpo0.01 (two-tailed). Controls for EEOC district office are included in the models predicting nonmonetary and monetary benefits but are not presented due to confidentiality issues. EEOC district office measures were omitted from the model predicting policy change mandates because of convergence problems. a
The second column in Table 3 presents results of a similar model but predicts the likelihood that the settlement results in monetary relief for the complainant. The number of prior charges is positively related to the receipt of monetary benefits; for every one additional charge, the odds that the complainant receives monetary benefits increase by about 1 percent. Evidently, in situations where complainants can reach the settlement stage, employers with prior charge resolution experience are more likely to turn over monetary settlements. Neither establishment size nor parent organization size are significant predictors of monetary benefits among charges that settle. The analysis also provides little support for the effects of indicators of EEO compliance. Among the compliance measures, only the proportion of female employees is a significant predictor of monetary settlements, suggesting that establishments with more female employees are more likely to turn
Hirsh
267
over monetary relief. With regard to administrative factors, claims that involve disputes over promotion and the terms and conditions of employment are marginally more likely to receive monetary benefits, while claims that involve other employment issues are less likely to receive monetary relief, as compared to claims involving termination charges. Finally, the third column of Table 3 presents estimates of the effects of organizational and administrative factors on the likelihood that charge resolutions involve a mandated policy change for the accused employer among cases that settle. Again, the number of prior charges is positively related to policy change; for every previous charge, the odds that the charge results in a policy change mandate increases by 3 percent. This finding suggests that establishments with a history of prior charges are more likely to receive mandates for policy change. Sex segregation and the proportion of female employees are positively related to the likelihood that the resolution involves a policy change, though these effects are only marginally significant. Nevertheless, these results suggest that complainants and the EEOC may be more likely to push for policy change if the workplace is highly sex-segregated and if there are a substantial number of female employees who may benefit from the policy change. Several administrative characteristics are significantly related to the odds that charges result in policy change mandates. Nearly all employment issuesFincluding hiring, promotion, terms and conditions, and harassmentFare more likely to result in policy change mandates as compared to claims of termination. This is likely because policy change mandates are an appropriate remedy when the complainant remains employed at the charged establishment and thus can benefit from the policy change. Notably, while the results do not indicate an advantage for sex over race charges as in previous models, they do suggest that black and other nonwhite complainants are less likely than white complainants to receive policy change mandates. Finally, class action resolutions are considerably more likely to involve mandates for policy change, likely due to the systematic nature of class charges, as compared to stand-alone charges. Indeed, more than half of all settlements with mandated policy changes involve charges that are part of larger class action claims.
Discussion The foregoing results have several implications for the charge resolution process and the study of organizations’ response to the law more broadly. First, employers that have prior experience with
268
Determinants of Discrimination-Charge Outcomes
the charge resolution process maintain an advantage in resolving future charges. The more experience employers have with the charge resolution process, the less likely they are to receive unfavorable charge outcomes, net of the administrative characteristics of the charge. In keeping with Galanter’s thesis (1974), this finding suggests that larger and more experienced organizations can draw on their institutional know-how to navigate the charge resolution process to avoid unfavorable outcomes. Second, in situations where employers are willing to settle, however, establishments with prior experience with the charge resolution process are less likely to receive nonmonetary benefits but more likely to pay monetary damages and receive mandates to change their employment policies. In other words, among cases that settle, legally experienced actors receive tougher penalties. One possible explanation for this pattern is that although prior charges do not necessarily indicate a history of discriminatory practices, they do provide a record of employee grievances over discrimination at the charged organization. Complainants as well as EEOC officials may take a harder line, pushing for monetary relief and policy changes, in negotiating settlements with such employers. Indeed, workers often inquire about employers’ prior record when initiating their own grievances in an effort to legitimize their workplace experience and claim. It is also possible that experienced organizations are more likely to recognize cases in which the evidence against them is especially strong and will accept stricter penalties at the EEOC in order to avoid litigation. In other words, repeat players may strategically incur penalties from the EEOC to avoid potentially more costly payouts in court. Third, indicators of organizations’ EEO compliance afford employers some bargaining power in the charge resolution process. The results indicate a negative relationship between federal contract status and the likelihood that complainants receive a favorable outcome. This finding suggests that federal contract status and the affirmative action requirements that follow lend credibility to employers’ defense, making complainant-favorable outcomes less likely. As I observed at the district EEOC office, many organizations argue this point directly when responding to discrimination claims; employers often highlight the success of their affirmative action programs or provide copies of their affirmative action goals and timetables as evidence of their commitment to fair employment practices. The results presented here suggest that such displays are convincing to the extent that they protect employers against unfavorable charge outcomes. However, other forms of EEO complianceFincluding measures of gender and racial equity in the workplaceFdo not offer organizations much mileage in terms of avoiding unfavorable
Hirsh
269
charge resolutions. Evidently, the more consequential signal of EEO compliance for charge outcomes is contractor status as compared to compositional measures of sex and race equity in the workplace. Moreover, the effect of contractor status on complainant-favorable outcomes holds net of the presence of women and racial minorities in the workplace and levels of occupational segregation by sex and race. Consistent with an institutional view of the law and organizations, these findings suggest that structural indicators of legal compliance, such as affirmative action requirements, can take on legitimacy in the legal realm regardless of establishments’ empirical progress toward gender and race equality. While affirmative action requirements are suggestive of increased attention to fair employment practices, they do not guarantee antidiscrimination or gender and race equity in practice (Edelman & Petterson 1999; Hirsh & Kornrich 2008). Yet in the context of charge resolution, such EEO compliance structures lend legitimacy to employers’ claims. Fourth, this analysis identifies a clear advantage for charges that are part of larger class action cases. Charges designated as class action claims are more likely to receive favorable outcomes, nonmonetary benefits, monetary relief, and changes in employment policy, all else remaining equal. Presumably, complainants involved in a class action claim can share resources, evidence, and bargaining power to boost their chances against employers. Thus although rare, class designations afford complainants considerable leverage in negotiating charge outcomes and the terms of settlements. Finally, these results confirm that complainants alleging race discrimination are less likely to prevail as compared to complainants alleging sex discrimination. Indeed, race claims are less likely to result in complainant-favorable outcomes, net of employers’ experience and resources as well as other administrative characteristics of the claim. The source of this disparity may be due to variation in the issues raised in sex and race claims. As shown in Figure 3, sex charges are more likely to involve harassment while race charges more typically involve disputes over personnel-related issues, such as termination, promotion, and hiring. While not the focus of the current study, additional analysis (available upon request) suggests that sexual harassment charges are especially likely to result in complainant-favorable outcomes as compared to all other claims, when considered independent of organizational and administrative controls. This disparity in favorable outcomes for sex versus race claims is particularly interesting given that Title VII was originally introduced to eradicate a history of discrimination against racial minorities, specifically African Americans; sex was added to the list of protected classes as a last-ditch effort by opponents to thwart its
270
Determinants of Discrimination-Charge Outcomes
passage. Moreover, recent research documents a declining significance of race in terms of the prevalence of discrimination-charge filings relative to other protected classes (Wakefield & Uggen 2004). The evidence presented here provides one possible explanation for this decline: claims of discrimination based on race or national origin seldom produce favorable outcomes. Given the odds of success, coupled with the social, psychological, and economic costs of formally alleging discrimination, workers may be reluctant to mobilize their rights. As Bumiller explains, many workers who perceive discrimination on the job do not pursue claims because they perceive their struggle as a doomed battle against ‘‘the corporation’’ (1988:25). In addition, in studies of discrimination lawsuits, Donohue and Siegelman (1991, 2005) demonstrate that the likelihood and size of rewards play an important role in determining the number of cases brought as well as the rate of plaintiff victory. In short, if workers perceive charges of race discrimination as largely unwinnable, they are left with little incentive to mobilize their rights.
Conclusion This article began with a question regarding employing organizations’ response to EEO law enforcement and the extent to which charge resolution provides redress for potential victims of discrimination. In general, the analysis presented here raises concerns about the capacity of the current discrimination claiming model to remedy discrimination in the labor market. The discrimination-charge resolution process offers relief to a small share of all workers who file formal charges, and organizational attributesF most notably, employers’ legal experienceFinform charge outcomes net of the administrative characteristics of the case. These findings have broad theoretical implications for the study of the law, organizations, and regulatory systems. First, this study reiterates and extends the insight that the law is an evolving institution, subject to influence by the social actors and institutions that it both protects and regulates. Antidiscrimination law enforcement is not an objective, clear-cut, or linear process; rather, regulatory efforts are subject to influence and redirection even as they are applied. For instance, the results presented here suggest that employing organizations affect how EEO violations are understood and punished; however, the nature of organizations’ impact varies as the regulatory process unfolds. The pivotal point in the charge resolution process is settlement. If employers avoid settlement, charge outcomes advantage repeat players; yet when employers
Hirsh
271
are compelled to settle, repeat players are especially likely to suffer penalties. This formative nature of the charge resolution process complicates the incentive structure for compliance. As Rosenberg (1991:33) points out, the effectiveness of the law for producing social change depends in part on the credible threat of sanctions in the event of noncompliance. However, if employers approach discrimination-charge adjudication as a system that is subject to influence, there is more reason for employers to ‘‘bulletproof ’’ (Bisom-Rapp 1999) against potential liability rather than interrogate employment practices that might, often unintentionally, disadvantage protected groups. Second, the foregoing results are consistent with both resourcedriven and cultural views of organizational response to the legal process. Evidently, employers can effectively mobilize their legal experience and resources to avoid unfavorable outcomes, suggesting that experienced and well-endowed organizations maintain an advantage in legal resolutions. However, the resource advantage is only part of the story. Consistent with Edelman and colleagues’ (Edelman 2005; Edelman et al. 1999) theory of legal endogeneity, I find that organizations influence the charge resolution process through demonstrations of compliance. Organizations’ indicators of EEO compliance play an important role in determining whether potential legal violations are resolved in favor of complainants. Thus while EEO law constrains the behavior of organizations by licensing and limiting certain employment behaviors, organizations also influence how the law is applied and with what outcomes. Moreover, to the extent that EEO structures provide some protection against unfavorable outcomes, organizations may be quick to adopt such structures without examining substantive patterns of inequity, such as racial and gender divisions of labor and workplace diversity. While this research is an initial step toward unraveling the role employing organizations play in shaping legal outcomes in the context of employment discrimination law, its findings as well as limitations suggest several avenues for future research. First, given the importance organizations’ federal contractor status and related affirmative action requirements play in structuring charge outcomes, future research should more thoroughly detail how organizations leverage such policies to their advantage. For instance, what proportion of organizations mention their affirmative action plans in defending claims, and how is this defense received by regulatory officials? Moreover, how do additional indicators of EEO compliance, including EEO units and officers, diversity training programs, and internal grievance procedures, affect the evolution of legal resolutions?
272
Determinants of Discrimination-Charge Outcomes
Second, while findings suggest that experienced organizations are more likely to incur stiff penalties at the point of settlement, the mechanism driving this effect is unclear. It may be the result of the EEOC taking a hard line against organizations with prior charges, it may reflect a tendency among experienced organizations to strategically accept tough penalties at the EEOC to avoid the possibility of litigation, or it may be a combination of both. Future research might consider these alternatives by examining charge negotiations in more detail. Additional qualitative research on charge resolution at the EEOC or other claims processing agencies would be particularly useful in this regard. Third, to the degree that sex and race charges have different fates in the administrative process and the potential consequence of this on workers’ rights mobilization, future research should examine the race/sex disparity in charge outcomes in greater depth. One plausible direction is exploring how the issues raised in disputes may contribute to disparate charge outcomes by sex and race. Finally, while the findings presented here raise some concern regarding the capacity of the EEO regulatory process to redress individual claims of employment discrimination, it is important to recognize that the EEOC process represents one step in the larger EEOC regulatory framework. Disputes over discrimination also play out within employing organizations through internal dispute resolution and in the federal courts through private lawsuits. Thus future research should address if and how organizations leverage their legal experience, resources, and compliance symbols in resolving discrimination disputes in these additional legal contexts.
References Albiston, Catherine (2003) ‘‘The Rule of Law and the Litigation Process: The Paradox of Losing by Winning,’’ in H. M. Kritzer & S. Silbey, eds., In Litigation: Do the Haves Still Come Out Ahead? Stanford, CA: Stanford Univ. Press. FFF (2005) ‘‘Bargaining in the Shadow of Institutions: Competing Discourses and Social Change in the Workplace Mobilization of Civil Rights,’’ 39 Law & Society Rev. 11–50. Bisom-Rapp, Susan (1999) ‘‘Bulletproofing the Workplace,’’ 26 Florida State University Law Rev. 959–1049. Blumrosen, Alfred (1993) Modern Law: The Law Transmission System and Equal Employment Opportunity. Madison: Univ. of Wisconsin Press. Bumiller, Kristin (1988) The Civil Rights Society. Baltimore: Johns Hopkins Univ. Press. Burstein, Paul (1989) ‘‘Attacking Sex Discrimination in the Labor Market: A Study in Law and Politics,’’ 67 Social Forces 641–65. Burstein, Paul, & Kathleen Monaghan (1986) ‘‘Equal Employment Opportunity and the Mobilization of Law,’’ 20 Law & Society Rev. 355–88. Dobbin, Frank, et al. (1993) ‘‘Equal Opportunity Law and the Construction of Internal Labor Markets,’’ 99 American J. of Sociology 396–427.
Hirsh
273
Donohue, John J., & Peter Siegelman (1991) ‘‘The Changing Nature of Employment Discrimination Litigation,’’ 43 Stanford Law Rev. 983–1033. FFF (2005) ‘‘The Evolution of Employment Discrimination Law in the 1990s: A Preliminary Empirical Investigation,’’ in L. B. Nielson & R. L. Nelson, eds., Handbook of Employment Discrimination Research: Rights and Realities. New York: Springer Press. Edelman, Lauren B. (1990) ‘‘Legal Environments and Organizational Governance: The Expansion of Due Process in the American Workplace,’’ 95 American J. of Sociology 1401–40. FFF (1992) ‘‘Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law,’’ 97 American J. of Sociology 1531–76. FFF (2005) ‘‘Law at Work: The Endogeneous Construction of Civil Rights,’’ in L. B. Nielson & R. L. Nelson, eds., Handbook of Employment Discrimination Research: Rights and Realities. New York: Springer Press. Edelman, Lauren B., & Stephen M. Petterson (1999) ‘‘Symbols and Substance in Organizational Response to Civil Rights Law,’’ 17 Research in Social Stratification and Mobility 107–35. Edelman, Lauren B., & Mark C. Suchman (1999) ‘‘When the Haves Hold Court: Speculations on the Organizational Internalization of the Law,’’ 33 Law & Society Rev. 941–91. Edelman, Lauren B., et al. (1999) ‘‘The Endogeneity of Legal Regulation: Grievance Procedures as Rational Myth,’’ 105 American J. of Sociology 406–54. Galanter, Marc (1974) ‘‘Why the ‘Haves’ Come Out Ahead: Speculation on the Limits of Legal Change,’’ 9 Law & Society Rev. 95–160. Graham, Hugh D. (1990) The Civil Rights Era: Origins and Developments of National Policy. New York: Oxford Univ. Press. Heinz, John P., et al. (1998) ‘‘The Changing Character of Lawyer’s Work: Chicago in 1975 and 1995,’’ 32 Law & Society Rev. 751–75. Hirsh, C. Elizabeth, & Sabino Kornrich (2008) ‘‘The Context of Discrimination: Workplace Conditions, Institutional Environments, and Sex and Race Discrimination Charges,’’ 113 American J. of Sociology 1394–432. Kalev, Alexandra, et al. (2006) ‘‘Best Practices or Best Guesses? Diversity Management and the Remediation of Inequality,’’ 71 American Sociological Rev. 589–917. Lancaster, Ryon, et al. (2006) ‘‘Social Structure and Formal Law: Social Attributes and the Outcomes of Employment Discrimination Cases.’’ Presented at the American Sociological Association, Montreal, Quebec (August). Lillard, Margaret (2005) ‘‘NC Department Deemed Racially Hostile,’’ Yahoo News, http://news.yahoo.com/s/ap/noose_discrimination (accessed 18 May 2005). Long, J. Scott (1997) Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Marshall, Anna-Marie (2005) ‘‘Idle Rights: Employees’ Rights Consciousness and the Construction of Sexual Harassment Policies,’’ 39 Law & Society Rev. 83–123. Massey, Douglas S., & Nancy A. Denton (1988) ‘‘The Dimensions of Residential Segregation,’’ 67 Social Forces 281–315. Nielson, Laura Beth, & Robert L. Nelson (2005) ‘‘Scaling the Pyramid: A Sociolegal Model of Employment Discrimination Litigation,’’ in L. B. Nielson & R. L. Nelson, eds., Handbook of Employment Discrimination Research: Rights and Realities. New York: Springer Press. Nelson, Robert L., & William P. Bridges (1999) Legalizing Gender Inequality: Courts, Markets, and Unequal Pay for Women in America. New York: Cambridge Univ. Press. Robinson, Corre L., et al. (2005) ‘‘Studying Race or Ethnic and Sex Segregation at the Establishment Level: Methodological Issues and Substantive Opportunities Using EEO-1 Reports,’’ 32 Work and Occupations 5–38. Rosenberg, Gerald N. (1991) The Hollow Hope: Can Courts Bring About Social Change? Chicago: Univ. of Chicago Press.
274
Determinants of Discrimination-Charge Outcomes
Schultz, Vicki (1990) ‘‘Telling Stories about Women and Work: Judicial Interpretations of Sex Segregation in the Workplace in Title VII Cases Raising the ‘Lack of Interest’ Argument,’’ 103 Harvard Law Rev. 1750–843. Sturm, Susan (2001) ‘‘Second Generation Employment Discrimination: A Structural Approach,’’ 101 Columbia Law Rev. 458–568. Sutton, John R., & Frank Dobbin (1996) ‘‘Two Faces of Governance: Responses to Legal Uncertainty in U.S. Firms, 1955–1985,’’ 61 American Sociological Rev. 794–811. U.S. Equal Employment Opportunity Commission (2004) ‘‘Charge and Litigation Statistics from the U.S. Equal Employment Opportunity Commission, FY 1992 through 2003,’’ http://www.eeoc.gov/stats/all.html (accessed 18 Dec. 2004). U.S. Equal Employment Opportunity Commission (2005) ‘‘Home Depot to Pay $5.5 Million to Resolve Class Discrimination Lawsuit in Colorado,’’ http://www.eeoc.gov/ press/8-25-04.html (accessed 24 May 2005). U.S. Equal Employment Opportunity Commission (2007) ‘‘Definition of Terms,’’ http:// www.eeoc.gov/stats/define.html (accessed 1 March 2007). Wakefield, Sara, & Christopher Uggen (2004) ‘‘The Declining Significance of Race in Federal Civil Rights Law,’’ 74 Sociological Inquiry 128–57. Westin, Alan F., & Alfred G. Feliu (1988) Resolving Employment Disputes without Litigation. Washington, DC: Bureau of National Affairs. Wheeler, Stanton, et al. (1987) ‘‘Do the Haves Come Out Ahead? Winning and Losing in State Supreme Courts, 1870–1970,’’ 21 Law & Society Rev. 403–45.
Statutes Cited Civil Rights Act of 1964, 42 U.S.C. 2000e to 2000e-17. Equal Pay Act of 1963, 29 U.S.C. 206(d). Pregnancy Discrimination Act of 1978, 42 U.S.C. 2000e(k).
Elizabeth Hirsh is Assistant Professor in the Department of Sociology at Cornell University. Her research interests are in the areas of workplace inequality, organizational structures, and the relationship between law and organizations, particularly with respect to equal employment opportunity law. Her current research focuses on employment discrimination and the consequences of alternative organizational and legal arrangements for gender and race inequality in the workplace.
275
Incarceration, Health, and Racial Disparities in Health
Michael Massoglia
This article addresses two basic questions. First, it examines whether incarceration has a lasting impact on health functioning. Second, because blacks are more likely than whites to be exposed to the negative effects of the penal systemFincluding fractured social bonds, reduced labor market prospects, and high levels of infectious diseaseFit considers whether the penal system contributes to racial health disparities. Using the National Longitudinal Survey of Youth and both regression and propensity matching estimators, the article empirically demonstrates a significant relationship between incarceration and later health status. More specifically, incarceration exerts lasting effects on midlife health functioning. In addition, this analysis finds that, due primarily to disproportionate rates of incarceration, the penal system plays a role in perpetuating racial differences in midlife physical health functioning.
T
he rapid expansion of the correctional system is one of the most significant and dramatic trends in the legal system and contemporary American society. As of 2004, there were approximately six times more inmates and ex-inmates than in the mid-1970s. Presently, there are more than 16 million felons and ex-felons in the United States (Uggen et al. 2006). The growth of penal law in recent decades represents an increase in government social control, which in theory falls disproportionately on the lower classes and has implications for other institutions, such as employment, education, medicine, and public health (Black 1976, 1998). Little research, however, examines exposure to the penal system as an explanatory factor in health outcomes or racial disparities in health. Prior research in the law and society tradition has investigated how mental health influences criminal sanctions (Hochstedler 1986) and how law and medicine can represent This research was supported by the National Institute of Mental Health (#MH19893). I am especially indebted to Ryan King, Glenn Firebaugh, Christopher Uggen, Jason Schnittker, Carroll Seron, and three anonymous reviewers for their comments on an earlier draft of this manuscript. Please direct all correspondence to Michael Massoglia, Department of Sociology, Penn State University, 211 Oswald Tower, University Park, PA 16802; e-mail:
[email protected].
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
276
Race, Health, and Prison
competing institutions (Heimer 1999), yet little work to date explicitly documents the health consequences of an ever-expanding system of penal social control. The present research undertakes that task by empirically examining two questions. First, to what extent does exposure to penal incarceration influence midlife physical health functioning? Second, given the racial disparities in the criminal justice system, does the ever-growing penal system account for some of the persistent racial disparities in health? While all demographic groups are impacted by the expansion of the penal state, the phenomenon has disproportionately affected various subgroups of the population, in particular black males (Western 2006). In 2002, approximately 12 percent of black males were in correctional facilities (Harrison & Karberg 2003). The lifetime cumulative risk (measured to age 34) of imprisonment for all African American males is more than 20 percent (Pettit & Western 2004). Among African American males without a high school diploma, the lifetime risk of incarceration is 58.9 percent (Pettit & Western 2004). While strikingly high, these estimates carry additional meaning given that they are at least five times higher than the rates of comparable whites. In perhaps the most striking assessment of the scope and reach of the correctional system, Uggen et al. (2006) argue that correctional policies have caused the emergence of a new ‘‘felon class’’ in society. They estimate that this new ‘‘class’’ comprises approximately 7.5 percent of the adult population, 22.3 percent of the black adult population, and 33.4 percent of the black adult male population. In light of this rapid expansion of the penal system, a number of observers have considered the implications of the growing size and racial composition of the incarceration system. There is an extensive and growing literature on how crime and punishment impact later life chances and outcomes. Research links earlier crime and punishment with later educational outcomes, employment and marital processes, and the labor market (Hagan 1993, 1997; Lopoo & Western 2005; Pager 2003; Sampson & Laub 1990; Pettit & Western 2004; Tanner et al. 1999; Western 2002, 2006; Western et al. 2001). Research consistently finds that contact with the penal system both lowers the likelihood of obtaining gainful employment and depresses wages in the event of employment (Pager 2003; Western 2002), and disrupts marital stability as well (Lopoo & Western 2005). As a function of differential incarceration rates, minorities disproportionately carry these labor and marriage market deficits. Moreover, Pager concludes that a criminal record is more detrimental to the employment prospects of blacks than whites (2003:961). Other studies reach similar conclusions, finding that
Massoglia
277
the correctional system disproportionately impacts the marriage market in African American communities (Staples 1987; Wilson 1987). Thus both as a function of who is incarcerated, meaning racial differences in rates of incarceration, and as a function of how, meaning racial differences in the consequence of incarceration, the penal system appears to impact minorities more severely. To the extent that research has examined the health consequences of incarceration, the focus has been on the rather immediate impact of prisons on health outcomes such as suicide, depression, and coping (Liebling 1999; Liebling & Maruna 2005; Kruttschnitt & Gartner 2005; Porporino & Zamble 1984; Toch & Adams 1989) or problems that impact a relatively small percentage of the population such as severe health limitations (Schnittker & John 2007). Emerging work has considered how incarceration may contribute to patterns of HIV infections (Johnson & Raphael 2006). The current research furthers that tradition by investigating the lasting consequences of incarceration on general health functioning for a large sample of midlife adults. The article first elaborates on the theoretical and empirical linkages suggestive of a relationship between incarceration status and health, considering such factors as exposure to stress and major life events. The data, methods, and logic of analysis are then presented. Using both propensity score and regression estimators, the results show a significant effect of incarceration on later health and indicate that the penal system accounts for a sizeable proportion of racial disparities in general health functioning. Finally, in the conclusion the article argues that the penal system has grown to the point where it is now a system of stratification touching almost all aspects of contemporary American society, including health functioning, and merits a position alongside traditional systems of health stratification such as the occupational and educational system.
Theoretical and Empirical Linkages Between Incarceration and Health: Inside the Prison and After Incarceration Multiple research traditions are suggestive of a significant association between exposure to the penal system, for instance, incarceration, and later health. These include the literature on exposure to stress, research in the social gradient tradition, work on stratification and health, and life course studies. Stress
The stress literature has traditionally classified exposure as either a major ‘‘life event’’ or ‘‘chronic stress’’ associated with given
278
Race, Health, and Prison
social roles, positions, or life events (Thoits 1995). Incarceration appears to map over both classifications. On the one hand, ethnographic accounts of the prison experience suggest that incarceration is a dramatic ‘‘life event’’ (Sykes 1971; Hassine 2004). On the other hand, incarceration places individuals at a disadvantaged social position (Pager 2003; Western 2002) that likely exposes them to more chronic stress over the life course. Pearlin’s (1989) theoretical framework of primary and secondary stressors appears particularly informative, as the physical spell of incarceration may act as a primary stressor, and upon release individuals are exposed to a number of secondary stressorsFbe they family, employment, or socialFthat result in prolonged exposure to stress. Prolonged exposure to stress leaves the body in a heightened state of awareness that ultimately taxes the cardiovascular and immune systems. This leaves individuals at increased risk for both mental and physical health problems (Lazarus & Folkman 1984; Pearlin 1989). More recently it has become clear that severe or chronic stress can fundamentally alter the body and permanently alter and weaken its ability to respond to additional stressors (McEwen 1998; Fremont & Bird 2000). That is, the body’s ability to maintain health is permanently damaged. While the stressors of incarceration differ from those after release, the totality of the incarceration experienceFfrom fear or isolation while incarcerated (Sykes 1971) to labor market and family problems that released inmates face (Western 2002; Lopoo & Western 2005)Fmay fundamentally alter an individual’s ability to effectively regulate health functioning. Social Gradient and Social Location
The Whitehall Studies, which investigate the relationship between social status and health, are among the most influential works on the life course determinants of health (Marmot et al. 1984; Marmot 2004; Bosma et al. 1998). Based on British social servants, the Whitehall Studies (Marmot et al. 1984; Marmot 2004) show that health is related to individuals’ abilities (or perceived abilities) to control their life and participate fully in society. Even after accounting for factors such as access to health care and financial circumstances, those at the higher end of the social gradient are better able to exercise control over their life, while those in lower social classes are less able to control their life and participate fully in society. It is reasonable to hypothesize that incarceration lowers an individual’s position in the social hierarchy, through both the stigma of ex-con status and the related economic and employment deficits. It logically follows that incarceration may impact health. In
Massoglia
279
addition, given the emphasis of the legal and penal system on removing individuals and regulating and controlling behavior upon release, from Marmot’s theoretical perspective it appears that the specific intent and goal of incarceration and post-correctional policies is inconsistent with positive health outcomes from a social hierarchy perspective. The Life Course and Life Events
Given the disruptive impact of incarceration on wages, employment, and marriage (Western 2002; Pager 2003; Lopoo & Western 2005), one mechanism linking incarceration and health is through the social and life course processes associated with both health and incarceration, in particular marriage and employment. Married individuals are in better physical health than the nonmarried across a variety of indicators (Ross et al. 1990; Anson 1989; Litwak & Messeri 1989).1 Employment also affects health status. Relative to those without jobs, people who are employed report better health (Turner 1995; Ross & Bird 1994; Ross & Mirowsky 1995; Verbrugge 1989). One could hypothesize that incarceration impacts health by lowering income and employability and by severing the social bonds, such as marriage, that are associated with health. Second, and closely related to the stress literature, research has paid particular attention to how major life events such as divorce, loss of a job, or loss of a loved one adversely impact health (Barrett 2000; Kurdek 1990, 1991; Kessler et al. 1989; McLeod 1991; Mechanic & Hansell 1989; Turner 1995). The key theoretical notion is that these events are moments in the life course that require major behavioral adjustments in a relatively short period of time (Thoits 1995). It is reasonable to hypothesize that incarceration is such a moment in the life course (Thoits 1995:54) and thus affects health in a manner consistent with the life events framework. Indeed, there is a relatively developed literature that examines the problems inmates face as they adjust to prison (MacKenzie et al. 1987; Jiang & Winfree 2006). Moreover, ethnographic accounts of prison almost universally identify the entrance to a spell of incarceration as a period characterized by rapid transition and adjustment (see for instance Hassine 2004). Finally, when taken in conjunction with the social epidemiology literature, incarceration may heighten exposure to disease. Data from the National Commission on Correctional Health Care (2002) suggest that incarceration exposes inmates to a number of 1 For competing perspectives on the role of marriage and health, see, for instance, Fu and Goldman (1996).
280
Race, Health, and Prison
infectious diseases. The commission report estimates that more than 20 percent of all individuals in the United States infected with HIV and almost 40 percent of individuals infected with hepatitis C pass through correctional institutions in a given year. These high rates of infectious disease provide another explanation for the relationship between incarceration and health; prisons differentially expose individuals to infectious diseases. Contemporary evidence supports this position. In recent decades, several outbreaks of tuberculosis in the United States have been traced to correctional facilities (Farmer 2002). In fact, Farmer (2002) argues that rates of tuberculosis in the New York facility Rikers Island are higher than those of many third world countries.
Racial Differences in Health and the Penal System In light of racial differences in incarceration rates, an explicit consideration of whether the incarceration system contributes to racial disparities in health is warranted. By almost any objective standard, the health of blacks is disadvantaged relative to whites (Smaje 2000; Hayward & Heron 1999). While specific estimates vary slightly, whites can expect to live approximately six years longer than African Americans, with the gap being greater for males than females (Ventura et al. 1997; Williams & Collins 1995; Rubio & Williams 2004; Hoyert et al. 2006). In addition, blacks are likely to live a greater portion of their lives with chronic health conditions (Hayward & Heron 1999; Hayward et al. 2000). Blacks can thus expect to live shorter lives, while dealing with more chronic pain and illness than whites. Efforts to explain racial differences in health focus on factors ranging from structural conditions to employment and lifestyle choices to psychological factors (Bird et al. 2000). While Schnittker and John (2007) find the incarceration system is unrelated to racial differences in severe health problems, to date no research has assessed whether the penal system impacts racial differences in general health functioning. As discussed, the rapid expansion of the penal system has not affected all racial groups equally. While specific estimates vary by subgroup (e.g., high school degree or not), minority males are five to eight times more likely to be incarcerated than comparably educated whites (Pettit & Western 2004). While health disparities were evident well before the expansion of the penal system, Link and Phelan (1995) argue that the cause of these health disparities likely changes over time. In light of evidence suggesting that the health status of blacks, relative to whites, has worsened in the same time period that roughly corresponds to the implementation of
Massoglia
281
punitive incarceration initiatives (Cooper 1993; Williams & Collins 1995; National Center for Health Statistics 2001; Freeman 1996), it may be that incarceration serves as an important and underresearched factor in the study of race and health. Consistent with the theoretical notion of Link and Phelan (1995), the penal system may have emerged as a system of health inequality. Such findings call for an investigation into the potential contribution of incarceration to racial differences in health outcomes. The following sections of the article outline the logic, data, and methods used to examine the impact of exposure to incarceration on health, and the contribution of the penal system to racial disparities in health. While a number of theoretical and empirical linkages are evident, it is important to craft the analysis while mindful of the potential for spuriousness. It may be that some individuals engage in behavior Ffor instance, illicit drug use or repeated violenceFthat increases both the likelihood that they are unhealthy or injured and the likelihood that they will be incarcerated.
Data, Logic of Analysis, and Methods As with related research on the life course consequences of incarceration (Lopoo & Western 2005; Western 2002) data from the National Longitudinal Survey of Youth 1979 (NLSY79) is used for analysis. The NLSY79 allows for an assessment of the lasting relationship between incarceration and health using a number of factors that predict incarceration and are associated with health. Moreover, the analysis can speak to specific theoretical frameworks previously introduced. For instance, the NLSY79 includes factors such as employment and poverty, which address Marmot’s theoretical notions of ‘‘social gradient’’ (2004). Moreover, while direct measures of stress are rare in social science research, the analysis includes measures of mental health that help speak to factors such as anxiety and stress levels. Finally, measures such as marriage or education level are consistent with prior empirical and theoretical treatments of the life course determinates of health. Complete descriptions of the data and variables are presented in Table 1. In conjunction with the Bureau of Labor Statistics, data collection began in 1979, when the respondents were between ages 14 and 22 (respondents were drawn from a nationally representative sample and were incarcerated at any point over the survey period). As part of the NLSY79 collection protocol, a comprehensive health questionnaire was given to respondents during the survey period immediately after they turned age 40. As of 2002, approximately 5,500 respondents had turned 40 and answered the health 40
Cocaine use
Weight Cigarette use Binge drinking
Lifestyle (1996–2000) Insurance Workout
Welfare Drug use Self-reported crime
Control orientation
Intact family Parents’ education
Ascribed Characteristics Gender Race Background Variables (1979–80) Prior health problems
Variable
Does the respondent have health insurance? Number of times weekly the respondent participates in vigorous physical activities Respondent’s weight in pounds Does the respondent report daily cigarette use? Does the respondent drink more than five drinks, at least 2–3 times a month? Has the respondent ever used cocaine?
0 5 No, 1 5 Yes
0 5 No, 1 5 Yes 1 5 Never to 5 5 More than 3 times a week 0–370 0 5 No, 1 5 Yes 0 5 No, 1 5 Yes
0 5 No, 1 5 Yes 0 5 No, 1 5 Yes 0 5 Never 1 5 Once 2 5 Twice 3 5 3–5 times 4 5 6–10 times 6 5 More than 10 times
1 5 Less control over life 2 5 Greater control over life
0 5 No, 1 5 Yes Grades completed 0–20
0 5 No, 1 5 Yes
Does the respondent have health problems that limit the type or amount of work they can do? Was the respondent raised in a two-parent household? Highest level of schooling completed for either mother or father Does the respondent . . . ? Have control over the direction of their life? Feel things in life are a matter of fortune or planning? Feel getting what they want has nothing to do with luck? Feel that chance or luck plays an important role in life? How close is this statement to your opinion? (asked for each item) Did the respondent’s family of origin receive welfare? Ever use drugs other than marijuana In the past year had the respondent . . . Intentionally damaged property Been in school/work fights Shoplifted Attacked someone Sold/used marijuana
Coding 0 5 Female, 1 5 Male 0 5 Other, 1 5 White
Description
Self-reported sex Self-reported race
Table 1. Descriptive Statistics
24%
174.9 27% 19%
81% 2.76
13% 47% 3.64
8.21
77% 10.79
06%
48% 71%
Mean
41.4
.47
4.47
1.58
3.31
Std. Dev.
282 Race, Health, and Prison
75% 27% 56% 13.08 73% 52.03 52.93 4.8%
0 5 No, 1 5 Yes 0 5 No, 1 5 Yes 0 5 No, 1 5 Yes 0–20 0 5 No, 1 5 Yes 0–100 0–100 0 5 No, 1 5 Yes
(8.08) (8.39) (0.20)
SF-12 Health Questionnaire: Specific Question-and-Response Wording In general, would you say your health is: Possible responses: Excellent, Very good, Good, Fair, Poor The following questions are about activities you might do during a typical day. Does your health limit you in these activities? If so, how much? (a) Moderate activities, such as moving a table (b) Pushing a vacuum cleaner, bowling, or playing golf (c) Climbing several flights of stairs Possible responses: Yes, Limited a lot, Yes, a little, Not limited at all During the past four weeks, how much of the time have you had any of the following problems with your work or other regular daily activities as a result of your physical health? (a) Accomplished less than you would like (b) Were limited in the kind of work or other activities Possible responses: Yes, No During the past four weeks, how much did pain interfere with your normal work (including both work outside the home and housework)? Possible responses: Yes, Limited a lot, Yes, a little, Not limited at all These questions are about how you feel and how things have been with you during the past four weeks. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time during the past four weeks . . . (a) Did you have a lot of energy? Possible responses: All the time, Some of the time, None of the time During the past four weeks, how much of the time has your physical health interfered with your social activities (such as visiting friends, relatives, etc.)? Possible responses: Not at all, A little bit, Moderately, Quite a bit, Extremely
Does respondent live in an urban setting? Does the respondent live at or below 125 percent of the poverty level? Marital status Is the respondent married? Education level Respondent’s years of school completed Labor force participation Is the respondent in the labor force? Dependent and Focal Independent Variables Mental Health Respondent’s Mental Health Composite Score Physical Health Respondent’s Physical Health Composite Score Incarceration Respondent Incarcerated?
Life Course/Social Location Urban residence Poverty status
Massoglia
283
284
Race, Health, and Prison
module of the survey.2 This group serves as the analytic sample for analysis. Respondents answered the health 40 module only once, and those who were not 40 by 2002 were omitted from analysis because they had not answered the module.3 The health 40 module employs the physical health component of the Short Form 12 (SF-12) to ask a series of questions on the physical and mental health functioning of respondents. The SF-12 serves as the dependent variable in this analysis and is an established indicator of health functioning (McDowell & Newell 1996) that probes the degree to which health problems affect the respondent’s lifestyle. The Bureau of Labor Statistics and NLSY79 standardize these multiple responses into a single, normally distributed, variable of self-reported physical health functioning with a theoretical range of 0–100 and a mean of 50 (sample range is 29–72, sample mean is 51.11).4 This indicator of self-reported physical health serves as the main dependent variable for this analysis. Self-reported health is the subject of considerable scholarly attention (Bound 1991; Browning & Cagney 2002; Hayward & Heron 1999; Ross & Wu 1996; Ferraro & Farmer 1996, 1999; Idler & Benyamini 1997).
Background and Demographic Variables
This analysis accounts for different phases of the life course. The first is a series of variables that measure background characteristics, taken in survey years 1979–1980. In addition to indicators of gender and race, this analysis considers measures of family welfare status, parents’ education level, whether respondents lived with two parents, and control orientation.5 Descriptive statistics for all variables are reported in Table 1. 2
As noted elsewhere, sample attrition is not impacted by incarceration status (Western 2002). 3 While all respondents answered the module at the same point in life, age 40, the differences in respondents’ age (14–22) at the beginning of data collection correspond to variation in the year when they reached age 40 and responded to the health module. The NLSY79 creates one variable for each question on the module and then creates a trichotomy indicating when an individual responds to the health 40 module. The analysis uses that trichotomy to test for cohort effects in response years 1998, 2000, and 2002. The results of this analysis indicate no significant health differences based on the survey response year. 4 The specific wording of each question in the SF-12, the survey used to create the self-reported health score that serves as the dependent variable for analysis, is presented at the end of Table 1. Readers interested in the procedure used by the NLSY79 and Bureau of Labor Statistics to create the composite indicator should consult the manual by Ware et al. (1995). 5 Control orientation represents the degree to which respondents feel they are able to control the direction of their lives.
Massoglia
285
In addition to family background measures, other key individual-level indicators are measured prior to respondents’ incarceration. Measures of criminal behavior, including prior substance use and self-reported crime, are included to separate the effect of punishment from that of crime and are a proxy for risk-taking behavior (Gottfredson & Hirschi 1990). Thus key control variables such as crime, psychological processes, social standing, and family structure are measured prior to risk of incarceration. Lifestyle and Life-Course Indicators
Consistent with existing literature, the analysis utilizes lifestyle indicators that tap both healthy and unhealthy aspects of respondents’ behavior and lifestyle choices. The analysis considers the respondent’s self-reported levels of cigarette use, binge drinking, and illicit drug use. In addition to substance use measures, selfreported weight, exercise habits and insurance status after the incarceration risk period are included. Contemporaneous life events are also included in the analysis. This block of factors includes whether respondents lived in an urban location, their poverty status and marital status, and employment indicators and educational attainment. Finally, given the strong relationship between mental and physical health, the analysis also includes contemporaneous measures of mental health. Focal Independent Variable and Lagged Measure of Health
The focal independent variable is incarceration. The data collection protocol is structured so that shorter incarceration spells, those less than one year, may be missed, but spells greater than one year will not be missed (Western 2002). The average length of incarceration is slightly more than three years.6 Accordingly, the incarceration indicator depicts invasive and prolonged exposure to a correctional setting rather than more fleeting involvement in the justice systemFfor instance, a short stay in jail resulting from a drunken driving arrest. Approximately 5 percent of the sample has been incarcerated, and these individuals are disproportionately members of racial minority groups. More specifically, during the time period under investigation, 270 individuals (182 minorities) were incarcerated. 6 Consistent with other work in the area (Schnittker & John 2007) additional analysis found that exposure rather than length of incarceration was more informative for understanding the relationship between incarceration and health. Supplementary analysis with incarceration treated as a continuous variable (mean 3.3, median 2, range 1–13) was consistent with the substantive findings presented here. Supplementary analysis is available upon request.
286
Race, Health, and Prison
As a key control variable, a lagged measure of problematic health is included in all models. Respondents are asked to report whether they have a health problem that prevents them from performing social or work functions. The data are arrayed to ensure that the measurement of prior health problems precedes risk of incarceration. Prior to incarceration, slightly more than 6 percent of the sample report having health problems. Including a lagged measure of health helps attenuate bias introduced from unmeasured heterogeneity in the sample. While incarceration is measured each year, this analysis considers only those who were incarcerated after 1979 and prior to 1996. This is done to maintain the temporal sequence underlying the model predicting health. In addition, the models include a lagged indicator of health functioning. Formally, the final models can be expressed as: Formula 1: Health40 ¼ a þ X1 b1 ðbgÞ þ X2 b2 ðprisonÞ þ X3 b3 ðlsÞ þ X4 b4 ðlcÞ þ e where X1b1(bg) is a vector of background variables, including prior health and accompanying parameter estimates, and X2b2(prison) represents incarceration history (0 for those never incarcerated, and 1 for those who were incarcerated) and associated parameter estimates of incarceration on health. Finally, X3b3(ls) and X4b4(lc) represent vectors of lifestyle and life course markers and their accompanying parameter estimates. Finally, in the regression models, the data are weighted to make the sample nationally representative.7 Logic of Analysis
There are three main analytical issues in this analysis. The first phase of analysis uses regression analysis and covariate adjustment to examine the enduring impact of incarceration on health. The second phase examines the incarceration-health relationship using methods that correct for the nonrandom nature of incarceration and examines whether the prison ‘‘selects’’ individuals who are unhealthy. Finally, the article explicitly considers whether the incarceration systemFeither through differential effects of incarceration or through differential rates of ‘‘exposure’’ to incarceration Fcontributes to racial differences in midlife health functioning. This analysis begins with regression models, which provide both a baseline to assess the effect of incarceration relative to other social processes and an incarceration parameter for comparisons 7 I thank Jay Zagorsky and Steve McClaskie, NLSY79 user services, for valuable assistance in constructing the weights used for analysis.
Massoglia
287
with later propensity models. Covariate adjustment, however (specified in Formula 1), can be problematic when attempting to make causal inferences. Broadly stated, standard regression models estimate an average effect. That is, the estimate is based upon those who have experienced the event or treatment, in this case incarceration, and is assumed to be an average effect generalizable to the entire population (Rosenbaum & Rubin 1983; Rubin & Thomas 1996; Winship & Morgan 1999). Under conditions in which the treatment is random and the population is homogeneous prior to treatment (or if differences are unrelated to either the treatment or the outcome), the ordinary least-squares estimate is efficient and unbiased and the estimate can be taken to represent the actual treatment effect. If, however, treatment is nonrandom, as is the case with incarceration, then estimates generated from regression models are likely biased. To correct for this potential bias, propensity matching estimators are utilized. The overall intent of employing this estimator is to approximate the conditions of an experiment and calculate the average treatment effect among the treated (ATT). Two characteristics of propensity matching procedures (formally presented below) have important substantive implications. First, if the matching equation adequately models the propensity for individuals to experience prison and the sample is balanced, the remaining individuals are homogeneous and differ only in whether they have experienced incarceration, given the covariates in the model. Second, as a function of the homogeneous sample, incarceration can be treated as though it occurs at random, also known as the ignorable treatment assumption (or the conditional independence assumption, in the economics literature). Given the ignorable treatment assumption and the random nature of treatment status, factors unrelated to treatment, for instance prior health (see footnote 14 and the Appendix), can be seen as occurring at random across both the treatment and control sample. Assuming the treatment state is properly modeled, in situations where the data do not have mirror measures of the dependent variable at earlier points in time, propensity models allow such factors to be treated as random and thus help alleviate concerns about sample heterogeneity. Treatment Effects and Propensity Score Models
The estimation of treatment effects is dependent upon a counterfactual. As noted by others (Winship & Morgan 1999; Morgan 2001), it is impossible for any individual to be in both the treated and untreated state (both ever incarcerated and never incarcerated). As such, the problem can be seen as one of missing data.
288
Race, Health, and Prison
Propensity score methods attempt to address this problem by using relevant background characteristics and behavioral indicators to match individuals on the basis of their likelihood to experience a treatment, in this case their likelihood of experiencing prison (Rosenbaum & Rubin 1983; Winship & Morgan 1999; Morgan 2001; Harding 2003). The propensity score, which represents the conditional probability of incarceration, can be written as (Rosenbaum & Rubin 1983): Formula 2: pðincarcerationÞ ¼ PrðTi ¼ 1jXi Þ Where Ti 5 1 if individual i has ever been incarcerated, and Xi is a vector of covariates, often including multiple interaction terms, for individual i that predict incarceration or are potential confounding variables in the association between incarceration and health.8 After the propensity scores and matched sample have been created, the analysis then utilizes nonparametric procedures to estimate the treatment effect of incarceration on later health. This analysis generates treatment effects using four different matching algorithms to pair incarcerated individuals (treated) with similar non-incarcerated individuals (controls) on the basis of their likelihood to experience incarceration. This analysis uses four different matching procedures to assess the effect of incarceration on health: neighbor matching, radius matching, kernel matching, and stratification matching. As each matching procedure has potential strengths and weaknesses, the results from all four matching estimators are presented. While variants are evident across different matching procedures, the guiding principle behind each method is to create homogeneous samples by matching individuals based on their propensity scores, allowing incarceration to be treated as though it has occurred at random; the ignorable treatment assumption (Rosenbaum & Rubin 1983; Morgan & Harding 2005). Under conditions where the treated and untreated samples have identical scores across relevant variables (the covariates), the samples are considered to be balanced (Rubin 1985; Morgan 2001; Harding 2003).9 It is important to note that the analysis specifically models prior health problems as a risk factor for exposure to incarceration. This helps address concerns of spuriousness and selection. See the Appendix for more detailed information on the matching methods and sensitivity analysis. 8 The vector of covariates can, and in this analysis does, include multiple interactions. For ease of presentation. Formula 2 presents the simplest form of the models. 9 This analysis uses Version 2.0 of the Stata programs for ATT estimation based on propensity score matching developed and discussed by Becker and Ichino (2002). In addition to calculating the treatment effect, this program specifically tests for covariate balance. In all cases, the balancing property was satisfied.
289
Massoglia
To further ensure the homogeneity of the matched sample and confidence in the estimates, all analysis is restricted to regions of common support. In practice, restricting analysis to regions of common support excludes outliers, identified on the basis of their propensity score, from analysis. Finally, no distributional assumptions are imposed in analysis, ensuring that the estimates are derived from the actual distribution of the data (Morgan 2001; Harding 2003). The potential advantages of propensity estimators over traditional regression are discussed in more detail in Joffe and Rosenbaum (1999), Morgan (2001), and Morgan and Harding (2005).
Findings Table 2 presents the results from a series of regression equations predicting general health functioning at age 40. Model 1 shows the effect of a series of background and demographic variables on health. The results are consistent with prior research; those whose family receives public assistance, those who feel that Table 2. Background, Lifestyle, Life Course, Social Location, and Incarceration Effects on Physical Health Functioning at Age 40: Regression Results Model
1
Incarceration Race (white 5 1) 0.623 Background Gender (male 5 1) 0.912 Prior health problems 4.136 Intact family 0.261 Control orientation 0.127 Parents’ education 0.218 Welfare (1 5 yes) 1.557 Drug use 0.286 Self-reported crime 0.048 Lifestyle Insurance (1 5 yes) Workout Weight Cigarette use Binge drinking Cocaine use Life course/Social gradient Urban residence Poverty status (1 5 yes) Marriage status (1 5 yes) Education level Labor force partic. (1 5 yes) Mental health Constant 51.367 R-Squared 0.05 N 5,043
2 (0.28)n
3
4
3.553 (0.52)n 2.208 (0.55)n 0.769 (0.28)n 0.239 (0.29) 0.520 (0.32)
(0.22)n 1.721 (0.48)n 3.633 (0.29) 0.489 (0.06)n 0.058 (0.04)n 0.124 (0.42)n 0.932 (0.22) 0.074 (0.03) 0.026
(0.25)n 1.997 (0.46)n 3.620 (0.29) 0.264 (0.07) 0.070 (0.04)n 0.111 n (0.41) 0.889 (0.23) 0.075 (0.03) 0.008
(0.25)n 1.562 (0.46)n 3.459 (0.29) 0.430 (0.06) 0.039 (0.04)n 0.034 n (0.41) 0.614 (0.22) 0.226 (0.03) 0.055
(0.25)n (0.46)n (0.29) (0.06) (0.03) (0.40) (0.22) (0.03)
0.915 0.598 0.028 1.789 0.155 0.437
(0.29)n 0.764 (0.07)n 0.612 (0.01)n 0.030 (0.24)n 1.760 (0.37) 0.357 (0.26) 0.430
(0.29)n 0.051 (0.07)n 0.632 (0.01)n 0.029 (0.25)n 1.099 (0.36) 0.128 (0.26) 0.455
(0.29) (0.07)n (0.01)n (0.25)n (0.36) (0.26)
0.032 1.793 0.452 0.243 1.453 0.062 54.879 (1.01) 47.996 0.11 0.15 5,043 5,043
(0.23) (0.28)n (0.23)n (0.05)n (0.26)n (0.01)n (1.41)
(.92)
54.028 (1.02) 0.09 5,043
po0.05; nonsignificant 98-00-02 cohort dummies not reported.
n
290
Race, Health, and Prison
they have less control over the direction of their lives, and those with prior health problems have poorer midlife health. Not surprisingly, prior health problems are the strongest predictor of later health. Growing up in an intact family, drug use, and self-reported crime in youth are unrelated to health at age 40. Consistent with prior research, Model 1 indicates that whites have significantly better health functioning than minorities. Model 2 both attempts to explain racial differences in health functioning with a number of lifestyle variables and establishes a baseline model for the introduction of the incarceration indicator. Again, the results are generally consistent with prior research; for instance, those who exercise more frequently and have health insurance report better health, while cigarette smoking and increased weight have detrimental effects on health. Despite the introduction of these relevant lifestyle factors, whites still have significantly better health functioning than minorities. In fact, racial differences in health are slightly stronger in magnitude after incorporating these factors (0.769). Model 3 introduces incarceration. The magnitude of the incarceration effect ( 3.553) exceeds other factors commonly used to explain health functioningFfor instance, access to insurance. This indicates that incarceration has a lasting and significant impact on physical health. Moreover, with one notable exception, the direction and significance of all other processes in Model 3 are similar to those presented in Model 2. The notable exception is race: the introduction of incarceration attenuates racial difference in health by almost 70 percent and renders racial differences nonsignificant. In addition to having a significant effect on later health functioning, this suggests that incarceration contributes to persistent racial differences in midlife health functioning. Model 4 in Table 2 presents results using the full model specification presented in Formula 1. That is, the effect of incarceration on health is estimated when controlling for demographic, background, and health indicators prior to incarceration, and lifestyle and life course processes after incarceration. This model specification tests whether the relationship between incarceration and health is explained by other contemporaneous life course processes. The results presented in Model 4 are again consistent with much prior work: smoking and increased weight are negatively associated with health, as is poverty status. Conversely, education, exercise, and mental health all positively affect physical health. In addition, those who are married and are in the labor force have increased health functioning.10 10 Given that the majority of incarcerated individuals are men, it is informative to report incarceration effects among men only. If women are removed from the analysis,
Massoglia
291
In the full models, incarceration continues to exert a significantly negative effect on health. Introducing the final block of life course variables reduces the incarceration effect approximately 40 percent, from 3.553 to 2.208.11 These results indicate that, in part and in a manner predicted by prior research, incarceration works through life course processes to affect health by lowering marriage, employment, and educational prospects. To provide additional evidence of the health-incarceration relationship, the matching methods previously described are used to further assess the relationship between incarceration and later health functioning. Given the apparent significance of incarceration in explaining racial differences in health, these methods further test the robustness and magnitude of the incarceration effect ( 2.208) reported in the final model of Table 2. All indicators used to create the matching equation are measured prior to the period of risk of incarceration, and earlier regression analysis informs the selection of variables for the propensity models. A few notes warrant discussion before moving to a consideration of the propensity models. First, in contrast to the average treatment effects estimated with regression models, the propensity estimates are derived from a sample matched on their likelihood to experience incarceration. As such, the estimates represent the average treatment effect among the treated (ATT). In addition, the matched sample is significantly more disadvantagedFboth in terms of health and other social indicatorsFthan the general population. Thus while both estimators speak to the impact of incarceration on health, the specific inferences and generalizations drawn from the estimators vary based on the analytic sample. As such, it is perhaps most informative to view the propensity models as a further, yet complementary, test of the relationship between incarceration and health. Depending on the specific matching procedure, almost all incarcerated individuals have appropriate control matches. After the matching procedure, the sample size of the treatment groups ranges from 214 to 242 (approximately 87–98 percent of the incarcerated sample, depending on the matching procedure), and the remaining individuals in the control group are homogeneous
using the model specification in Table 2, Model 4, the effect of incarceration on health is 2.691, indicating a significant relationship. 11 I thank a reviewer for the informative suggestion to assess the robustness of the finding by re-estimating the incarceration effect in Table 2, Model 4, after restricting the sample to sibling data. This helps further the homogeneity in the sample. While the size of the analytic sample decreases substantially, even with this restriction the negative effect of incarceration remains significant and large, albeit at slightly attenuated levels ( 1.88, po0.01).
292
Race, Health, and Prison
Table 3. Propensity Estimators With Multiple Matching Procedures: The Treatment Effect of Incarceration on Physical Health
Matching Estimator Nearest neighbor (237) Kernel (242) # Radius (214) Stratification (242)
Incarceration Effect 2.011n 1.659n 1.820n 1.505n
Normal Theory 95% Bounds
Empirical Distribution 95% Bounds
Lower
Upper
Lower
Upper
3.752 3.066 3.255 2.893
0.697 0.242 0.384 0.139
3.801 3.077 3.331 2.913
0.593 0.233 0.442 0.138
Notes: npo0.05. Numbers in parentheses represent treated individuals with matched controls for each estimator. All analysis restricted to regions of common support. # Radius width set to 0.001.
given the covariates in the models. In all cases the balancing property is satisfied. As reported in Table 3, incarceration significantly lowers later health functioning. While there is some variation in the magnitude of the effect based on the matching procedure, ranging from 1.505 to 2.011, the consistency of the results is striking. This demonstrates the robustness of the incarceration-health relationship even when controlling for the likelihood of experiencing incarceration with multiple matching techniques. Moreover, because nonparametric methods are utilized, the relationship is not dependent on any distributional assumptions for either the matching estimator or the incarceration effect. From the first part of the analysis, two significant findings emerge. First, regardless of the analytic procedure and controls imposed, a significant relationship between incarceration and later health is evident. Second, using multiple methods, the effect of incarceration on health is similar or greater in magnitude to many factors traditionally associated with health, such as marital and employment status. This analysis now moves to an explicit consideration of incarceration and racial differences in health. As shown in Table 2, racial differences in health are mediated by approximately 70 percent and become nonsignificant when exposure to incarceration is considered. Two plausible scenarios could explain this finding. First, minorities are significantly more likely to be incarcerated. As a function of this differential exposure, even if the effect of incarceration on later health is the same across races, the impact will be greater on the aggregate health of minorities because minorities are considerably more likely to be exposed to the detrimental health effects of incarceration. Essentially, this is a consideration of racial variation in aggregate exposure to the penal system. Alternatively, the impact of exposure to the penal system may be worse for minorities, a finding consistent with other work on the justice
Massoglia
293
Whites Minorities Disadvantaged 55 54 53 52 51 50 49 48 47 46 45 Baseline
College Grad
Employment
Prison
Figure 1: Short form 12 reported physical health values-racial differences Notes: Disadvantage is characterized by failure to graduate college, unemployment, or having been incarcerated. The baseline represents average health differences between whites and blacks. To help assess the general impact of disadvantage across these social domains, the average health score of individuals disadvantaged in each domain is graphed next to the health status of whites and blacks who are not disadvantaged in those social domains. Accounting for disadvantage in any of these social domains has a similar impact on racial health inequalities. In addition, this shows the importance of incarceration relative to education and employment in perpetuating racial health inequalities.
system (Pager 2003). This suggests an interaction between incarceration exposure and minority status. Figure 1 explores the relationship between incarceration and racial disparities in general health functioning in more detail by considering aggregate exposure to the penal system. Moreover, Figure 1 places incarceration along other systems of stratification and disadvantage. Attempts to examine racial health disparities often focus on employment and education, arguing in part that the labor market and education are systems of stratification that place minorities in a disadvantaged position relative to whites. Figure 1 locates incarceration next to these other systems of health stratification. For illustrative purposes, in addition to showing baseline health differences for whites and minorities, Figure 1 graphs physical health values (higher scores indicate better health) and shows the impact of disadvantage across three different social arenas. Those labeled as disadvantaged in Figure 1 are those without a college diploma, without a job, or those who have been incarcerated.
294
Race, Health, and Prison
Controlling for gender, Figure 1 shows significant baseline differences in health between whites and minorities. Consistent with other work, accounting for socioeconomic status such as education or employment minimizes general differences in health (Williams & Collins 1995; Robert 1999). In part, this is because minorities disproportionately carry the weight of disadvantage in these areas. For instance, while comprising only 36 percent of the overall sample, nonwhites represent more than half (52 percent) of those without jobs and approximately 77 percent of those without a college degree. Figure 1 shows that incarceration works in a manner consistent with other systems of stratification; after separating those who have been incarcerated, the health profiles of whites are similar to those of minorities. As with disadvantage in educational and occupational domains, this is because minorities disproportionately face disadvantage in the penal system. With incarceration rates five to eight times higher than those of comparable whites, the aggregate health impact of incarceration is greater for the minority community. Clearly these systems of stratification are not independent of one another. Figure 1 suggests, however, that incarceration is likely another system of inequality that produces variation in health functioning. Finally, the main effect reported in Table 2 (Model 4) and further assessed by the propensity models reported in Table 3 is decomposed. To examine a possible differential impact of incarceration, an interaction term for minorities and prison was constructed. Such a term examines if the impact of incarceration is more detrimental to the general health functioning of minorities. Consistent with other research (Schnittker & John 2007), the magnitude of the coefficient ( 0.249) does not reach conventional levels of statistical significance. The evidence then is equivocal when linking incarceration to racial disparities in health. Less clear is whether the individual-level impact of incarceration on health varies by race; the available evidence from the NLSY79 does not suggest that it does (see also Schnittker & John 2007). What is clear, however, is the strong negative effect of incarceration on health, and the disproportionate likelihood that minorities will be in the penal system. Figure 1 suggests that this disproportionate probability of incarceration has implications for racial differences in general health status.
Discussion The penal and legal system has reached the point where the pervasive use of prison as a mechanism of social control has
Massoglia
295
touched almost every aspect of contemporary American society. More than 600,000 individuals are released from prisons yearly (Mauer 2006). As these individuals are released, the burden of their problems is shifted from the penal and legal system to general society. The health problems revealed in this analysis present yet further barriers to the successful reintegration of ex-offenders. As the number of individuals released from correctional institutions increases, so too do the social, economic, and legal implications of offender reintegration. A developed literature on the consequences of incarceration explains part of the relationship between incarceration and physical health. Incarceration negatively impacts a number of factors, such as wages (Western 2002), that are associated with health functioning. Still, even after accounting for these processes, the analysis shows incarceration to be a strong predictor of health. In so doing, this article adds to an emerging literature that finds the consequences of contact with the penal system extend into areas well outside the legal system that have not been fully appreciated in other work. That is not to suggest this analysis is not without limitations. While the NLSY79 has many strengths, it was not specifically designed for an investigation of health and incarceration. As such, there are some data limitations. First, the dependent variable is measured at only one point in time. While propensity estimators help alleviate this concern through randomization, future work would be well served to collect repeated measures and incorporate models focusing on change in health status. Along similar lines, more specific health measures emphasized in other work, such as low birth weight (Johnson & Schoeni 2003, 2007), would further refine an understanding of the prison-health relationship. Future work in the area would benefit from data collection designed specifically to examine the health-incarceration relationship. In addition, both the regression and propensity models are sensitive to bias from unmeasured processes. While sensitivity analysis suggests that the problem is not a significant contaminant (see Appendix), with social science data it is virtually impossible to definitively rule out unmeasured heterogeneity. Future work might consider if the impact of incarceration on health varies by years since release. Indeed, there is rationale to suggest both that the negative effects may lessen over time and that the negative effects may worsen over time. Along similar lines, it is plausible that the incarceration spell itselfFperhaps as a function of increased access to health care, daily meals, or even removal from an abusive relationshipFprovides some immediate health benefits to some inmates. The impact of incarceration may well be conditioned by a host of socioeconomic or other contextual factors. Finally, while this
296
Race, Health, and Prison
analysis considers only the United States, there may be crossnational variation in the impact of the penal system on health. Given these many possible contingencies, this analysis should be seen as but an initial step in specifying the complexities in the penal state-health relationship. This relationship between the penal state and health is partially explained by the concept of ‘‘exposure.’’ Incarceration exposes individuals to a number of infectious diseases including HIV, hepatitis B, hepatitis C, and tuberculosis (Farmer 2002; Talvi 2003; Massoglia 2008; National Commission on Correctional Health Care 2002).12 Even short prison sentences may provide ample exposure to infectious disease and alter the potential social and medical consequences of contact with the penal system. Research on social epidemiology, stress, and the social hierarchy suggests that the relationship between health and punishment may be driven by mechanisms beyond life course processes and infectious disease; the status of ex-con itself signals a lower position on the social hierarchy. As public awareness and concerns have risen, few labels are as stigmatizing as ex-con. Felons are viewed as the ‘‘underclass’’ of society (Irwin 1985). This lower standing has implications for health (Marmot 2004). Moreover, the processes by which social standing impacts healthF the inability of individuals to exercise control over their lives and participate fully in societyFappear particularly relevant for understanding the strong association between incarceration and health. Both while inmates are incarcerated and upon their release, legislative initiatives and legal mechanisms assure that inmates face both significant public stigma and legal regulation over their lives. In explaining the relationship between incarceration and health, there is perhaps no clearer relationship than the restrictive capacity of incarceration and the justice system to prevent individuals from participating fully in society. To be sure, some of these legal restrictions are rooted in public safety, but the public good, legislative goals, and legal aims of many post-release legal sanctionsFfor instance, bans on voting and many government subsidy programs, as well recent initiatives to place identification markers on license platesFappear less clear. From Marmot’s (2004) perspective, such restrictions could have negative implications for health by restricting individuals’ societal participation and control over their life. As has been noted in other places (Mauer & Chesney-Lind 2002), at a minimum the 12 As noted, supplementary analysis indicated that ex- inmates are more than twice as likely as non-inmates to report having hepatitis, tuberculosis, and urinary tract infections. Estimates were derived using both logistic regression and propensity models. This supplementary analysis is available upon request.
Massoglia
297
cost and value of these collateral consequencesFand indeed the cost and value of the widespread use of incarcerationFshould become a greater part of the ongoing debate about the penal and legal systems in contemporary American society.
Incarceration and Racial Inequalities in Physical Health The second major thrust of this research assesses whether the penal system plays a role in average health differences between whites and nonwhites. Given the makeup of the NLSY79, this analysis speaks best to differences in health between blacks and whites.13 Attempts to explain racial disparities in health often focus on socioeconomic indicators such as education and employment. Figure 1 indicates that incarceration effects parallel those of other classic markers of socioeconomic status and suggests that the penal system merits consideration as one of the fundamental systems of stratification that contributes to racial health disparities in general health functioning. It is noteworthy that this finding is in contrast to other work on incarceration and severe health problems (Schnittker & John 2007). Similar to earlier suggestions to further examine the contingencies in the incarceration-health relationship, further research is needed to definitively examine the conditions or measures of health status where the penal state perpetrates racial differences in health. The significant contribution of the penal state to racial differences in general health functioning may not extend to all health outcomes or specific ailments. That said, a number of mechanisms help explain the importance of the penal system in explaining the racial differences in general health functioning found in this analysis. First, in the labor and marriage marketFtwo key tenets of adult healthFblacks appear disproportionately impacted by incarceration or contact with the legal system (Pager 2003; Staples 1987; Wilson 1987). Accordingly, as compared to whites, two key venues to partially overcome the detrimental health effects of incarceration are more likely to be closed for black ex-inmates. Second, as compared to whites, the application of the law is such that blacks have a much greater likelihood of contact with the penal system (Pettit & Western 2004). As a percentage of the overall population, many more blacks are incarcerated, which this analysis suggests has implications for aggregate minority general health functioning. 13 Slightly more than 5 percent of the overall sample consists of nonblack minorities. Exploratory analysis indicates that these individuals have health outcomes and social status indicators, such as education and income, consistent with blacks in the sample.
298
Race, Health, and Prison
While blacks are disproportionately incarcerated, the analysis suggests that the penal system produces significant stratification in physical health for all incarcerated individuals. The penal system has expanded so tremendously that its scope and stratifying impact appear similar to that of other systems of health stratification. Recent estimates indicate that the legal and penal systems have grown to the point where there are roughly as many felons and ex-felons as college students, and there are as many inmates and ex-inmates as men in college (Uggen et al. 2006; United States Department of Education 2003). As the footprint of the penal system takes an ever larger place in the landscape of American society, debates about the scope, goals, costs, and benefits of the penal system are needed. This research seeks to inform and contribute to those debates. To that end, it reaches two basic conclusions about the long-term consequences of incarceration. Incarceration has significant long-term effects on physical health, and incarceration plays a role in perpetuating racial inequalities in health. The effect on health appears so robust and the incarceration disparities so great, that the penal system likely warrants consideration alongside other systems of health stratification.
Appendix: Matching Models and Sensitivity Analysis This Appendix briefly discusses each of the four matching procedures used in the analysis, beginning with neighbor matching.14 Neighbor matching is a relatively straightforward and parsimonious way to create a matched sample for subsequent analysis. A propensity score is estimated for every individual in the sample, independent of incarceration history. The nearest neighbor matching procedure identifies the propensity score of all incarcerated individuals and then selects a non-incarcerated individual whose propensity score is closest to each incarcerated individual. In its simplest form, and using a 1–1 match, nearest neighbor matching pairs each incarcerated individual to the closest non-incarcerated unit on the basis of the estimated propensity score. Analysis is then done on the matched sample to determine the treatment effect of incarceration. In many respects, stratification matching operates similarly to neighbor matching. However, it divides the range of variation of propensity scores into strata, or intervals, such that within each strata the treated and control units have, on average, the same propensity score. The estimated treatment effect is the average 14 This section focuses on the conceptual and substantive distinction between different matching procedures. Readers interested in a statistical and mathematical presentation of the differences across matching procedures should consult Becker and Ichino (2002:5–7).
Massoglia
299
difference in health across treated and nontreated individuals in each stratum. While neighbor and stratification matching are straightforward and conceptually simple, they are not without potential problems. In particular, if no restrictions are imposed, nearest neighbor matching, by design, selects the nontreated individuals closest to the treated individuals, independent of the actual differences in propensity scores. That is, neighbor matching takes the closest match to any treated individual, even if the closest individual is, in fact, relatively dissimilar in terms of propensity score. This problem is conceptually similar to that of outliers in traditional regression models in that individuals with poor matches (meaning a significant difference between the propensity score of the two individuals) contribute the same to the calculation of the treatment effect as individuals with identical matches. Radius and kernel matching offer solutions to this potential problem. Radius matching allows the user to define the radius, or neighborhood, around each treated individual from which matches can be drawn. The user sets a radius, for example, 0.001 in this analysis, and all nontreated individuals whose propensity score is within 0.001 of a treated individual will be selected for analysis.15 Finally, kernel matching invokes a matching procedure where all treated individuals are matched with a weighted average of all untreated controls that is inversely proportional to the distance between the propensity scores of the treated and the controls. Substantively, this procedure uses weights to balance the contribution of each individual in the control group. This ensures that the contribution of an individual in the control group to the calculation of the treatment effect is based on how closely the individual is matched to an individual who has experienced incarceration. The equation used to generate the matched sample is formally presented in Formula 2. As presented in the body of the article, one key component of the equation is the vector of covariates, Xi, used to create the propensity scores. In this article, the matching equation includes measures of race, gender, whether the respondents came from an intact family, whether the respondents’ family received public assistance, high school degree completion, employment status, juvenile involvement with the justice system, self-reported drug use, acts of violence, and a series of gender and race interactions. Because 15 There is an obvious tension between the size of the radius and those included in the control group. The smaller the radius, the more homogeneous the experimental and the control group, but the greater the likelihood that treated cases will not have a match. In contrast, the larger the radius, the greater the number of matches for each treated individual, but greater heterogeneity is introduced across the matched and treated samples.
300
Race, Health, and Prison
individuals with unhealthy lifestyles may be disproportionately incarcerated, propensity equations used to generate the results in Table 3 also include prior health problems. In so doing, the estimates specifically account for and control whether individuals with health problems are more likely to be ‘‘selected’’ into incarceration.16 The inclusion of indicators of crime and involvement with the justice system has multiple functions. First, they help ensure that the treatment effect reported in Table 3 represents an effect of incarceration as opposed to an effect of criminal behavior or more fleeting contact with the justice system, such as arrest. Moreover, including factors such as arrest, drug use, and selfreported crime has another important function: it helps create a matching equation well-suited to place individuals in the treatment state of incarceration. Indicators of criminal behavior and correctional intervention are highly predictive of incarceration status. Moreover, the specific criminal behavior indicators included, violence and drug use, are particularly predictive of incarceration (Blumstein & Beck 1999). In conjunction with socio-demographic indicators, the matching equation includes many factors that prior research indicates are key predictors of incarceration. Depending on the matching algorithm, 87–98 percent of incarcerated individuals are matched with appropriate nonincarcerated controls, with a slight variation in the number of matched individuals based on the specific matching procedure employed. While left-skewed, estimated propensity scores range from 0.001 to approximately 0.7. Important to note, the range of propensity scores is consistent across both treated and nontreated groups. Even after restricting the analysis to regions of common support, the majority of incarcerated individuals are matched with appropriate controls. The radius matching procedure finds suitable matches for the lowest number of incarcerated individuals, 214. In part, this is a function of the highly restrictive conditions placed on how individuals are selected as controls. The radius width is set to 0.001, indicating that for an individual to be selected as a control, that individual’s propensity score has to be within 0.001 of an incarcerated individual. This highly restrictive criterion further ensures the homogeneity of the sample, but it comes at the expense of the number of paired individuals. The less restrictive the radius bandwidth, the more individuals matched. In subsequent analysis, with the bandwidth increased to 0.005 and 0.01, 229 and 240 16 There is no evidence that individuals with prior health problems are more likely to be incarcerated. The propensity matching equation is a logit model, and the results indicate that health problems are a nonsignificant predictor of incarceration.
Massoglia
301
individuals respectively are matched. The substantive results do not change when these different radius bandwidths are used for analysis. One potential weakness of propensity models is bias from unobservables. As noted by others (Caliendo & Kopeinig 2005; Becker & Caliendo 2007), there are a number of procedures that help assess the role of unobservables, speak to the quality of the matching procedure, and test the robustness of the treatment effect. Following the lead of Caliendo and Kopeinig (2005:23) multiple steps were undertaken to ensure the adequacy of the matching procedure. First, as noted earlier, in all cases the balancing property was tested for and satisfied. Moreover, tests for standardized bias indicated that the bias reduction was below the 3–5 percent generally accepted as sufficient (Sianesi 2004; Caliendo and Kopening 2005:15). Finally, t-tests were utilized to examine mean differences between the treated and the control group, and in all cases the tests failed to reach conventional levels of statistical significance. The results of all tests point to an adequate matched and control sample. With regard to the calculation of the treatment effect(s) shown in Table 3, one sensitivity check is to estimate the parameter using multiple matching methods (e.g., kernel, neighbor) and using bootstrapped standard errors. Both were done for all the analysis presented in Table 3. Moreover, as outlined in Rosenbaum (2002), DiPrete and Gangl (2004), and Becker and Caliendo (2007), estimates were derived using the complimentary psmatch2 method, and the Rosenbaum bounds were calculated. This sensitivity analysis specifically addresses the potential role of unobservable processes in the estimation of the treatment effects presented in Table 3. It examines the departure from an analysis free of hidden bias to empirically assess how much bias from unobservables would have to be introduced to significantly alter the inferences drawn from the parameter estimates. The results indicate that the estimates in Table 3 are not overly sensitive to potential bias from unobservables. More specifically, the odds that any matched pair differs in their probability of receiving treatment due to unobservables (g or G) would have to exceed 30 percent (G 5 1.3) to alter the substantive conclusions here. As such, the estimates appear robust. All sensitivity analyses are available upon request. In addition, as noted in footnote 11, the regression results remain significant even when restricting the analysis to the sibling data. Thus while almost no social science models can rule out the possibility of some unmeasured heterogeneity impacting the statistical results or substantive inferences, the available evidenceFgathered through a series of sensitivity analysesFsuggests that the findings presented herein are robust.
302
Race, Health, and Prison
References Anson, Ofra (1989) ‘‘Marital Status and Women’s Health Revisited: The Importance of a Proximate Adult,’’ 51 J. of Marriage and the Family 185–94. Barrett, Anne (2000) ‘‘Marital Trajectories and Mental Health,’’ 41 J. of Health and Social Behavior 451–64. Becker, Sascha, & Andrea Ichino (2002) ‘‘Estimation of Average Treatment Effects Based on Propensity Scores,’’ 2 The Stata J. 358–77. Becker, Sascha, & Marco Caliendo (2007) ‘‘Sensitivity Analysis for Average Treatment Effects,’’ 7 The Stata J. 71–83. Bird, Chloe, et al., eds. (2000) Handbook of Medical Sociology, 5th ed. Upper Saddle River, NJ: Prentice Hall. Black, Donald (1976) The Behavior of Law. New York: Academic Press. FFF (1998) The Social Structure of Right and Wrong, 2d ed. New York: Academic Press. Blumstein, Alfred, & Allen J. Beck (1999) ‘‘Population Growth in U.S. Prisons. 1980– 1996,’’ in M. Tonry & J. Petersilia, eds., Crime and Justice: A Review of Research, Vol. 26, Prisons. Chicago: Univ. of Chicago Press. Bosma, Hans, et al. (1998) ‘‘Two Alternative Job Stress Models and the Risk of Coronary Heart Disease,’’ 88 American J. of Public Health 68–74. Bound, John (1991) ‘‘Self-Reported Versus Objective Measures of Health in Retirement Models,’’ 26 The J. of Human Resources 106–38. Browning, Christopher, & Kathleen A. Cagney (2002) ‘‘Neighborhood Structural Disadvantage, Collective, and Self-Rated Physical Health in an Urban Setting,’’ 43 J. of Health and Social Behavior 383–99. Caliendo, Marco, & Sabine Kopeinig (2005). ‘‘Some Practical Guidance for the Implementaion of Propensity Score Matching,’’ Journal of Economics Literature Classification System (JEL) C40 H43. Working paper. Reprinted in Microeconometric Evaluation of Labour Market Policies, Berlin and Heidelberg: Springer-Verlag, 2006. Cooper, Richelle (1993) ‘‘Health and the Social Status of Blacks in the United States,’’ 3 Annals of Epidemiology 137–44. DiPrete, Thomas, & Marcus Gangl (2004) ‘‘Assessing Bias in the Estimation of Causal Effects: Rosenbaum Bounds on Matching Estimators and Instrumental Variables with Imperfect Instruments,’’ 34 Sociological Methodology 271–310. Farmer, Paul (2002) ‘‘The House of the Dead: Tuberculosis and Incarceration,’’ in M. Mauer & M. Chesney-Lind, eds., Invisible Punishment: The Collateral Consequences of Mass Imprisonment. New York: The New Press. Ferraro, Kenneth, & Melissa M. Farmer (1996) ‘‘Double Jeopardy to Health Hypothesis for African Americans: Analysis and Critique,’’ 37 J. of Health and Social Behavior 127–43. FFF (1999) ‘‘Utility of Health Data from Social Surveys: Is There a Gold Standard for Measuring Morbidity?,’’ 64 American Sociological Rev. 303–15. Freeman, Richard (1996) ‘‘Why Do So Many Young American Men Commit Crimes and What Might We Do About It?,’’ 10 J. of Economic Perspectives 25–42. Fremont, Allen, & Chloe Bird (2000) ‘‘Social and Psychological Factors, Physiological Processes, and Physical Health,’’ in C. Bird et al., eds., Handbook of Medical Sociology, 5th ed. Upper Saddle River, NJ: Prentice Hall Publishers. Fu, Haishan, & Noreen Goldman (1996) ‘‘Incorporating Health into Models of Marriage Choice: Demographic and Sociological Perspectives,’’ 58 J. of Marriage and the Family 740–58. Gottfredson, Michael, & Travis Hirschi (1990) A General Theory of Crime. Palo Alto, CA: Stanford Univ. Press. Hagan, John (1993) ‘‘The Social Embeddedness of Crime and Unemployment,’’ 31 Criminology 465–91. FFF (1997) ‘‘Defiance and Despair: Sub-Cultural and Structural Linkages Between Delinquency and Despair in the Life Course,’’ 76 Social Forces 119–34.
Massoglia
303
Harding, David J. (2003) ‘‘Counterfactual Models of Neighborhood Effects: The Effect of Neighborhood Poverty on Dropping Out and Teenage Pregnancy,’’ 109 American J. of Sociology 676–719. Harrison, Paige, & Jennifer Karberg (2003) ‘‘Prison and Jail Inmates at Midyear 2002.’’ Bureau of Justice Statistics Bulletin, NCJ 198877. Washington, DC: U.S. Department of Justice. Hassine, Victor (2004) Life Without Parole: Living in Prison Today, 3d ed. Ed. R. Johnson & T. Bernard. Los Angeles: Roxbury Press. Hayward, Mark D., & Melonie Heron (1999) ‘‘Racial Inequality in Active Life among Adult Americans,’’ 36 Demography 77–91. Hayward, Mark D., et al. (2000) ‘‘The Significance of Socioeconomic Status in Explaining the Racial Gap in Chronic Health Conditions,’’ 65 American Sociological Rev. 910–30. Heimer, Carol (1999) ‘‘Competing Institutions: Law, Family, and Medicine in Neonatal Intensive Care,’’ 33 Law & Society Rev. 17–66. Hochstedler, Ellen (1986) ‘‘Criminal Prosecution of the Mentally Disordered,’’ 20 Law & Society Rev. 279–92. Hoyert, Donna, et al. (2006) ‘‘Deaths: Final Data for 2003,’’ 54 National Vital Statistics Reports 1–120. National Center for Health Statistics. Idler, Ellen, & Yael Benyamini (1997) ‘‘Self-Rated Health and Mortality: A Review of Twenty-Seven Community Studies,’’ 38 J. of Health and Social Behavior 21–37. Irwin, John (1985) The Jail: Managing the Underclass in American Society. Berkeley: Univ. of California Press. Jiang, Shanhe, & L.Thomas Winfree Jr. (2006) ‘‘Social Support, Gender, and Inmate Adjustment to Prison Life: Insights From a National Sample,’’ 86 The Prison J. 32–55. Joffe, Marshall M., & Paul R. Rosenbaum (1999) ‘‘Invited Commentary: Propensity Scores,’’ 150 American J. of Epidemiology 327–31. Johnson, Rucker C., & Steven Raphael (2006). The Effects of Male Incarceration on AIDS Infection Rates among African-American Women and Men.’’ National Poverty Center Working Paper Series #06-22, University of Michigan, Ann Arbor. Johnson, Rucker C., & Robert F. Schoeni (2003). ‘‘Health Dynamics and the Evolution of Health Inequality over the Life Course: The Importance of Neighborhood and Family Background.’’ Working paper, http://socrates.berkeley.edu/ruckerj/HlthnHood.pdf (accessed 23 Nov. 2006). FFF (2007). The Influence of Early Life Events on Human Capital, Health Status, and Labor Market Outcomes over the Life Course.’’ Working paper, http://socrates.berkeley.edu/ruckerj/johnson_schoeni_earlylifetext_NPC.pdf (accessed 7 June 2007). Kessler, Ronald J., et al. (1989) ‘‘Unemployment, Reemployment, and Emotional Functioning in a Community Sample,’’ 4 American Sociological Rev. 648–57. Kruttschnitt, Candace, & Rosemary Gartner (2005) Marking Time in the Golden State: Women’s Imprisonment in California. Cambridge, United Kingdom: Cambridge Univ. Press. Kurdek, Lawrence (1990) ‘‘Divorce History and Self-Reported Psychological Distress in Husbands and Wives,’’ 52 J. of Marriage and the Family 701–8. FFF (1991) ‘‘The Relations between Reported Well-Being and Divorce History, Availability of a Proximate Adult, and Gender,’’ 53 J. of Marriage and the Family 71–8. Lazarus, Richard, & Susan Folkman (1984) Stress, Appraisal, and Coping. New York: Springer. Liebling, Alison (1999) ‘‘Prison Suicide and Prisoner Coping,’’ in M. Tonry & J Petersilia, eds., Prisons. Chicago: Univ. of Chicago Press. Liebling, Alison, & Shadd Maruna, eds. (2005) The Effects of Imprisonment. Portland, OR: Willan Publishing.
304
Race, Health, and Prison
Link, Bruce, & Jo Phelan (1995) ‘‘Social Conditions as Fundamental Causes of Disease,’’ 35 J. of Health and Social Behavior 80–94. Extra Issue: Forty Years of Medical Sociology: The State of the Art and Directions for the Future. Litwak, Eugene, & Peter Messeri (1989) ‘‘Organizational Theory, Social Supports, and Mortality Rates: A Theoretical Convergence,’’ 54 American Sociological Rev. 49–66. Lopoo, Leonard, & Bruce Western (2005) ‘‘Incarceration and the Formation and Stability of Marital Unions,’’ 67 J. of Marriage and the Family 721–34. MacKenzie, Doris L., et al. (1987) ‘‘Personal Control and Prisoner Adjustment: An Empirical Test of a Proposed Model,’’ 24 J. of Research in Crime and Delinquency 49–68. Marmot, Michael (2004) Status Syndrome - How Your Social Standing Affects Your Health and Life Expectancy. London: Bloomsbury, and New York: Henry Holt. FFF, et al. (1984) ‘‘Inequalities in Death: Specific Explanations of a General Pattern,’’ 323 Lancet 1003–6. Massoglia, Michael (2008) ‘‘Incarceration as Exposure: The Prison, Infectious Disease and Other Stress-Related Illness,’’ 49 The J. of Health and Social Behavior 56–71. Mauer, Marc (2006) Race to Incarcerate, 2d ed. New York: The New Press. Mauer, Marc, & Meda Chesney-Lind, eds. (2002) Invisible Punishment: The Collateral Consequences of Mass Imprisonment. New York: The New Press. McDowell, Ian, & Claire Newell (1996) Measuring Health: A Guide to Rating Scales and Questionnaires, 2d ed. Oxford, United Kingdom: Oxford Univ. Press. McEwen, Bruce (1998) ‘‘Protective and Damaging Effects of Stress Mediators,’’ 338 New England J. of Medicine 171–9. McLeod, Jane (1991) ‘‘Childhood Parental Loss and Adult Depression,’’ 32 J. of Health and Social Behavior 205–20. Mechanic, David, & Stephen Hansell (1989) ‘‘Divorce, Family Conflict, and Adolescents’ Well-Being,’’ 30 J. of Health and Social Behavior 105–16. Morgan, Stephen L. (2001) ‘‘Counterfactuals, Causal Effect Heterogeneity, and the Catholic School Effect on Learning,’’ 74 Sociology of Education 341–74. Morgan, Steven, & David J. Harding (2005) ‘‘Matching Estimators and Causal Effects: From Stratification to Weighting to Practical Data Analysis Routines.’’ Unpublished manuscript. National Center for Health Statistics (2001) Health, United States, 2001. Hyattsville, MD: National Center for Health Statistics. National Commission on Correctional Health Care (2002) The Health Status of Soonto-be- Released Inmates. Chicago: National Commission on Correctional Health Care. Pager, Devah (2003) ‘‘The Mark of a Criminal Record,’’ 108 American J. of Sociology 937–75. Pearlin, Leonard (1989) ‘‘The Sociological Study of Stress,’’ 30 J. of Health and Social Behavior 241–56. Pettit, Becky, & Bruce Western (2004) ‘‘Mass Imprisonment and the Life Course: Race and Class Inequality in U.S. Incarceration,’’ 69 American Sociological Rev. 151–69. Porporino, Frank, & Edward Zamble (1984) ‘‘Coping with Imprisonment,’’ 26 Canadian J. of Criminology 403–21. Robert, Stephanie (1999) ‘‘Socioeconomic Position and Health: The Independent Contribution of Community Socioeconomic Context,’’ 25 Annual Rev. of Sociology 489–516. Rosenbaum, Paul R. (2002) Observational Studies. New York: Springer Verlag. Rosenbaum, Paul R., & Donald B. Rubin (1983) ‘‘The Central Role of the Propensity Score in Observational Studies for Causal Effects,’’ 70 Biometrika 41–55. Ross, Catherine E., & Chloe E. Bird (1994) ‘‘Sex Stratification and Health Lifestyle: Consequences for Men’s and Women’s Perceived Health,’’ 35 J. of Health and Social Behavior 161–78.
Massoglia
305
Ross, Catherine E., & John Mirowsky (1995) ‘‘Does Employment Affect Health?,’’ 36 J. of Health and Social Behavior 230–43. Ross, Catherine E., & Chia-Ling Wu (1996) ‘‘Education, Age, and the Cumulative Advantage in Health,’’ 37 J. of Health and Social Behavior 104–20. Ross, Catherine, et al. (1990) ‘‘The Impact of the Family on Health: The Decade in Review,’’ 52 J. of Marriage and the Family 1059–78. Rubin, Donald (1985) ‘‘Constructing a Control Group Using Multivariate Matched Sample Methods,’’ 39 American Statistician 33–8. Rubin, Donald B., & Neal Thomas (1996) ‘‘Matching Using Estimated Propensity Scores: Relating Theory to Practice,’’ 52 Biometrics 249–64. Rubio, Mercedes, & David R. Williams (2004) ‘‘The Social Dimension of Race,’’ in M. B. Beech & M. Goodman, eds., Race and Research. Washington, DC: American Public Health Association. Sampson, Robert J., & John H. Laub (1990) ‘‘Crime and Deviance over the Life Course: The Salience of Adult Social Bonds,’’ 55 American Sociological Rev 609–27. Schnittker, Jason, & Andrea John (2007) ‘‘Enduring Stigma: The Long-Term Effects of Incarceration on Health,’’ 48 The J. of Health and Social Behavior 115–30. Sianesi, Barbara (2004) ‘‘An Evaluation of the Swedish System of Active Labor Market Programs in the 1990’s,’’ 86 Rev. of Economics and Statistics 133–55. Smaje, Chris (2000) Natural Hierarchies: The Historical Sociology of Race and Class. Boston: Blackwell. Staples, Robert (1987) ‘‘Social Structure and Black Family Life,’’ 17 J. of Black Studies 267–86. Sykes, Gresham (1971) The Society of Captives: A Study of a Maximum Security Prison. Princeton, NJ: Princeton Univ. Press. Talvi, Silja (2003) ‘‘Hepatitis C,’’ in T. Herivel & P. Wright, eds., Prison Nation: The Warehousing of American Poor. New York: Taylor and Francis. Tanner, Julian, et al. (1999) ‘‘Whatever Happened to Yesterday’s Rebels? Longitudinal Effects of Teenage Delinquency on Education and Occupational Outcomes,’’ 46 Social Problems 250–74. Thoits, Peggy (1995) ‘‘Stress, Coping, and Social Support Processes: Where Are We? What Next?,’’ 35 J. of Health and Social Behavior 53–79. Extra Issue: Forty Years of Medical Sociology: The State of the Art and Directions for the Future. Toch, Hans, & Kenneth Adams (1989) The Disturbed Violent Offender. New York: VailBallou Press. Turner, Blake J. (1995) ‘‘Economic Context and the Health Effects of Unemployment,’’ 36 J. of Health and Social Behavior 213–29. Uggen, Christopher, et al. (2006) ‘‘Democracy and the Civic Reintegration of Criminal Offenders,’’ 605 The Annals of the American Academy of Political and Social Science 281–310. United States Department of Education (2003) Integrated Postsecondary Education Data System, Fall Enrollment Survey and Projections of Education Statistics to 2011. National Center for Education Statistics Washington, DC: U.S. Government Printing Office. Ventura, Stephanie, et al. (1997) ‘‘Birth and Death, United States, 1996,’’ 46 Monthly Vital Statistics Report S2. Verbrugge, Lois (1989) ‘‘The Twain Meet: Empirical Explanations of Sex Differences in Health and Mortality,’’ 30 J. of Health and Social Behavior 282–304. Ware, John E., et al. (1995) SF-12: How to Score the SF-12s Physical and Mental Health Summary Scales, 2d ed. Boston: The Health Institute, New England Medical Center. Western, Bruce (2002) ‘‘The Impact of Incarceration on Wage Mobility and Inequality,’’ 67 American Sociological Rev. 526–46. FFF (2006) Punishment and Inequality in America. New York: Russell Sage Foundation. FFF, et al. (2001) ‘‘The Labor Market Consequences of Incarceration,’’ 47 Crime & Delinquency 410–27.
306
Race, Health, and Prison
Williams, David R., & Chiquita Collins (1995) ‘‘U.S. Socioeconomic and Racial Differences in Health: Patterns and Explanations,’’ 21 Annual Rev. of Sociology 349–86. Wilson, William Julius (1987) The Truly Disadvantaged. Chicago: Univ. of Chicago Press. Winship, Christopher, & Stephen L. Morgan (1999) ‘‘The Estimation of Causal Effects from Observational Data,’’ 25 Annual Rev. of Sociology 659–706.
Michael Massoglia is an Assistant Professor of sociology and Crime, Law & Justice at The Pennsylvania State University. His research interests include crime and deviance over the life course, health stratification, and research methods. He is currently examining the relationship between criminal behavior and identity transformation during the transition to adulthood (with Christopher Uggen).
307
Prisoners’ Adjustment, Correctional Officers, and Context: The Foreground and Background of Punishment in Late Modernity
Mike Vuolo
Candace Kruttschnitt
Past research indicates that front-line criminal justice workers are the critical players in determining whether innovations in penal policy are realized. Recent attempts to understand the diversity in the application of the penal harm movement have, however, sidestepped the primary audience of these policies, the population of convicted offenders. This article uses data from two prisons to examine the effects of correctional officers on women prisoners’ adjustment to prison life. Using regression models and interview data, we find that correctional officer behavior has a profound impact on women’s ability to adjust to prison, and this effect is largely independent of the prisoners’ characteristics and the institutions in which they are housed. On a theoretical level, the findings speak to recent calls to examine the background and foreground of penal culture. On a practical level, they highlight the need to understand the environments from which women are emerging, not just the communities into which they are released.
The way that these people treat us, it’s as though emotionally and physically they feel we will never get out of prison, so they can do whatever they want to us. They forget that . . . the way they treat them will be reflect[ed] back on them, because these people, some of them will get out . . . but they don’t think about that. And it sounds like a threat, but it’s not really a threat. That’s just an old saying that goes around prison, because the way that some of these people treat the inmates, you would swear that they think we were rabid dogs or something like that. (California inmate)
P
risons changed dramatically over the course of the twentieth century in this country. Inmate populations expanded; new prisons, including high-security, ‘‘super-max’’ facilities and private prisons were built at an alarming rate; and those who were charged with working in these institutions generally have inadequate The authors would like to thank Joachim Savelsberg and Joshua Page for helpful comments and Rosemary Gartner for her role in data collection. Please direct all correspondence to Mike Vuolo, Department of Sociology, University of Minnesota, 909 Social Sciences Bldg., 267 19th Ave. S, Minneapolis, MN 55455; e-mail:
[email protected].
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
308
Prisoners’ Adjustment, Correctional Officers, and Context
training (Britton 2003; Irwin 2005). Public concern shifted to isolating and managing the ‘‘dangerous classes’’ rather than reforming or helping them to change their ways. While all of this has been well documented and heralded as hallmarks of the ‘‘penal harm movement’’ or the ‘‘new penology’’ (Feeley & Simon 1992), an equally compelling thesis points to an uneven and incomplete transformation in penal policy, one that varies across social contexts and one that reflects criminal justice actors’ differential abilities to absorb new ideologies about punishment (Kruttschnitt & Gartner 2003:59; Lynch 1998; O’Malley 1992, 1996; Garland 1985, 1990, 1997, 1999). According to Garland, ‘‘[t]his ongoing attempt to reorient criminal control institutions and revise their relations to a changing social environment [is] very much a matter of patchwork repairs and interim solutions rather than well thought-out reconstruction’’ (2001:103). Some scholars have tried to explain this ‘‘patchwork’’ effect, or the noted variation in the assimilation of the penal harm movement. Simon and Feeley (1995), acknowledging the limitations of their own conceptualization of the ‘‘new penology,’’ argue that there is a disjuncture between populist views about crime and criminality and actual penal policies. More recently, Cheliotis (2006) suggests three rationales. First, at the point of implementation of criminal justice policy, the new penology downplays the role of human agency. Second, the new penology ignores the positive potential of managerialism; and third, it misses the continuity between past and contemporary penal features. Notably, Lynch’s (2001) research on parole officers in California provides empirical support for two of these hypotheses. Policy initiatives redefined some of the roles of parole officers, including characterizing their job as being more about instrumental needs of the system rather than providing help to their clients. Lynch (1998) demonstrates that while parole officers were aware of these initiatives, they were reluctant to put them into practice. Instead of implementing the ‘‘waste manager’’ role and abandoning notions of transforming parolees’ lives, agents continued to take an individualistic and intuitive approach to their clients. Lynch (2001) also vividly shows that the conflict between different discourses, one that continues to tout the importance of rehabilitation and the other that focuses on ‘‘coercive control,’’ has important implications for how agents construct the needs and problems of their clients. Kruttschnitt and Gartner’s (2005) study of two women’s prisons in California also suggests that penal cultures can be highly resistant to change. While they found evidence of transition in both prison regimes, the older facility retained elements of the rehabilitative era. Further, one of the most visible signs of the uneven application of the postmodern penal era was women’s attitudes
Vuolo & Kruttschnitt
309
toward correctional officers (COs) and their carceral experience. The prisoners generally concurred that rules were applied consistently regardless of the institution in which they were housed; yet their subjective assessment of COs and their own feeling about their institutional lives appeared to vary between prisons. Given that COs are one of the primary actors in the penal system and the individuals who are directly responsible for implementing new penal policies, they are a critical link in understanding how variations in the penal harm movement are occurring and what they mean for ‘‘the most immediate audience for the practical rhetoric of punishment’’Fthe population of convicted offenders (Garland 1990:260). This frame resonates with the various ways in which culture is currently being deployed in the sociology of punishment as the discipline seeks to clarify the relationship between the analytic aspect of penal reforms and the routine encounters of prisoners’ lives that give meaning to these reforms (Garland 2006). We draw attention to this distinction between the background of punishment in late modernity and the foreground, or behavioral aspects. Our background consists of two women’s prisons, which symbolize very different penal eras, and our foreground is the behavioral consequences of COs on women prisoners’ subjective views of their carceral experience.
The Importance of COs As Garland (1990) argues, any external force, whether law, policy, or culture, which seeks to change penal practices must first transform the local penal culture. ‘‘The primary ‘bearers’ of this penal culture, and the agents who do most to transform cultural conceptions into penal actions, are, of course, the ‘operatives’ of the penal systemFthe personnel who staff the courts, the prisons, the probation offices, and the state departments’’ (Garland 1990:210). Arguably, COs are the most visible and important connection prisoners have to the outside world. Prisoners’ views of how COs implement their job have consequences both for themselves and for the effective running of the institution. As such, this is a site ripe for the investigation of whether the penal harm movement is in fact emerging, yet most of the research on COs is quite outdated (Crouch & Marquart 1980; Duffee 1980; Poole & Regoli 1981; Lombardo 1981; Kauffman 1988) or takes a decidedly occupational focus. Here we are referring to the research that either describes how officers think and feel about their job and whom they work with (Thomas 1972; Jacobs 1978; Lombardo 1981; Kauffman 1988; Herberts 1998; Liebling & Price 1999, 2001; Stojkovic 2003; Crawley 2002), or focuses on the effects of working in an
310
Prisoners’ Adjustment, Correctional Officers, and Context
institution, the occupational culture it instills, and the stress, conflict, anxiety, alienation, and job turnover it induces (e.g., Poole & Regoli 1981; Sharma & Sharma 1989; Walters 1991; Wright 1993; Camp 1994; Triplett et al. 1996; Lancefield et al. 1997).1 What, however, can be said about the role of COs in implementing the new penal policy and shaping prisoners’ perceptions of institutional life? Sparks and colleagues’ (1996) study of how order was maintained in two high-security prisons in England, in the context of the prison service’s introduction of ‘‘managerialism’’ and standardization, suggests one answer to this question that resonates with Lynch’s (1998) research. They argue that prisons, at least in part, are capable of rebuffing policy initiatives precisely because of the resilience of staff identities and cultures (Sparks et al. 1996:135). While the officers they studied expressed many similarities in core beliefs about what being a ‘‘good’’ or helpful prison officer meant, how this was implemented and how it translated into interactions with prisoners varied between carceral contexts. Operating within the same penal political climate, officers in one prison maintained relatively close relations with prisoners, whereas in the other, they kept their distance. These different styles of control influenced prisoners’ relations with staff and their perspectives of prison life. In the analysis that follows, we bring this research up to the present and expand it to include women prisoners’ relations to COs. Although women have been the fastest growing sector of the mass imprisonment movement in this country, they remain an understudied segment of the prison population (Kruttschnitt & Gartner 2003). This is unfortunate because women’s prisons provide a particularly important site for investigating the changing prison environment. In the heyday of prison sociology, female offenders were viewed as maladjusted, misguided, and in need of treatment (Giallombardo 1966) but not dangerous or fully responsible, agentic actors. Despite media attempts to sensationalize violent female offending, such as ‘‘gangsta girls’’ and crack moms (Gomez 1997; Miller 2001; Morrissey 2003), staff and prison administrators continue to see their charges as criminally immature and more often victimsFof abuse and bad relationshipsFthan culpable offenders (see Gartner & Kruttschnitt 2004). Women’s prospects for rehabilitation have also always been seen as greater than men’s prospects. Even with the demise of rehabilitation, and the movement for gender equity in corrections, claims about women’s distinctive life circumstances and special needs have provided considerable continuity in the treatment accorded female prisoners 1 More recently, a limited number of autobiographical accounts of prison work by COs have also been published (Dickenson 1999; Conover 2000; Papworth 2000).
Vuolo & Kruttschnitt
311
(see, e.g., Rafter 1990; Bosworth 2000). These trends, coupled with the frequently noted qualities assigned to female prisoners by COs Fthey are emotional, manipulative, and impulsive, and they pose relatively little danger (Carlen 1983, 1985, 1998; Pollock 1986; Rasche 2001; Britton 2003)Fsuggest that a study of women’s prisons can provide important insights into the questions of how daily routines and standard arrangements reflect or deflect the recent innovations in penal policy (Garland 2006). In what follows, we first offer a set of hypotheses explaining why we might expect variation in women’s responses to their prison environments contingent on, and net of, their assessment of COs’ behavior. Next we present our data, beginning with an explanation for why we believe that California is a particularly poignant example of the emergence of the new penology, followed by a description of the two prisons we studied. We then set out our methods, analyses, and findings. Finally, we discuss the implications of our results for penal theory and practice.
Hypotheses The theoretical perspective that views the modern culture of punishment as a background causal force that shapes penal routines and specific behavioral practices would predict little variation in COs’ behavior and, thus, inmate experiences (Feeley & Simon 1992; Garland 2006). If prisons in California have adopted a managerial character that emphasizes the uniform treatment of offenders within different risk categories, as opposed to the individualization of offenders (Feeley & Simon 1992), we would expect: Hypothesis 1: COs should have no effect on women’s perceptions of prison life, regardless of the institution in which they are housed.
We offer two additional hypotheses that acknowledge Garland’s (2006:438–9) recent efforts to integrate culture and conduct, or the practices of interpretation. As other scholars have shown (Lynch 1998; Sparks et al. 1996), attention to the foreground issues of cultural shifts in penality reveal the importance of specific institutional actors in determining whether and how policies are implemented. If policies are filtered through COs who have their own ‘‘corporate ethos’’ (Sparks et al. 1996:134) and, as some argue, considerable power to shape policy (Pens 1998), we would expect that: Hypothesis 2: COs have a significant impact on prisoners’ perceptions of prison life, and this will vary across prisons.
Another way of approaching the behavioral aspects of the current ‘‘culture of control’’ is to acknowledge the ‘‘embodied habits of
312
Prisoners’ Adjustment, Correctional Officers, and Context
social actors who have been ‘acculturated’ to the norms of life in a specific setting’’ (Garland 2006:433). Contemporary prison guards, like police officers, have a well-developed culture that embodies not only their attitudes and perceptions of ‘‘criminals’’ (Irwin 2005:63–6) but also the gendered nature of corrections. If COs operating in the era of hyper-control have an identified subculture and set of expectations and responses to working with female prisoners (Carlen 1998; Britton 2003), we would expect: Hypothesis 3: Women’s perceptions of their carceral lives will be impacted in a similar fashion by COs regardless of where they are imprisoned.
Data The Case of California
California’s penal policy changed dramatically in the late 1970s as the rehabilitative era came to a close with the passage of the Uniform Determinate Sentencing Act (1976). The declining crime rates over the subsequent decade did little to dampen voters’ enthusiasm for law-and-order issues as Californians set into motion a series of bonding bills that fueled the largest prison building initiative in history. Despite the construction of nine new prisons between 1984 and 1989, public concern about crime continued unabated, resulting in the passage of hundreds of bills that created new crimes and lengthened sentences on existing crimes. Faced with detaining and managing thousands of convicted offenders, the state approved a $450 million prison bond to construct 11 additional prisonsFincluding two for women offendersFthat opened between 1991 and 1997, further fortifying what Simon (2000) calls the ‘‘era of hyper-incarceration’’ (2000:288). The California Correctional Peace Officer’s Association (CCPOA) grew in size, wealth, and power alongside these developments. In 1980, it had only about 5,600 members, but over the course of the next two decades, the CCPOA joined with the Youth Authority and parole officers. In addition to meeting the needs of the surge in prison building, these combined forces produced a sixfold increase in union membership by 2002, and what is commonly referred to as the ‘‘most powerful and influential lobbying group in the state’’ (Pens 1998:135–6). Their contribution of $101,000 to get Proposition 184, the ‘‘Three Strikes’’ initiative (1994), on the ballot suggests that this reference to their relative political power is more than just rhetoric (Pens 1998; Biewen 2002). Interestingly, then, the California case suggests that COs may not be just the carriers of penal policy but, in part, the architects of it.
Vuolo & Kruttschnitt
313
The California Institution for Women and Valley State Prison for Women
The two prisons in our study mirror the distinctive shifts that occurred in California’s penal policy over the course of the twentieth century. The California Institution for Women (CIW) is the oldest prison for women in the state, opening its doors to female offenders in 1952, during the height of the rehabilitative era. Today, despite the addition of a fence and guard towers, it still retains the campus-style architecture and atmosphere of this era in corrections. The prison was designed to hold 900 women, but when we conducted our research it housed almost 1,800; all women were double-bunked in cells designed to hold only one prisoner. Reflecting both the growth in the inmate population and the CO union, custody and support staff services grew from 200 in the 1960s to 600 in the mid-1990s. While this meant that COs were more varied in their prior work experience and training than they had been in the past, a number of the correctional staff had spent their entire careers at CIW. By contrast, Valley State Prison for Women (VSPW), our second research site, was opened in 1995, and it reflects the iconography of the new penal era with a modular design and extensive perimeter security. Despite being a recent addition to the California Department of Correction’s prison portfolio, VSPW also suffered from extreme overcrowding. At the time we conducted our research, it was the largest prison for women in the world; the prison held close to 3,500 women, and cells that were supposed to house four women contained eight. While VSPW had a slightly higher ratio of prisoners to staff than CIW (roughly 4:1 vs. 3:1), the COs at VSPW tended to be younger and more recent graduates of the state’s training academy compared to COs at CIW. These structural differences also reflect differences in the culture of the two institutions, but as noted elsewhere, these differences were a matter of degree rather than kind (Kruttschnitt & Gartner 2005). Prisoners’ perceptions of staff and other inmates, while generally distrustful, were patterned by the distinctions in these two carceral experiences. At VSPW, the prison that best captures the elements of the new penology, women were particularly disaffected and isolated despite conditions of extreme overcrowding (Kruttschnitt & Gartner 2005). At CIW, an institution that retains elements of the maternal-rehabilitative regime of the past, women’s reactions to the prison were more likely to reflect their individual experiences. However, even with these institutional differences, it is important to remember the larger underlying similarities of the life of these two institutions working within the California Department of Corrections (CDC) at the end of the twentieth century. Core
314
Prisoners’ Adjustment, Correctional Officers, and Context
features of women’s daily livesFwearing uniforms, eating prison food, programming,2 being subject to multiple counts throughout the day, and having telephone calls and mail monitoredFwere governed by Title 15 of the California Code of Regulations.
Data Collection
Interviews with 73 women prisoners (35 at CIW and 38 at VSPW) formed the basis for the development of a survey that was completed by 1,821 prisoners in two women’s prisons between 1995 and 1998. Interviewed prisoners were selected at random from lists of the entire prison populations that were dichotomized on length of time served.3 Administering the prison surveys required a method that was tailored somewhat to the individual needs of each prison. As documented elsewhere (Kruttschnitt & Gartner 2005: Chapter 3), this produced very different response rates. At CIW, where we were allowed to individually distribute and collect the surveys, and where the women were locked in to complete the surveys, we had a relatively high response rate (72 percent). At VSPW, by contrast, the correctional staff oversaw the administration of the surveys after we had left the institution for the day; here we had a notably lower response rate (37 percent).4 Nevertheless, the women completing the surveys were generally representative of their respective prison populations (see Kruttschnitt & Gartner 2005:56). The survey respondents in both prisons slightly overrepresent women convicted of person offenses and underrepresent women convicted of drug law violations. In addition, these data slightly underrepresent African Americans and overrepresent women of ‘‘other’’ racial categories,5 relative to their actual representation in the prison populations from which they were drawn. The overrepresentation of women in the ‘‘other’’ racial category is due to women self-identifying multiple racial origins on the survey as opposed to adopting the racial category the CDC assigns them. 2 Programming refers to prisoners’ daily activities. Prisoners could earn credit for ‘‘good time’’ if they stayed actively engaged in the prisons’ educational classes, work assignments, and other related offerings. 3 We wanted to ensure that we talked with women who were both ‘‘old-timers’’ (those women who had served five or more years on their current sentence) and recently admitted to prison (those who were admitted on their current sentence within the last six months). 4 See Kruttschnitt and Gartner (2005:54–5) for an explication of the possible reasons for these different response rates. 5 The survey data category other includes Native Americans, Asians, and women who only indicated that they are of mixed racial origin.
Vuolo & Kruttschnitt
315
Methods Our outcome variable was a scale that measures how difficult it is for inmates to adjust to various aspects of their prison life: rules, other inmates, lack of privacy, absence of home and family, lack of outside social life, food, medical care, lack of programs,6 and overcrowding. While we cannot determine these prisoners’ actual levels of adjustment to prison life, their perceptions about any difficulties they have in these particular domains have been, and remain, important indicators of adjustment (Adams 1992:284–5; Kruttschnitt & Gartner 2003).7 Further, as noted previously, adjustment to the carceral experience is an important but understudied aspect of the new penology, and we suspect that COs may be a critical component of this process (Rhodes 2004). Each question asked how difficult it was to adjust to a particular aspect of prison life, with the possible responses being ‘‘not at all difficult,’’ ‘‘a little difficult,’’ ‘‘difficult,’’ ‘‘very difficult,’’ and ‘‘extremely difficult,’’ with the responses being valued from 1 to 5, where 1 represents ‘‘not at all difficult’’ and 5 represents ‘‘extremely difficult.’’ Thus higher values on the scale represent increased difficulty adjusting to prison life. Respondents were only included in the scale if they responded to all nine items. The scale had a range of 9 to 45 and a mean value of 31.54. Overall, this adjustment scale demonstrated moderate reliability (a 5 0.77; DeVellis 2003).8 The descriptive statistics for this response variable and all variables in our analysis are shown in Table 1. The analysis includes two types of variables consistent with the empirical literature on female inmate adjustment: background characteristics and criminal justice experiences. In the case of the former, studies show that that women who are young, are non6 Examples include vocational, substance use, anger management, and parenting programs. 7 Our focus is not on the concept of prisonization or misconduct, which has an unknown relationship with adjustment (Adams 1992), although we do control for the effect of these variables in our analyses. In assessing inmates’ adjustment, ideally we would also want to know their mental state prior to entering prison. While we did not have access to this information, we attempted to address this issue by controlling for their self-reported pre-institutional mental health in our models. 8
A reviewer questioned whether perceived self-adjustment to prison is a valid measure of overall perceptions of prison life. Our measure of adjustment taps virtually all the critical aspects of prison life (the rules, other inmates, food, medical care, programs, privacy, etc). Taken together, these items reflect what is both physically and emotionally critical in living in an institutional setting. Further, to the extent that they constitute a valid indicator of adjustment, we would expect that they would have some relationship to another measure in our survey that reflects how bad doing time has been relative to what the prisoners expected. These measures were correlated in the expected direction, such that those who had difficulty adjusting also found prison worse than they expected (r 5 0.380, po0.001).
316
Prisoners’ Adjustment, Correctional Officers, and Context
Table 1. Descriptive Statistics and Tests for Differences and Associations between Prisons Variable Background characteristics Age Ethnicity: white Ethnicity: black Ethnicity: Latino Ethnicity: other Child under 18 Education Had job Received welfare Homeless at arrest Drug abuse Prior mental health prescription Criminal justice experience Friend been to prison Years on priors (Log) Number of prisons (Log) Months on current sentence (Log) Life sentence v115s (Log) Custody level: 1 Custody level: 2 Custody level: 3 Custody level: 4 Custody level: Don’t know Offense: violent Offense: drug Offense: other Used drug in prison No control over day-to-day life Have close friends in prison Homosexual activity Prison CIW VSPW CO experiences CO helped COs treat work as just a job COs go by the rulebook New COs write up everything Dependent variable Adjusting difficulty scale
Mean for VSPW (St. Dev.)
Mean for CIW (St. Dev.)
Overall Mean (St. Dev.)
34.34 (6.49) 0.42 0.26 0.16 0.16 0.72 5.01 (1.04) 0.37 0.74 0.07 0.80 0.22
29.61 (7.31) 0.45 0.28 0.14 0.13 0.68 5.02 (1.03) 0.36 0.71 0.07 0.78 0.21
35.02n (7.80) 0.44 0.27 0.15 0.15 0.70 5.01 (1.03) 0.36 0.73 0.07 0.79 0.22
0.78 1.02 (0.88) 0.65 (0.70) 2.67 (0.87) 0.17 0.40 (0.70) 0.46 0.23 0.16 0.11 0.04 0.35 0.29 0.37 0.22 0.62 0.89 0.38
0.73 0.95 (0.79) 0.65 (0.66) 2.89 (1.26) 0.12 0.42 (0.72) 0.56 0.16 0.08 0.08 0.12 0.32 0.34 0.34 0.18 0.45 0.89 0.40
0.75 0.99 (0.84) 0.65 (0.68) 2.78n (1.07) 0.15n 0.41 (0.71) # 0.51 # 0.19 # 0.12 # 0.10 0.08# 0.31 0.36 0.33 0.20 0.54n 0.89 0.39
F F
F F
0.47 0.53
0.40 0.85 0.43 0.92
0.59 0.78 0.52 0.91
0.49n 0.82n 0.47n 0.91
33.24 (6.49)
29.61 (7.31)
31.54n (7.12)
n
po0.05 for t-test (continuous and dichotomous variables) between prisons. po0.05 for w2-test of independence (categorical variables) between prisons.
#
white, come from an urban background, are single with no children, have prior institutional experience, and have been convicted of a violent crime or a drug crime tend to score higher on traditional indicators of male adaptation, such as prisonization, opposition to staff, and misbehavior (Jensen & Jones 1976; Alpert et al.
Vuolo & Kruttschnitt
317
1977; Jensen 1977; Faily & Roundtree 1979; Zingraff & Zingraff 1980; Kruttschnitt 1981; Mandaraka-Sheppard 1986; Bondeson 1989). In addition, many qualitative studies of, and government reports on, female inmates draw attention to the role that women’s economic marginality and their substance abuse and mental health problems play in their adjustment to prison and life thereafter (Kruttschnitt & Gartner 2003). Besides these background variables, scholars also draw attention to the effects of sentence length, time served, and time left to serve on women’s adjustment to prison. The effects of these variables on carceral adjustment are, however, somewhat inconsistent (Tittle 1969; Jensen & Jones 1976; Alpert et al. 1977; Kruttschnitt 1981; Mawby 1982; Mandaraka-Sheppard 1986; Bondeson 1989; Craddock 1996). The perception that one has relatively little control over the prison environment has also been shown to be a recurrent obstacle to successful adjustment to prison (Ruback & Carr 1984; Ruback et al. 1986). With a continuous outcome variable, we used linear regression models to predict inmate adjustment. We only included variables in our models that had a significant bivariate relationship with inmate adjustment according to preliminary linear regression models (not shown). The covariates that had significant bivariate relationships with adjustment were operationalized as follows.9 Economic marginality was measured with a dummy variable for having a job at arrest, having received welfare, and having been homeless at the time of arrest, as well as a seven-item ordinal measure of education. Family background was measured with a dummy variable indicating whether a woman had a child younger than 18. Histories of substance abuse and mental health problems were measured using dummy variables for drug abuse and whether the respondent had received any prescriptions for mental health problems prior to incarceration. Finally, we included as controls age and race. Race was measured with four categories: white, black, Latino, and other. In terms of criminal justice experience, we first included variables indicative of the women’s prior carceral experiences. These measures included the number of prisons they had been housed in and the total number of years they had been incarcerated on prior commitments. Sentence length and time served are clearly important according to the literature, but given the proportion of women 9 Several variables did not have significant bivariate relationships with adjustment. These variables were marital status, alcohol abuse, U.S. nativity, having lived in a treatment facility, whether they have self-harmed prior to their sentence, whether any family members have been to prison, whether they were incarcerated as a minor, whether they have participated in any programs, whether they felt they were ‘‘in the mix,’’ and whether they believe they have a ‘‘prison family.’’ These predictors were also checked for interactions with prison and CO behavior.
318
Prisoners’ Adjustment, Correctional Officers, and Context
serving life sentences in our sample (14 percent), quantifying this concept is difficult. In an attempt to get at both concepts, we included a measure for time served on current sentence and a dummy variable for a life sentence. We also included three dummy variables for offense of convictionFviolent, drug, or other offense Fwith violent offenses serving as the baseline category. Several variables also measured a woman’s involvement in prison culture: the number of disciplinary actions she had, whether she had used drugs in prison, whether she had any close friends in prison, whether she had been involved in any homosexual activity, and her perception of the amount of control she had over her daily life.10 Because prison adjustment may be affected by having close friends incarcerated together, we also included such an indicator in our analyses. Finally, we used a five-category measure for custody level (levels 1 through 4 and a category for ‘‘don’t know’’). A dummy variable indicated the prison where the respondent was incarcerated, with VSPW as the baseline. We included four indicators of COs’ behavior. These measures assessed whether the COs take a strictly security-driven, managerial approach to their job, indicative of the new penology, or whether they maintain a more individualistic approach toward inmates. First, the survey asked respondents to indicate if COs and staff had ever been helpful in any of the following respects: ‘‘listened to my problems,’’ ‘‘gave good advice,’’ ‘‘increased my self-esteem,’’ ‘‘treated me with respect,’’ and ‘‘kept me out of trouble.’’’ If the prisoner responded yes to any of the above questions, we assumed that the COs provided some tangible assistance to the prisoner. Thus the variable ‘‘CO helped’’ was created by giving respondents a score of 1 if they responded yes to any of these items and a score of 0 if they indicated they received no help on each of these items.11 As shown in Table 1, almost 50 percent of the respondents indicated they received at least some support from COs. The second and third measures asked inmates if they agreed that COs treated their work as ‘‘just a job’’ (82 percent agreed) and if COs usually ‘‘go by the rulebook’’ (47 percent agreed). Finally, women were asked how annoying they felt it was that new and inexperienced COs wrote up every infraction. Ninety-one percent of inmates found this aspect of prison annoying.12 10 Number of years on priors, number of prisons and disciplinary actions, and months served on current sentence were all skewed. As a result, these variables were logged in the analyses. 11 We also ran the models with this variable as a simple additive scale. While the results were similar, we chose the indicator variable due to the skewed nature of the scale. 12 As with the ‘‘CO helped’’ variable, we chose to dichotomize these measures of COs because they were skewed and it is unclear if one-unit increases across response categories represent equal values.
Vuolo & Kruttschnitt
319
With the exception of new COs writing up every infraction, the two prisons differed significantly on the measures of CO behavior in ways consistent with their histories and cultures, according to t-tests as shown in Table 1 (po0.001). At CIW, the oldest prison for women, roughly 61 percent of respondents indicated that COs helped them in some way, while only 39 percent of the prisoners at VSPW, the prison that grew up in the new penal era, acknowledged receiving help from COs. These prison-based discrepancies were somewhat less obvious in the case of whether prisoners thought that the COs treated their work as ‘‘just a job’’: at CIW, 77 percent of the inmates agreed with this statement, whereas at VSPW 86 percent agreed. Finally, in terms of perceived fairness of COs’ actions, 54 percent of the women at CIW agreed that COs went by the rulebook, while only 45 percent felt this way at VSPW. According to analysis of variance models, these indicators of CO behavior had a significant effect on women prisoners’ ability to adjust to carceral life (po0.001). Inmates who received no help from COs, who believed that COs viewed their work as just a job, who thought COs did not go by the rulebook, and who believed that new COs wrote up everything had significantly more difficulty adjusting to prison.13 In regression models, we also explored interaction effects between CO behavior and the offenders’ characteristics and between offenders’ characteristics and the prison. While the prior literature points to the importance of the prison context in shaping prison officers’ attitudes toward their job (Sparks et al. 1996), we also anticipated that women offenders’ life experiences are important determinants of their reactions and responses to COs (see Kruttschnitt et al. 2000). After these variables were entered into nested models, we retained 62 percent of the sample due to missing values. This subset of the sample was generally very similar to the entire sample on all our predictors and the response variable. Finally, we note that we supplemented our statistical findings with interview data when the information provided by the prisoners shed light on the empirical results.
Findings The results of the linear regression models are shown in Table 2. Background characteristics, criminal justice experience, prison, 13 In addition to CO behavior influencing perceptions of carceral experiences, another interpretation of this relationship could be that COs provide preferential treatment to inmates who are already better adjusted. Other recent research supports the former interpretation. This research found that CO help had no effect on measures of mental health among the same sample of California inmates (Kruttschnitt & Vuolo 2007), indicating that COs do not specifically target better- (or worse-) adjusted inmates for help.
(Constant) Age Ethnicity: black vs. white Ethnicity: Latino vs. white Ethnicity: other vs. white Child under 18 Education Had job Received welfare Homeless at arrest Drug abuse Prior mental health prescription Friend been to prison Years on priors (Log) Number of prisons (Log) Months on current sentence (Log) Life sentence v115s (Log) Custody level: 2 vs. 1 Custody level: 3 vs. 1 Custody level: 4 vs. 1 Custody level: DK vs. 1 Offense: drug vs. violent Offense: other vs. violent
s.e.
(1.713) (0.030) (0.540) (0.668) (0.657) (0.543) (0.224) (0.477) (0.546) (0.868) (0.570) (0.525)
B
27.763nnn 0.051 0.934 2.335nnn 1.694nn 1.090n 0.530n 1.194n 0.228 1.445 1.394n 0.797
Model 1 s.e. (2.026) (0.031) (0.523) (0.632) (0.624) (0.514) (0.213) (0.449) (0.519) (0.818) (0.595) (0.494) (0.530) (0.344) (0.417) (0.265) (0.790) (0.360) (0.545) (0.687) (0.794) (0.779) (0.768) (0.757)
Model 2 23.263nnn 0.013 1.017 2.173nnn 1.341n 0.658 0.573nn 1.109n 0.113 0.596 0.526 0.758 0.130 0.133 0.130 1.240nnn 0.305 0.762n 0.475 0.151 0.289 1.837n 0.732 1.108
B
Table 2. Linear Regression of Inmate Adjustment Difficulty s.e. (1.978) (0.030) (0.505) (0.611) (0.604) (0.497) (0.206) (0.434) (0.502) (0.791) (0.575) (0.478) (0.513) (0.333) (0.404) (0.257) (0.768) (0.349) (0.530) (0.673) (0.774) (0.756) (0.746) (0.735)
Model 3 25.646nnn 0.025 0.959 2.037nnn 1.129 0.524 0.491n 1.039n 0.074 0.599 0.442 0.745 0.095 0.032 0.324 1.404nnn 0.378 0.864n 0.404 1.096 1.102 1.262 1.355 1.667n
B
s.e. (2.047) (0.030) (0.490) (0.596) (0.586) (0.481) (0.201) (0.422) (0.488) (0.768) (0.558) (0.464) (0.498) (0.323) (0.392) (0.251) (0.749) (0.341) (0.516) (0.653) (0.750) (0.736) (0.726) (0.716)
Model 4 24.269nnn 0.033 0.950 1.563nn 1.094 0.526 0.375 0.829n 0.216 0.272 0.674 0.684 0.027 0.048 0.389 1.319nnn 0.075 0.778n 0.349 1.037 1.060 0.665 1.380 1.389
B
s.e. (3.696) (0.029) (0.487) (0.592) (0.581) (0.477) (0.649) (0.419) (0.484) (0.760) (0.554) (0.461) (0.496) (0.320) (0.389) (0.249) (0.743) (0.339) (0.684) (0.860) (0.930) (1.117) (0.723) (0.712)
Model 5 20.353nnn 0.030 1.043n 1.427n 1.163n 0.560 1.239 0.872n 0.249 0.229 0.481 0.708 0.113 0.053 0.345 1.353nnn 0.129 0.817n 1.471n 1.913n 2.975nnn 0.064 1.397 1.359
B
320 Prisoners’ Adjustment, Correctional Officers, and Context
nnn
po0.001;
nn
po0.01; po0.05.
n
Used drug in prison No control over day-to-day life Have close friends in prison Homosexual activity Prison: CIW vs. VSPW CO helped COs treat work as just a job COs go by the rulebook New COs write up everything CO helped n Education CO write up n Education CO rulebook n Custody level 2 CO rulebook n Custody level 3 CO rulebook n Custody level 4 CO rulebook n Custody level DK R2
0.062
0.199
0.794 3.665nnn 1.147 0.169
(0.575) (0.409) (0.657) (0.486)
0.252
0.520 3.076nnn 1.221 0.098 3.511nnn
(0.557) (0.401) (0.635) (0.470) (0.414)
0.300
0.394 2.460nnn 1.044 0.193 3.086nnn 1.858nnn 1.570nn 1.139nn 2.461nnn
(0.540) (0.398) (0.619) (0.459) (0.406) (0.417) (0.523) (0.391) (0.692)
0.444 2.493nnn 1.022 0.187 2.976nnn 5.827nn 1.494nn 2.176nnn 9.451nn 0.785n 1.404n 2.355n 1.721 4.585nnn 0.837 0.319
(0.536) (0.396) (0.613) (0.455) (0.403) (1.890) (0.521) (0.533) (3.232) (0.368) (0.628) (1.008) (1.219) (1.340) (1.452)
Vuolo & Kruttschnitt
321
322
Prisoners’ Adjustment, Correctional Officers, and Context
perceptions of CO behavior, and significant interactions are sequentially added, producing five models. The addition of each set of variables improves the fit of the models. Before discussing specific effects, we examine mediating effects across our nested models.14 Adding criminal justice experience measures in Model 2 mediates the effect of past drug abuse and having a child under 18. In Model 3, we see that when the prison is included in the analyses, being white, as opposed to having a mixed racial heritage or being of Asian descent, and having a minimum custody level or being unsure of one’s custody level have no effect on adjustment. Finally, we also find that our measures of CO behavior (added in Model 4) mediate the effects of conviction for a violent offense and conviction for a nonviolent or nondrug offense on adjustment. These findings suggest that many personal attributes and experiences thought to affect prison adjustment are, in reality, linked to the type of prison in which a woman is housed and her interactions with COs. The importance of these two covariates is further underscored by examining the amount of variance we can explain in prison adjustment across these different models. Inmates’ demographic and background characteristics (Model 1) and their criminal justice experiences (Model 2) only explain, respectively, 6 and 20 percent of the variance in prison adjustment. Yet our final model, which includes indicators of where the inmates are incarcerated and their perceptions of CO behavior, explains almost onethird (32 percent) of the variance in adjustment. The first significant background characteristic is ethnicity. According to Model 5, blacks, Latinos, and ‘‘other’’ ethnicities all have significantly less difficulty adjusting to prison than whites. While it might be tempting to ascribe this racial difference in adjustment to differences in prior experience with the criminal justice system, our analysis takes account of this potential distinction between white inmates and inmates of color by controlling for such experiences. A more likely explanation is white women’s simple lack of knowledge about, and familiarity with, minority cultures and the discomfort they may feel in becoming the minority racial group in this context (see, e.g., Anderson 1990). One white prisoner described it to us this way: They are all black. I don’t understand; white women are the minority here. Although in . . . society [it] seems like, you know, we’re not the minority but in here we are. And we have little 14 Several of the variables in our models were highly correlated. Therefore, we ran the models removing highly collinear terms one at a time; the magnitude of the coefficients of the remaining terms were similar, and no new terms became significant. In addition, an analysis of residuals showed that this model satisfies assumptions of normality and constant variance.
Vuolo & Kruttschnitt
323
consideration. We are not allowed to form organizations in here, but every ethnic in here has an organization for their ethnic [group] but whites. I understand what their method of thinking is, to a degree, but I feel my limitations there, as being white. So I feel that I have to be careful how I walk . . . . and that’s where I feel my threat is . . . It’s just where the state is at and I feel threatened by that.
The only other background characteristic that has a significant effect on adjustment is having employment at the time of arrest: those who had a job when they were arrested have more difficulty adjusting to prison life than inmates who were unemployed. While this effect could be tapping social class differences in adjustment (with those working having more income), it may also reflect the lack of work experience some women had prior to coming to prison and the clear benefits they feel they have gained from their prison jobs. Consider, for example, the following prisoner’s description of how her prison job shaped her views of her prison life: Yeah, you know the good thing about it, it gives me structure. I’m working; OK, I get up at 5:30 every morning. I work from 7 to 3 every day, five days a week, plus I go to classes twice a week, and I have my schedule . . . You’re on the streets before you came in here, you were just on the streets running amuck, OK. This gives you a little bit of substance and [a] little bit of stability. Well maybe if I can do it in here, I’m sure I can do it out there.
Women’s prior and current criminal justice experiences represent the remaining main effects. Both the number of months served on the current sentence and the number of disciplinary actions received impact prison adjustment. As time served increases and as the number of rule infractions increase, women’s adjustment to prison becomes more strained. Other carceral experiences also shape women’s ability to adjust to prison life. Relative to those who feel that they have some control over their daily lives, those who believe they have no control have more difficulty adjusting to prison. Two inmates at CIW expressed it to us this way: the first described, with some pride, the control she feels and the influence it has on her interactions with COs and other prisoners. I believe that I have mastered my environment, for lack of a better term. Sometimes I’m amazed at how outrageous I get, because I will walk up to people and say the darndest things. On the other hand, I usually know when it’s all right to do that. I taught myself that . . . . I am so used to people treating me like a real person, and I’m not, I guess I’m not used to being treated like an infant. I’m not used to being hassled, and all the time that I’ve been here, cops have always said to me, ‘‘You don’t act like you’re an inmate.’’ Inmates have always said, ‘‘You don’t act like an inmate.’’ To this day, people will say to me, they’ll ask me ‘‘Are
324
Prisoners’ Adjustment, Correctional Officers, and Context
you an inmate?’’ . . . And somehow or another, I don’t have the consciousness that I’m an inmate. I know I am. I’m proud of it. Which is strange, you know, but I am, because I survived it.
By contrast, the second prisoner expressed considerable frustration at not being able to do even the simplest daily tasks without acceding to someone else’s authority. OK, my rights as an individual; I don’t have that here because I’m decided upon. What time I have to get up in the morning; what time I have to program; what time I eat; what time I have to shower; what time I have to be locked back in my room, you know. It’s a constant thing every day, you know, and . . . it’s hard adjusting to that. It took me a long time to get used to that, you know, somebody telling me, you know, you got to go lock in at this time, and you can’t wear this. I’ve never been in [a place] where so many individuals had control over me. You know, I’ve always been in control of my own self; you know I made my own decisions. Now I have to make decisions according to these people who’s over me [sic] as far as authority figures, you know, and you have to watch what you say and how you say it, you know, because the least little thing out of context; they’ll write you up about it.
Finally, the prison itself also has a significant effect on adjustment. All else remaining constant, inmates at CIW have an easier time adjusting to prison life than inmates at the more custodial VSPW. This effect is not surprising in light of our interviews. As two women at CIW described: This place reminds me of a college campus at times, when I can drift far enough . . . . When my mind will let me drift far enough away, I can sit out there in the grass as if I’m sitting out at the park. The minute the guard blows the horn it’s all over. It’s such a relief to be able to walk around, and you almost feel like, you almost feel like you have freedom again in this place because of the campus atmosphere and . . . they put all the longtermers in close custody. You have freedom during the day, you can work and whatever, but at four o’clock every day, you’re locked back in your unit. You still can go around the unit, but . . . if you compare it to RC [reception center] and jail, there’s a lot of freedom in this place.
Since our other predictors of adjustment could be conditioned by the prison context and there is evidence of bivariate relationships as shown by the t-test and w2-test results in Table 1, we ran interactions with prison. Models were run with just the main and interaction effects for each variable. Those that were significant were subsequently added to the full model one at a time. No interaction with prison maintained its significance when added to the full model. Therefore, the effect of demographic and background
Vuolo & Kruttschnitt
325
characteristics and criminal justice experiences on adjustment is not conditioned by where the inmate is imprisoned. The prisoners’ perceptions of CO behavior represent the final set of variables. These variables are all statistically significant and indicate that CO behavior has a strong effect on women prisoners’ adjustment. From Model 4, we see that those who receive no help from COs, who believe that COs treat their work as if it is just a job, who feel that COs do not go by the rulebook, and those who believe that new COs write up everything have the most difficult time adjusting. One inmate described how important it is for COs to go by the book, as it sets the tone for her living unit. If your housing officers are fair, by the book, and you know this here, that unit is more relaxed than an officer that comes in and throws his power around, the control . . . . You can tell the big difference [in] those inmates of that unit.
Other inmates reacted to how some COs do not help and do not care about the prisoners as individuals, but rather view their role as just a job for them: They don’t holler at us. Like a lot of the staff that’s in other units, I’ve noticed they talk to the women’s [sic] with no respect. They demand all respect, and they yell a lot and . . . [it’s] always no’s and they don’t care. Like if you have a docket for instance, hey I have a docket at nine o’clock. Oh well, you just wait until I get there you know? You do have some officers that because they have a badge they’re . . . they have the attitude . . . they don’t treat [you] as human, you know? You’re just another inmate. You’re a piece of dirt. Then you have those that have been involved in the CDC system for years that work with men, that worked with women that give you respect if . . . if it’s given to them. They don’t um . . . they don’t push the power of authority on you.
We also explored the possibility that CO behavior would interact with women’s backgrounds and prior experiences. We found that three interactions with CO behavior are significant when added to the full model, and they remain significant when all interactions are in one model (Model 5). We found interactions between going by the rulebook and custody level, COs’ help and an inmate’s education level, and COs writing everything up and education level. Since these interaction effects can be difficult to interpret based only on the coefficients in Table 2, we include figures of the three interactions and discuss them in this context. These figures show the predicted value for these interaction effects holding the other predictors constant at their respective means. Figure 1 shows the interactions between custody level and whether the inmate agrees that COs go by the rulebook. There is
326
Prisoners’ Adjustment, Correctional Officers, and Context
33.5 disagree by rulebook
33.0
agree by rulebook
Adjustment Difficulty Scale
32.5 32.0 31.5 31.0 30.5 30.0 29.5 29.0 28.5 28.0 custody 1
custody 2
custody 3
custody 4
custody dk
Figure 1. Interaction between COs Go by the Rulebook and Custody Level.
relatively little effect of whether COs go by the rulebook in the middle two custody levels. For women in minimum custody (and for those who do not know their custody level15), consistency in the application of rules appears to make life easier, and inconsistency makes ‘‘programming,’’ or an inmate’s daily routine, unpredictable. As one minimum-security prisoner explained it to us: I think they’re too lax sometimes as far as letting things go. They’ll enforce go[ing] to eat, gotta have your ID or go back to your unit and get it. That lasted a whole week; well actually technically everybody is supposed to have your ID on you at all time[s] anyway. But a lot of people sometimes don’t have them on and they make you go back and get it. They just don’t follow through; they only follow through for a little bit of the time, and then the programs fall apart and break down and it just continues to break down.
By contrast, at the highest custody level (level 4), we see the opposite effect: those agreeing that COs go by the rulebook have a 15 Women who do not know their assigned custody level likely occupy the leastrestrictive custody level (level 1), because women in more secure custody levels have restrictions on how much time they get outside their cells and the amount of contact they can have with outsiders. These aspects of confinement are quite apparent to those who are subjected to them.
Vuolo & Kruttschnitt
327
36 CO no help
Adjustment Difficulty Scale
34
CO help
32
30
28
26
24
22 none
1st-4th grade
5th-8th grade
9th-11th grade
HS grad
some college
college grad
Education level
Figure 2. Interaction between Education and CO Help.
more difficult time adjusting. So while at the lowest level of custody, consistency according to the rulebook among COs is highly valued in terms of adjustment, at the highest custody level, this unwavering strictness may add insult to injury in the form of increased strictness in an already punitive environment. The interactions with education are shown in Figures 2 and 3. In Figure 2, we can see that adjustment difficulty increases as an inmate’s level of education rises, regardless of whether the inmate perceives COs as being helpful. However, this relationship is particularly attenuated for those who view COs as being unhelpful. In Figure 3, we see education as having somewhat of a similar impact on adjustment, but this relationship only appears for women who do not find COs writing women up for every infraction they see annoying. By contrast, for the inmates who find this behavior annoying, adjustment is poor, regardless of education level. What these interactions seem to suggest, then, is that while an inmate’s education is an important part of the picture in determining her ability to adjust to prison life, the degree to which COs are helpful and attend to ‘‘rules and regulations’’ has an important effect on determining how an educational advantage or disadvantage will play out. Perhaps this is because more-educated women have a very different understanding of staff-inmate relations than
328
Prisoners’ Adjustment, Correctional Officers, and Context
33 32
Adjustment Difficulty Scale
31 30 29 28 27 26 write up not annoying
25
write up annoying
24 none
1st-4th grade
5th-8th grade
9th-11th grade
HS grad
some college
college grad
Education level
Figure 3. Interaction between Education and COs Write Up Everything.
less-educated prisoners. Consider, for example, the extremely perceptive depiction this college-educated prisoner provided concerning what it is like trying to adjust to the demands of prison life and the place of COs in institutional life. You have to understand that . . . it takes a lot of patience and a lot of time, and a lot of diligence to choose your issues very carefully because you can’t take on the ‘‘I didn’t get a cookie’’ and ‘‘my clothes don’t fit’’ and ‘‘this isn’t right’’ and ‘‘that isn’t right’’ and ‘‘how can they live,’’ well none of it’s right, but it is. It’s a bunch of humans who don’t know any more than you know about what they’re doing. And they’re trying to just keep their own stuff straight. They’re in this just like we are. This staff will do far more prison time than I will . . . . They’re victims also . . . . They fight and complain, and argue, and go at each other just like the inmates do.
Perhaps most notable is the lack of an interaction between any of the variables representing CO behavior and prison. Thus net of our other variables, there is no difference between prisons in the way CO behavior affects adjustment. This finding lends support to our third hypothesis, that women’s adjustment to prison is
Vuolo & Kruttschnitt
329
impacted in a similar fashion by COs regardless of where they are imprisoned.
Discussion We began by noting that while the hallmarks of a ‘‘new penology’’ or postmodern penal movement have been carefully conceptualized and, at least, partially documented, they have also been questioned, most notably by those who have examined sites of implementation. Here the findings suggest that while some elements of the penal harm movement are flourishing, as evidenced by increases in managerialism, bureaucratization, and risk management (Adler & Longhurst 1994; Irwin & Austin 1994; Sparks et al. 1996; Simon 1993), others are perhaps best described as floundering. Those who have examined how the microprocesses that characterize these systemwide changes are being implemented find an uneven application of these new principles and practices (Haney 1996; Lynch 1998; Sparks et al. 1996). These studies resonate with the theories of Garland (1997, 1990) and others (O’Malley 1992, 1999; Cheliotis 2006) who argue that we need to examine the diversity and incoherence of current penal regimes and the ways that crime prevention policies and penal histories incorporate programs from different eras in an uneven and negotiated fashion. This study contributes to this line of research as it examines how the most immediate recipients of penal policyFconvicted offendersFare faring in light of a prison environment that now prioritizes truth-in-sentencing, security and classification, and responsibilization (O’Malley 1992, 1996) over rehabilitation. Specifically, focusing on the role of COs as conduits for the new policies, we tested three hypotheses. The first, in line with a strict constructionist interpretation of the new penology, suggests that COs will have little or no effect on women’s prison experiences as the CDC strives to ensure uniformity in the rules and regulations guiding all of its institutions. The second and third hypotheses are driven both by prior research on ‘‘front-line’’ workers that draws attention to the incomplete transformation in penal policy and by Garland’s (2006) recent attention to the way different levels of culture operate in the sociology of punishment. Here we reasoned that, on the one hand, COs could have a large effect on prisoners’ experiences and that this effect would vary depending on the institutional context. On the other hand, because women’s imprisonment has a specificity of its own, due to long-standing cultural attitudes toward women offenders and their punishment, COs should impact women’s adjustment but the effects should be consistent across institutions.
330
Prisoners’ Adjustment, Correctional Officers, and Context
While our findings provide the strongest support for the last hypothesis, they do not entirely dismiss the second hypothesis. Even though we found no evidence of an interaction between CO behavior and the institution in which they work, it is entirely possible, and even probable, that COs implement policies differently depending on where they work. Nevertheless, in the eyes of the primary audience of these policiesFthe prisonersFcentral aspects of their interactions with COs (whether they are seen as helpful, rule-bound, or ‘‘just doing a job’’) negatively impact their adjustment regardless of their carceral surrounding. As such, we think these findings support Garland’s (2006) recent call for examining multiple levels of culture in that they highlight the import of both the backdrop of a specific penal era and the foreground, or lived experiences, of those eras. The political and legislative shifts that moved women’s imprisonment in California from the rehabilitative era of CIW to the mass imprisonment era of VSPW are incomplete. Inmates still feel it is an advantage to be housed at CIW, relative to VSPW. But as we have seen, as important as these abstract distinctions are for women’s carceral lives, equally important is the fact that the new ‘‘culture of control,’’ or the ‘‘practices of interpretation’’ that it brings to bear upon prisoners, is being played out in very similar ways (Garland 2006:439). In addition, how the culture of control is brought to bear upon women offenders is more important to their ability to adjust to prison life than the prison institution itself. Another interesting feature of our findings, and one that also speaks directly to the invocation of a postmodern penality, is the relatively modest influence that women’s background characteristics have on their ability to adjust to their prison environment. This may well be a product of what the inmates call ‘‘take-aways,’’ or the CDC’s efforts to standardize the prison experience across all its 32 institutions. As the carceral experience becomes more extreme, individual experiences that marked a prisoner’s identity in the free world (e.g., her level of education, whether she was married or had children) may well fade in comparison to the prisoner’s immediate circumstances (Sykes 1958). So as we have seen, a prisoner’s ability to adjust to her prison environment is not derived from where she came from but rather from such pressing concerns as the length of time the prisoner is serving, her custody level, and how much control she feels she has over her day-to-day life. Can these findings be generalized beyond California or to men’s prison experiences? Our findings, and conclusions, may not reflect what is happening in other states. While California is often considered to be a bellwether state in penal policy, there is no doubt that the scope of its legislative reforms and the massive increases in its prison population over the last two decades of the
Vuolo & Kruttschnitt
331
twentieth century place it in a class by itself. From this perspective, we might think of California as the most rigorous test of the extent to which the new penology, or the postmodern penal movement, is affecting prisoners. It is possible, therefore, that in other parts of the country, where the changes in prison life have been less rapid and radical, we would see less continuity across institutions in both the extent and nature of COs’ influence on prisoner adjustment. It is also likely that our focus on women offenders’ experiences bias the direction of our results. As we noted at the outset, COs have long held rather rigid and stereotypical assumptions about female offenders. While these may be lessening over time as more women become COs and as men are exposed to more female offenders, they may also still influence the rhetoric and responses women incur from COs relative to their male counterparts (Britton 2003). Nevertheless, given our finding of institutional differences (see also Kruttschnitt & Gartner 2005) and the increasing security-driven focus in male institutions, we might expect male prisons to be more indicative of the penal harm movement. Finally, our findings also have important policy implications. Clearly, COs have a sizeable impact on prisoners’ ability to adjust to prison life. As many women prisoners revealed, there are some very good COs that manage to conduct themselves professionally and gain the respect and cooperation of the inmates. There are also those who fail miserably on both counts. While we will not be able to determine how these women who move out of CIW and VSPW fare in the future, we do know that most of them will in fact leave their carceral lives behind at some point and reenter society. Prisoner reentry is the focus of much current academic and policy concern, yet the focus of this concern seems to be disproportionately on the communities prisoners will be released into and not the prison communities from which they are emerging (Travis & Visher 2005). As the inmate we quote at the outset of this article noted, this may be quite shortsighted. Prison staff have the ability to improve the lives of their charges or to further damage them. To the extent that they engage in the latter rather than the former, they diminish not only these prisoners’ lives, but also the lives of their families and the communities to which they will return.
References Adams, Kenneth (1992) ‘‘Adjusting to Prison Life,’’ 16 Crime and Justice 275–359. Adler, Michael, & Brian Longhurst (1994) Discourse, Power and Justice: Toward a New Sociology of Imprisonment. London: Routledge. Alpert, Geoffrey P., et al. (1977) ‘‘A Comparative Look at Prisonization: Sex and Prison Culture,’’ 1 Q. J. of Corrections 29–34.
332
Prisoners’ Adjustment, Correctional Officers, and Context
Anderson, Elijah (1990) Streetwise. Race, Class, and Change in an Urban Community. Chicago: Univ. of Chicago Press. Biewen, John (2002) ‘‘Turning the Key: California’s Prison Guards,’’ American RadioWorks, http://www.americanradioworks.org/features/corrections/index.html (accessed 12 July 2006). Bondeson, Ulla V. (1989) Prisoners in Prison Societies. New Brunswick, NJ: Transaction. Bosworth, Mary (2000) ‘‘Confining Femininity: A History of Gender, Power and Imprisonment,’’ 4 Theoretical Criminology 265–84. Britton, Dana M. (2003) At Work in the Iron Cage. The Prison as Gendered Organization. New York: New York Univ. Press. Camp, Scott D. (1994) ‘‘Assessing the Effects of Organizational Commitment and Job Satisfaction on Turnover: An Event History Approach,’’ 74 The Prison J. 279–305. Carlen, Pat (1983) Women’s Imprisonment: A Study in Social Control. London: Routledge & Kegan Paul. FFF, ed. (1985) Criminal Women: Autobiographical Accounts. Cambridge, United Kingdom: Polity. FFF, ed. (1998) Sledgehammer: Women’s Imprisonment at the Millennium. Houndsmills, Basingstoke, United Kingdom: Macmillan. Cheliotis, Leonidas K. (2006) ‘‘How Iron Is the Iron Cage of New Penology? The Role of Human Agency in the Implementation of Criminal Justice Policy,’’ 8 Punishment & Society 313–40. Conover, Ted (2000) New Jack. Guarding Sing Sing. New York: Vintage. Craddock, Amy (1996) ‘‘A Comparative Study of Male and Female Prison Misconduct Careers,’’ 76 Prison J. 60–80. Crawley, Elaine (2002) ‘‘Bringing It All Back Home? The Impact of Prison Officers’ Work on Their Families,’’ 49 Probation J. 277–86. Crouch, Ben M., & James W. Marquart (1980) ‘‘On Becoming a Prison Guard,’’ in B. M. Crouch, ed., The Keepers: Prison Guards and Contemporary Corrections. Springfield, IL: Thomas Books. DeVellis, Robert F. (2003) Scale Development: Theory and Applications, Second Edition. Thousand Oaks, CA: Sage. Dickenson, Lee (1999) The Keepers of the Keys. Fort Bragg, CA: Lost Coast Press. Duffee, David (1980) ‘‘The Correction Officer Subculture and Organizational Change,’’ 11 J. of Research in Crime and Delinquency 155–72. Faily, Anwar, & A.Roundtree George (1979) ‘‘A Study of Aggression and Rule Violations in a Female Prison Population,’’ 4 J. of Offender Counseling, Services, and Rehabilitation 81–7. Feeley, Malcolm, & Jonathan Simon (1992) ‘‘The New Penology: Notes on the Emerging Strategy of Corrections and Its Implications,’’ 30 Criminology 449–74. Garland, David (1985) Punishment and Welfare: A History of Penal Strategies. London: Heinemann/Gower. FFF (1990) Punishment and Modern Society: A Study in Social Theory. Chicago: Univ. of Chicago Press. FFF (1997) ‘‘‘Governmentality’ and the Problem of Crime: Foucault, Criminology, Sociology,’’ 1 Theoretical Criminology 173–214. FFF (1999) ‘‘Editorial: Punishment and Society Today,’’ 1 Punishment & Society 5–10. FFF (2001) The Culture of Control: Crime and Social Order in Contemporary Society. Chicago: Univ. of Chicago Press. FFF (2006) ‘‘Concepts of Culture in the Sociology of Punishment,’’ 10 Theoretical Criminology 419–47. Gartner, Rosemary, & Candace Kruttschnitt (2004) ‘‘A Brief History of Doing Time: The California Institution for Women in the 1960s and the 1990s,’’ 38 Law & Society Rev. 267–304.
Vuolo & Kruttschnitt
333
Giallombardo, Rose (1966) Society of Women: A Study of a Women’s Prison. New York: Wiley. Gomez, Laura E. (1997) Misconceiving Mothers. Legislators, Prosecutors and the Politics of Prenatal Drug Exposure. Philadelphia: Temple Univ. Press. Haney, Lynn (1996) ‘‘Homeboys, Babies, and Men in Suits: The State and the Reproduction of Male Dominance,’’ 61 American Sociological Rev. 759–78. Herberts, Steven (1998) The Correctional Officer Inside Prisons. New York: Nova Science. Irwin, John (2005) The Warehouse Prison. Disposal of the New Dangerous Class. Los Angeles: Roxbury. Irwin, John, & James Austin (1994) It’s About Time: America’s Imprisonment Binge. Belmont, CA: Wadsworth. Jacobs, James (1978) ‘‘What Prison Guards Think: A Profile of the Illinois Force,’’ 24 Crime and Delinquency 185–96. Jensen, Gary F. (1977) ‘‘Age and Rule-Breaking in Prison,’’ 14 Criminology 555–68. Jensen, Gary F., & Dorothy Jones (1976) ‘‘Perspectives on Inmate Culture: A Study of Women in Prison,’’ 54 Social Forces 590–603. Kauffman, Kelsey (1988) Prison Officers and Their World. Cambridge, MA: Harvard Univ. Press. Kruttschnitt, Candace (1981) ‘‘Prison Codes, Inmate Solidarity, and Women: A Reexamination,’’ in M. Q. Warren, ed., Comparing Female and Male Offenders. Beverly Hills: Sage. Kruttschnitt, Candace, & Rosemary Gartner (2003) ‘‘Women’s Imprisonment,’’ 30 Crime and Justice 1–81. FFF (2005) Marking Time in the Golden State. Women’s Imprisonment in California. New York: Cambridge Univ. Press. Kruttschnitt, Candace, & Mike Vuolo (2007) ‘‘The Cultural Context of Women Prisoners’ Mental Health: A Comparison of Two Prison Systems,’’ 9 Punishment & Society 115–50. Kruttschnitt, Candace, et al. (2000) ‘‘Doing Her Own Time? Women’s Responses to Prison in the Context of the Old and the New Penology,’’ 38 Criminology 681–718. Lancefield, Kay, et al. (1997) ‘‘Management Styles and Its Effect on Prison Officers’ Stress,’’ 4 International J. of Stress Management 205–19. Liebling, Alison, & David Price (1999) An Exploration of Staff-Prisoners Relationships at HMP Whitemoor. Prison Service Research Report No. 6. London: HM Prison Services. FFF (2001) The Prison Officer. Leyhill, United Kingdom: Prison Service Journal. Lombardo, Lucien X. (1981) Guards Imprisoned: Correctional Officers at Work. New York: Elsevier. Lynch, Mona (1998) ‘‘Waste Managers? The New Penology, Crime Fighting, and Parole Agent Identity,’’ 32 Law & Society Rev. 839–70. FFF (2001) ‘‘Rehabilitation as Rhetoric: The Ideal of Reformation in Contemporary Parole Discourse and Practices,’’ 2 Punishment & Society 40–65. Mandaraka-Sheppard, Alexandra (1986) The Dynamics of Aggression in Women’s Prisons in England. Aldershot, United Kingdom: Gower. Mawby, Rob I. (1982) ‘‘Women in Prison: A British Study,’’ 28 Crime & Delinquency 224–39. Miller, Jody (2001) One of the Guys: Girls, Gangs, and Gender. New York: Oxford Univ. Press. Morrissey, Belinda (2003) When Women Kill: Questions of Agency and Subjectivity. New York: Routledge. O’Malley, Pat (1992) ‘‘Risk, Power, and Crime Prevention,’’ 21 Economy and Society 252–75. FFF (1996) ‘‘Risk and Responsibility,’’ in A. Barry et al., eds., Foucault and Political Reason: Liberalism, Neo-Liberalism, and Rationalities of Government. Chicago: Univ. of Chicago Press.
334
Prisoners’ Adjustment, Correctional Officers, and Context
FFF (1999) ‘‘Volatile and Contradictory Punishment,’’ 3 Theoretical Criminology 175–96. Papworth, Richard H. (2000) Key Man. Derbyshire, United Kingdom: Richard Papworth. Pens, Dan (1998) ‘‘The California Prison Guards’ Union. A Potential Interest Group,’’ in D. Burton-Rose & P. Wright, eds., The Celling of America: An Inside Look at the US Prison Industry. Monroe, ME: Common Courage Press. Pollock, Joycelyn M. (1986) Sex and Supervision: Guarding Male and Female Inmates. New York: Greenwood. Poole, Eric D., & Robert M. Regoli (1981) ‘‘Alienation in Prison: An Examination of the Work Relations of Prison Guards,’’ 19 Criminology 251–70. Rafter, Nicole Hahn (1990) Partial Justice: Women in State Prisons: 1800–1935, 2d ed. New Brunswick, NJ: Transaction. Rasche, Christine (2001) ‘‘Cross-Sex Supervision of Female Inmates: An Unintended Consequence of Employment Law Cases Brought by Women Working in Corrections.’’ Paper presented at the annual meeting of the American Society of Criminology, Atlanta. Rhodes, Lorna A. (2004) Total Confinement: Madness and Reason in the Maximum Security Prison. Berkeley: Univ. of California Press. Ruback, R. Barry, & Timothy S. Carr (1984) ‘‘Crowding in a Women’s Prison: Attitudes and Behavioral Effects,’’ 14 J. of Applied Social Psychology 57–68. Ruback, R. Barry, et al. (1986) ‘‘Perceived Control in Prison: Its Relation to Reported Crowding, Stress, and Symptoms,’’ 16 J. of Applied Social Psychology 375–86. Sharma, Sagar, & Devender Sharma (1989) ‘‘Organizational Climate, Job Satisfaction and Job Anxiety,’’ 34 Psychological Studies 21–7. Simon, Jonathan (1993) Poor Discipline: Parole and the Social Control of the Underclass, 1890–1990. Chicago: Univ. of Chicago Press. FFF (2000) ‘‘The ‘Society of Captives’ in the Era of Hyper-Incarceration,’’ 4 Theoretical Criminology 285–308. Simon, Jonathan, & Malcolm Feeley (1995) ‘‘True Crime: The New Penology and Public Discourse on Crime,’’ in T. G. Blomberg & S. Cohen, eds., Punishment and Social Control: Essays in Honor of Sheldon L. Messinger. New York: Aldine de Gruyter. Sparks, Richard, et al. (1996) Prisons and the Problem of Order. New York: Oxford Univ. Press. Stojkovic, Stan (2003) ‘‘Accounts of Prison Work: Corrections Officers’ Portrayals of Their Work Worlds,’’ in M. R. Pogrebin, ed., Qualitative Approaches to Criminal Justice: Perspectives from the Field. London: Sage. Sykes, Gresham M. (1958) The Society of Captives: A Study of a Maximum Security Prison. Princeton, NJ: Princeton Univ. Press. Thomas, J. E. (1972) The English Prison Officer Since 1850: A Study in Conflict. London: Routledge & Kegan Paul. Tittle, Charles R. (1969) ‘‘Inmate Organization: Sex Differentiation and the Influence of the Criminal Subcultures,’’ 34 American Sociological Rev. 492–505. Travis, Jeremy, & Christy Visher, eds. (2005) Prisoner Reentry and Crime in America. New York: Cambridge Univ. Press. Triplett, Ruth, et al. (1996) ‘‘Work-Related Stress and Coping among Correctional Officers: Implications from Organizational Literature,’’ 24 J. of Criminal Justice 291–308. Walters, Stephen (1991) ‘‘Alienation and the Correctional Officer: A Multivariate Analysis,’’ 16 American J. of Criminal Justice 50–62. Wright, Thomas A. (1993) ‘‘Correctional Employee Turnover: A Longitudinal Study,’’ 21 J. of Criminal Justice 131–42. Zingraff, Matthew T., & Rhonda Zingraff (1980) ‘‘Adaptation Patterns of Incarcerated Female Delinquents,’’ Juvenile and Family Court J. 35–47 (May)
Vuolo & Kruttschnitt
335
Statutes Cited Proposition 184, Cal. Penal Code §§ 667, 1170.12 (2006). Uniform Determinate Sentencing Act, Cal. Penal Code §§ 1170, 3000, 3040 (2006).
Mike Vuolo is a Ph.D. candidate in the Department of Sociology at the University of Minnesota. In 2007, he received an M.S. in statistics from the University of Minnesota. His most recent publications appear in Punishment and Society and Work and Occupations. Candace Kruttschnitt is a Professor in the Department of Sociology at the University of Toronto. Her recent books are Marking Time in the Golden State: Women’s Imprisonment in California (with Rosemary Gartner, Cambridge University Press, 2005) and Gender and Crime: Patterns of Victimization and Offending (with Karen Heimer, New York University Press, 2006).
336
Prisoners’ Adjustment, Correctional Officers, and Context
337
Elections as Focusing Events: Explaining Attitudes Toward the Police and the Government in Comparative Perspective
Lee Demetrius Walker
Richard W. Waterman
Traditional views hold that citizens’ attitudes toward the police are driven by local concerns. We contend that public attitudes toward the police are responsive to systematic and periodic national-level political factors. In particular, we show that national elections as a focusing event alter periodically the determinants of attitudes toward the police. Using a logistic regression model and diachronic data from Costa Rica, Mexico, and the United States, we find that attitudes toward the police and the national government are linked, and this linkage is responsive to the influence of national election campaigns in varying degrees. In addition, we find that attitudes toward the Mexican police are sensitive to partisan changes in the composition of the national political government. We find no such sensitivity in the police attitudes of Costa Rican and U.S. citizens. This suggests that police attitudes are not only affected by the performance of the national political government but also by the character (consolidated versus unconsolidated) of the national political government. In short, police attitudes in new democracies are an indication of the unconsolidated nature of the state apparatus.
C
lassic studies of public attitudes toward the police point to local factors as the predominant determinants in police attitudes (Whyte 1943; Wilson 1963). However, more recent studies indicate that attitudes toward national governmental structures also influence attitudes toward the police (Albrecht & Green 1977; Cao & Zhao 1998, 2005). Extending this research, we find that perceptions of the police are related to evaluations of the national government in three different countries. We also find that this relationship is stronger during national election years, when issues of crime and punishment are more likely to be on the national This article was written with support from the University of Kentucky, the University of South Carolina, and a resident fellowship at The Institute for Quantitative Social Science at Harvard University. We also wish to thank Jeff Gill, Gary King, James L. Gibson, Andrew Martin, Christopher Zorn, the political science departments of the University of Kentucky and University of South Carolina, and Law & Society Review’s anonymous reviews. Please address correspondence to Lee Demetrius Walker, University of South Carolina, Department of Political Science, 350 Gambrell Hall, Columbia, SC 29208; e-mail: walker23@ gwm.sc.edu; or Richard W. Waterman, University of Kentucky, Department of Political Science, 1637 Patterson Office Tower, Lexington, KY 40506-0027; e-mail:
[email protected].
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
338
Elections as Focusing Events
agenda. Elections serve as focusing events that impact public perceptions of the police in a systematic fashion. Focusing events can be particularly important cues for citizen attitudes toward the police. Tuch and Weitzer (1997:647), Shaw et al. (1998), and Sigelman et al. (1997) show that highly publicized incidents of police misconduct affect public confidence in the police at both the local and national level (i.e., the Los Angeles policemen’s 1991 beating of Rodney King). Cao and Hou (2001) also illustrate that public attitudes toward the police are related to major political events such as the 1990 Tiananmen Square incident. While these two examples indicate that idiosyncratic national focusing events affect the confidence that citizens hold for the police, we ask what effect national events that occur systematically, such as elections, exert on perceptions of the police.1 Local and state elections have definitive policy implications on police administration. For example, Levitt (1997) demonstrates that increases in the size of police forces are disproportionately concentrated in mayoral and gubernatorial election years. But what of national factors? During the 1996 presidential election, former U.S. President Bill Clinton proclaimed the deployment of 100,000 new local police officers as a legitimate accomplishment of his administration. Politicians often invoke themes of ‘‘law and order’’ in their campaigns (e.g., U.S. President George H. W. Bush in 1988). Moreover, McCann and Lawson (2003:69) show that Mexicans’ attitudes toward crime control were responsive to campaign effects from the 2000 Mexican presidential campaign. Given the evidence cited above, which demonstrates the effect of national events on public opinion toward the police and crime and the policy effects of election cycles on police administration, it is surprising that the relationship between national elections and police attitudes has not been the subject of more systematic and cross-national analysis. We argue that attitudes toward the police are more systematically linked to the national government than previously understood. Hence, using multinational and longitudinal survey data, we test whether and how citizens’ attitudes toward the police relate to their perceptions of the national political government. As a manifestation of the national government/police relationship, we argue that elections, with their greater focus in recent decades on crimerelated issues, provide a focusing event at the national level that influences citizen perceptions of the police at the local level. This is an important focusing event because the national election of the executive involves not just a vote on the national government but, 1 Kingdon (1995:98) describes crises such as the Rodney King incident as focusing events.
Walker & Waterman
339
more important, a dissemination of issues and a discussion of the greater need for law and order. Furthermore, these elections occur across all presidential countries. Accordingly, we examine this phenomenon in a comparative context over time. Using data from Costa Rica, Mexico, and the United States, we find that attitudes toward the police and the national government are linked, and this linkage is responsive to the influence of national election campaigns in varying degrees. We also find that attitudes toward the Mexican police are sensitive to partisan changes in the composition of the national political government, which suggests that police attitudes in new democracies are an indication of the unconsolidated nature of the state apparatus.
Perceptions of the Police There is an extensive literature on determinants of citizen attitudes toward the police. Research in the United States finds that contact with the police, neighborhood, race, and age affect attitudes toward the police (see Brown & Benedict 2002). Age is positively related to confidence in the police (Benson 1981; Zamble & Annesley 1987; Correia et al. 1996), with older citizens more likely to express confidence in the police than younger ones. In addition, much literature suggests that police community presence and public contacts affect public opinion of the police (Reiss 1967; Smith 1986; Tyler 1988; Shaw et al. 1998; Reisig & Parks 2000; Terrill & Reisig 2003; Sunshine & Tyler 2003). Favorable-contact arguments are consistent with the neighborhood thesis (Whyte 1943; Wilson 1963). Public perception of the police is a neighborhood or local phenomenon because the police are responding to local-oversight institutions and are subject to the approval of local public opinion (Bordua & Tifft 1971; Weitzer 2000; Seron et al. 2004). It follows necessarily, then, that public opinion must be based largely on the evaluation of the performance of the local police. More recently, much empirical evidence shows that public opinion of the police is highly sensitive to events that are heavily publicized by the national media (see, e.g., Albrecht & Green 1977; Tuch & Weitzer 1997; Shaw et al. 1998; Sigelman et al. 1997; Sampson & Bartusch 1998; Cao & Hou 2001). The urban riots of the middle to late 1960s, police violence against civil rights protestors and African Americans, and other high-profile incidents have attracted the attention of the national media. In addition, a national agenda that stressed the need for greater law and order, tougher sentencing guidelines, and more police on the streets became a part of the national debate. Often at the center of these
340
Elections as Focusing Events
debates were issues related to race and ethnicity. Lasley (1994), Tuch & Weitzer (1997), and Sigelman et al. (1997) demonstrate how short- and long-term effects of publicized police misconduct affect differently the opinions that majority and minority populations have concerning the national agenda of greater law and order.2 U.S. research provides considerable evidence for a significant relationship between the police and national governmental politics. This research, however, does not examine whether the determinants of national government/police relationships vary in some predictable and systematic fashion or provide a theory capable of explaining such systematic variations.
Police and Democratization in Latin America There is reason to believe that national effects are not simply related to the political dynamics in the United States (Weitzer 1995; Cao & Hou 2001; Cao & Zhao 2005).3 Cao and Zhao show that ‘‘[T]rust in the [national] political system’’ has the strongest influence on police attitudes in nine Latin American countries and the United States (2005:409). They also show that support for police varies predictably, with citizens in the more stable democracies of the United States, Uruguay, and Chile expressing greater confidence in the police than citizens in less well–performing democracies of the Dominican Republic, Peru, Argentina, and Mexico (2005:408).4 Regrettably, little public opinion work has been done to assess attitudes toward police in new Latin American democracies.5 Given 2 In addition, Howell et al. (2004) find that African American evaluations of the police are consistent even across majority black or majority white city contexts. 3
Weitzer (1995) finds that ethnic conflict is a significant component of policecommunity relations in Northern Ireland, while Cao and Hou (2001) study Chinese public opinion in the aftermath of the Tiananmen Square incident. 4 They use World Values Surveys from 1995 to 1997. The World Values Survey is a worldwide investigation of sociocultural and political change conducted by the World Values Survey Association. The organization compiled cross-national surveys in 1990, 1995, 2000, and 2005. The World Values Surveys grew out of a study launched by the European Values Survey group (EVS) under the leadership of Jan Kerkhofs and Ruud de Moor in 1981. The measured variable Police Support was 71 percent in the United States, 51.5 percent in Uruguay, 51 percent in Chile, 12 percent in the Dominican Republic, 20 percent in Peru, 21 percent in Argentina, and 32 percent in Mexico. The Dominican Republic, Peru, Argentina, and Mexico are less stable and less free, based on both Freedom House and Polity IV indexes (Cao & Zhao 2005:408). Freedom House is a nonprofit organization that annually publishes Freedom in the World and rates countries on their level of adherence to political rights and civil liberties. The Polity IV project provides annual information on regime and authority characteristics. The project assigns scores for overall polity performance, level of democratic performance, and level of autocratic performance. 5 Olivero and Murataya (1998) and Brown et al. (2006) are community-based examinations of public attitudes in both Guadalajara and Tampico, Mexico.
Walker & Waterman
341
the dependent nature of Latin American police forces during authoritarian periods and the important role that police play in democratic governance, changes in public opinion about the police tell us a great deal about regime transformation. Beyond public opinion research about Latin American police departments, Cao and Zhao point out that there has been ‘‘little systematic discussion of evolution of police systems since the 1980s’’ (2005:404). Call’s (2002) study of new civilian security forces in Latin America is a noteworthy exception to this lacuna. Call shows that the mode of democratic transition best accounts for reforms to internal security structures. Still, reforms have been incremental in most Latin American countries, and protecting the government from its citizens remains the first mission of most Latin American police departments (2002:7–8).6 Important to note, Call demonstrates that the transition from authoritarianism to democracy portends changes for police administration. Cao and Zhao suggest that democratic development matters in attitudes toward the police, and they posit a direct relationship between national governmental institutions and police administration (2005:404). Nevertheless, the research does not examine whether specific governmental arrangements or levels of institutional development influence the national government/police relationship. In addition, the research fails to provide a theory that is capable of explaining cross-national variation in the relationship. Our theory is based on the idea that attitudes about the police are conditioned by the nature of the relationship between the national government and the police. Moreover, national elections focus attention on the partisan struggle for control of the democratic state (of which the police power is an essential part). This is the focus of our research.
The Police and Political Government: A Model Most models of attitudes toward the police are based solely on local factors. The most prominent among these is provided by Wilson (1963). He offers a theory of police behavior and public support for the police at the local level that with modification has applicability to the national level. Wilson argues, ‘‘[i]f a professional police force can only exist in large cities as part of a set of political and civil institutions of certain character, the desirability of professionalism cannot be considered apart from the desirability of these institutions as a set’’ (1963:213). Wilson’s system contains three 6 Of 17 Latin American countries, only Costa Rica, Ecuador, Guatemala, and Nicaragua have formal police doctrines that charge the police with the defense of citizens and their rights.
342
Elections as Focusing Events
components: the police must attain a level of professionalism characterized by institutional autonomy; the other institutions of the system must attain a certain (nonpartisan) character; and the public must have sufficient representation and access to support the police. At the national level, Wilson’s professionalism component can be viewed as part of the usable state bureaucracy that operates within professional norms (Linz & Stepan 1996:55). This means the police are part of the state but not a part of political society’s electoral contestation. In turn, political society must allow the police bureaucracy to operate without unwarranted encroachment from national political actors (usability). This is consistent with Wilson’s notion that political institutions in the set of political and civil institutions attain a ‘‘certain nonpartisan’’ character. This dynamic should hold for both federal and unitary countries. That is to say, there is a relationship between the national government and the local police regardless of the structural division of power. It can be represented by the following equation: PðPSÞ ¼ b1 ðPGSÞ þ bj Zj ; where P is the probability of support for the police (PS), b1 is the coefficient estimated effect of support for the political government (PGS) on support for the police, Zj is a set of control variables that include the traditional determinants of police attitudes (age, race, neighborhood [local factors], party identification, and political ideology), and bj is the set of coefficient estimates for these control variables. At this point, we reiterate that all democratic governments are not the same. Cao and Zhao (2005) show that these differences matter in a country’s aggregate level of police support. Measures of democracy such as those of Freedom House and Polity IV align well with aggregate levels of support for the police. Freedom House and Polity IV measures are built on traditional theories of democracy that emphasize regime-level factors such as free and fair elections and observance of civil and political rights (Schumpeter 1975; Dahl 1989).7 Why might democratic development affect the level of police support? O’Donnell argues that a theory of democracy must go beyond the regime level and include aspects of the legal system of the state (2001:25). In turn, political (full) democracy differs from all other political types and has ‘‘four unique differentiating characteristics’’ (2001:25): (1) fair and institutionalized elections, 7 Dahl’s definition of polyarchy includes the following: elected officials are selected in free and fair elections, there is inclusive suffrage, practically all adults have the right to run for office, there is freedom of expression, there are alternative sources of information, and there is associational autonomy (1989:22).
Walker & Waterman
343
(2) an inclusive and universalistic institutionalized wager (this is what Linz and Stepan call a working consensus about procedures of governance [1996:10]), (3) a legal system that enacts and backs the rights included in the democratic regime, and (4) a legal system that prevents anyone from being above and beyond the law. This typology divides these four characteristics at two levels: Conditions 1 and 2 are located at the regime level, and Conditions 3 and 4 are located at the legal system of the state level. In addition, this typology not only distinguishes political democracy from other regime types but also distinguishes political democracy from ‘‘diminished kinds’’ of democracy (O’Donnell 2001:25).8 Even if new democracies meet regime-level conditions as prescribed by theories of democracy, failure to meet the legal system conditions relegates the nation-state to ‘‘consolidating’’ status. We use this full/consolidating dichotomy to differentiate democracies. In new democracies, the set of political and civil institutions has not attained the same character as the set of institutions in full democracies: most prominent, an autonomous legal system of the state. Nevertheless, Wilson’s model should apply in new democracies where the legal system has been subordinated to the executive government. This legal system subordination includes the subordination of the police. Wilson’s model applies because the police and the national political government are linked politically. That is, individuals who support the national political government are likely to support the police organization that they believe the national political government controls. However, we should expect the police in developing democracies to generate less public support because they are subordinate to the dominant political interest. Still, persons who politically support the government should also politically support the police. Accordingly, this model yields the following hypothesis for both full and consolidating democracies: Hypothesis 1: Controlling for the traditional determinants of police confidence, citizens positively and significantly link their perceptions of the police to their perceptions of the national government.
Elections as Focusing Events Wilson’s theory helps us understand how attitudes toward the police are associated with attitudes toward the national government. Still, it cannot account for idiosyncratic or periodic fluctuations in the strength of this association. Idiosyncratic fluctuations may relate directly to crises in the national government. Certainly, 8
See also Collier and Levitsky (1997).
344
Elections as Focusing Events
the urban riots of the middle to late 1960s were connected directly to shifting confidence in the U.S. government. Kingdon argues that crises are focusing events that call attention to a problem (1995:94). For example, the Rodney King beating was a focusing event that called attention to police brutality. Kingdon and others are interested in the effect of focusing events on agenda-setting and policy outcomes.9 We contend that national elections are periodic focusing events that affect the national government/police confidence relationship. Kingdon argues that focusing events are associated with powerful symbols (1995:97). Political leaders use symbols (such as patriotism) to focus attention on a subject that is normally in the background of people’s minds. Elections often serve as focusing events that call attention to the police as a symbol of crime prevention and law and order. As a focusing event, the rhetoric of campaigns focuses public attention on both the partisan nature of the national government and issues of law and order, as well as crime and punishment. Political rhetoric therefore is different in presidential election and nonpresidential election years, and this rhetoric intensifies the relationship between the symbol (the police) and the national political government. Our use of this agenda-setting theory is justified because symbols and focusing events only ‘‘rarely carry a subject to a policy agenda’’ (Kingdon 1995:98). This means that subjects and symbols may be deployed strategically during a focusing event with little concern for an accompanying policy agenda. While the goal of such rhetoric is to elect the politicians who use it, there is evidence that elections do have important policy consequences for the police as well. Police receive greater resources during election years (Levitt 1997). Here we examine whether elections also influence attitudes about the police. If the police are a symbol, why might attitudes toward the police shift during a presidential election? Kessel argues that there are three attitudinal properties to voting behavior: salience, partisan valence, and importance (1980:197–8). These three properties also help explain the shift in the determinants of public attitudes toward the police during presidential elections. Salience is the prominence of an attitudinal object. The more salient a topic (or symbol), the more likely the citizen is to have an attitude about it. Partisan valence concerns how the attitudes about the symbol sum to form a partisan valence; the attitude gives one party an electoral advantage over others. Importance concerns the extent to which votes depend on a given attitude. This means that ‘‘an 9
See Birkland (1997), Godwin and Schroedel (2000), and Shipan and Volden (2006).
Walker & Waterman
345
attitude that is salient and quite favorable to one party is usually important’’ (Kessel 1980:198). In short, attitudes toward the police are more strongly linked to the national government during presidential elections because the attitudes are salient, because they can be used for partisan advantage by a political party, and because they are related to an individual’s voting decision. Elections therefore are focusing events that are associated with shifts in perceptions of the police.10 This argument yields an enhanced predictive model: PðPSpy Þ ¼ ðdÞb1 ðPGSÞ þ bj Zj ; d > 1; where P is the probability of support for the police during a presidential year (PSpy), b1 is the coefficient estimated effect of support for the political government (PGS) on support for the police, Zj is a set of control variables that include the traditional determinants of police attitudes (age, race, neighborhood, party identification, and political ideology), and bj is the set of coefficient estimates for these control variables. We add the d tuning parameter to capture the intensified relationship between the national government and the police. d is a ratio that measures the strength of the political-government-support effect on police support in the presidential year divided by the strength of the average political-governmentsupport effect on police support in nonpresidential years. In short, d 5 [(b1(py)/sPGS(py))/(b1(bar)/sPGS(bar))].11 As b1 is multiplied by the d tuning parameter, d takes a value greater than 1 to account for the increased effect of national government support on police support in presidential election years. This argument yields our second hypothesis. Hypothesis 2: Citizens more greatly link their perceptions of the police to their perceptions of the national government during presidential election years.
Volatility in Partisanship We expect Hypothesis 2 to hold for both consolidating and full democracies. At the same time, ‘‘changing the cast of characters’’ may have different implications in these two types of democracy. These implications connect to O’Donnell’s (2001) argument about 10 Kingdon surmises that elections are important ‘‘simply because the cast of characters in positions of authority can change’’ (1995:62). 11 The d parameteris analogous to an interaction term between a presidential-year dummy variable and PGS if we had surveys over several years and were able to run responses across these several yearly surveys in a single regression model. Given the lack of such data, d offers a reasonable estimate of the effect.
346
Elections as Focusing Events
the legal system of the state. Because citizens in consolidating democracies are more likely to view the police as dependent on the political government (because the police have been historically), the partisan struggle for control of political government affects more greatly the attitudes that citizens hold about the police. This is because control of the executive branch infers partisan control of the coercive force of the democratic state (control of the legal system of the state). Because the set of national civil and political institutions have not reached the necessary level to infer police professionalization, individual-level attitudes about the police fluctuate due to changes in the partisan composition of the national political government. This means partisans are likely to shift their support for the police based on their relationship with the party that controls the national political government. In full democracies such as the United States, the national political government approximates the nonpartisan character that Wilson (1963) posits. Accordingly, partisans are not likely to shift their support for the police based on their relationship with the party that controls the national political government. Hypothesis 3: Partisan police attitudes in developing democracies are more likely to shift due to change in partisan control of the national government than are partisan police attitudes in full democracies.
The Police and Government in Three Countries We test our focusing-event thesis in three presidential democracies (Costa Rica, Mexico, and the United States). We do so because we employ a most different systems approach to comparative research (Przeworski & Teune 1970). We choose these three countries because we want to assess the national government/police support relationship in the context of varying governmental conditions. These countries vary in governmental (structure) division of power, level of democratic and party-system development, military/police relations, and level of governmental centralization. Costa Rica is a unitary country, and its police department is subject to national-level oversight. The Costa Rican national police structure is housed in two national institutional structures. The Ministry ´ blica (the National Police), while of Interior oversees the Fuerza Pu the Supreme Court of Justice oversees the Organismo de Invest´n Judicial (OIJ). The Fuerza Pu ´ blica oversees disorder patrol, igacio while the OIJ oversees investigative duties. Mexico and the United States are two federal cases. The United States is the model of the federal policing configuration. Local police are generally accountable to the mayor and city council.
Walker & Waterman
347
Wilson points out that many cities have established nonpartisan local elections to aid in the oversight of the police (1963:172). Nevertheless, national political leaders assert a national connection to the police. Meanwhile, the Mexican police are also subject to some local oversight. Nevertheless, the police, like much of the Mexican state prior to 1997, have been subjugated to Mexico’s hegemonic party system (Camp 1999), and the jurisdiction of the Mexican federal police overlaps with those of state and local police. The limited legitimacy of the Mexican police was related strongly to the legitimacy of the Partido Revolucionario Institucional (PRI), which lost in the 2000 presidential election. The relationship between the PRI and the Mexican police suggests that partisan identification should contribute to a person’s attitude about the police in the developing Mexico democracy. Mexico represents our developing democracy case, while we consider the United States and Costa Rica full democracies (see Linz & Stepan 1996 concerning Costa Rican democratic exceptionalism). Theoretically, we contend in Hypothesis 3 that police attitudes in developing democracies are likely to be sensitive to changes in partisan control of the national government. We expect that partisan attitudes in Mexico will change because the PRI lost control of the national government after the 2000 presidential election victory of Vicente Fox, the National Action Party (PAN) candidate. Alternatively, we do not expect changes in attitudes in the police attitudes of partisans in Costa Rica or the United States. That is, we do not expect National Liberation Party (PLN) partisans in Costa Rica to be less likely to support the police than Christian Social Unity Party (PUSC) partisans because the PLN lost control of the national government after the 1998 presidential election. Miguel Angel Rodrı´guez of the PUSC replaced Jose´ Marı´a Figueres of the PLN after the 1998 election. Meanwhile, we do not expect U.S. Republicans to become less likely to support the police than U.S. Democrats because the Democratic Party controlled the national executive government in 1998, 1999, and 2000. The military is also an important source of variation across the three countries. The subordination of the police to the military during previous authoritarian periods contributes to a lack of autonomy and legitimacy for many Latin American police departments (Call 2002). However in the two Latin American cases in this study, the military has a lesser role. Costa Rica eliminated its military in 1949. Consequently, Costa Rica has the least militarized police force in Latin America (Call 2002). For Mexico, the military has been responsible for and accountable to civilian control. Since the breakdown of the hegemonic party system in 1997, the Mexican military has had an expanded role in Mexican internal
348
Elections as Focusing Events
Table 1. Police and the Government in Three Countries Country Costa Rica Government Structure Military Role in Domestic Policing Party structure
Mexico
United States
unitary
federal
federal
none (no military) competitive two-party system
substantial and increasing three-party system emerging from hegemonic-party system some but increasingly subject to national oversight developing (consolidating) emerging from subordination to PRI
limited to national emergencies competitive two-party system
Local Oversight of Domestic Policing
little but increasing
Democratic Regime
full
Legal System Autonomy
high
Countries and Hypotheses H1: Police linked to yes National Government H2: Presidential-Year yes Increase H3: Significant no Partisanship Shift
substantial with little national interference full high
yes
yes
yes
yes
yes
no
security.12 Increasingly, the military has assumed positions of command over police officials during this nationalization. For its part, the U.S. military is firmly under civilian control and only functions in U.S. internal security in times of national emergencies. Meanwhile, local (see Wilson 1963) and state government have important relationships with the police. While the Costa Rican police are subject to national oversight, Costa Rica has begun efforts to decentralize state structures. Indeed, 2002 marked the first municipal elections for Costa Rican mayors. This infers that battling local crime should become an important campaign issue at the local level, and mayors and city councils should seek greater control of local police departments. Due to federal governmental structures in Mexico and the United States, local and state governments play the most important role in police oversight. A summary of important differences between the three countries is presented in Table 1, which also contains country-level expectations for our three hypotheses. We expect Hypotheses 1 and 2 to hold for individuals in the three countries. We expect Hypothesis 3 (a significant partisanship shift) to occur only in Mexico, because of Mexico’s consolidating status. 12 For example, Mexico City police were nationalized under military control in 1996 ´pez-Montiel 2000:79 and Pereira 2001). Also in 1997, the Mexican and 1997 (see Lo military took control of the police in Guerrero, Veracruz, Oaxaca, Yucata´n, and Chiapas.
Walker & Waterman
349
Data We use four data sets to examine citizens’ evaluations of the police: the 1998 Hewlett data set, the 1998 and 1999 University of New Mexico’s Institute for Public Policy (IPP) National Public Opinion Survey, the 2000 World Values System Survey, and the ´metro. The Hewlett data set comes from a study 2003 Latinobaro that examines democratic citizenship in Costa Rica and Mexico. Market and Opinion Research International in Mexico and the research firm Dichter and Neira in Costa Rica performed the surveys in July 1998. Total respondents for the two samples are 1,002 for Costa Rica and 1,200 for Mexico, and the same precoded questionnaire was used. The IPP at the University of New Mexico survey of public opinion in the United States used the same precoded questionnaire for its two waves: September to October 1998, and October to November 1999. Total respondents for the two samples are 1,085 for 1998 and 918 for 1999. The World Values System surveys were performed by the following polling agencies during the following time periods: for ´gico Auto ´nomo de Me´ xico from JanMexico, the Instituto Tecnolo uary 28 to February 7, 2000; and for the United States, the Institute for Social Research at the University of Michigan in two phases: November 19 to December 23, 1999, and August 4 to September 25, 2000. Total respondents for the two samples are 1,535 for Mexico and 1,200 for the United States. The Latino´metro Corporation compiled the 2003 Latinobaro ´metro data. baro The same precoded questionnaire was used in the two countries. Total respondents for the two samples are 1,004 for Costa Rica and 1,200 for Mexico. The Costa Rican and Mexican samples represent 100 percent of the national population.13 All samples have a margin of error between 3 and 3.5 percent.
Public Perception of Police and Government Our first task is to examine the correlation between perceptions of the national government and citizens’ attitudes toward the police. Table 2 shows the percentages of respondents who express ‘‘some’’ or ‘‘much’’ confidence in the police, the national government, the military, local government, and state government. For the 1998 and 1999 University of New Mexico surveys, the table shows the percentages of respondents who believe that the particular institution is doing a ‘‘good’’ or ‘‘excellent’’ job. Table 2 also 13 ´metro as unrepresentative and biased Scholars have criticized the Latinobaro toward urban respondents (e.g., Canache et al. 2001).
350
Elections as Focusing Events
Table 2. Confidence in Police, Government, and Sub-National Government Percent Expressing Confidence Country and Year a
Costa Rica 1998
c
Costa Rica 2003 a
Mexico 1998
b
Mexico 2000
c
Mexico 2003
United States 1998
d
United States 1999
d
United States 2000
b
Police
Government
Military
Local
State
38.7 (1.000) NA 36.2 (1.000) NA 33.4 (1.000) NA 28.3 (1.000) NA 15.7 (1.000) NA 73.0 (1.000) NA 70.7 (1.000) NA 71.4 (1.000) NA
39.8 (0.397) (1.000) 23.1 (0.304) (1.000) 30.0 (0.326) (1.000) 35.8 (0.542) (1.000) 23.5 (0.222) (1.000) 44.9 (0.154) (1.000) 36.4 (0.210) (1.000) 38.0 (0.320) (1.000)
NA
NA
NA
999
NA
31.9 (0.253) (0.261) NA
NA
1,004
NA
1,200
NA
NA
1,535
24.5 (0.184) (0.207) 56.5 (0.257) (0.221) 51.6 (0.230) (0.238) NA
NA
1,200
60.4 (0.210) (0.193) 54.7 (0.180) (0.197) NA
1,085
44.8 (0.259) (0.273) 51.7 (0.308) (0.365) 38.8 (0.233) (0.293) NA NA 81.3 (0.365) (0.284)
N
918 1,200
Notes: Correlation with confidence in the police is shown in first parenthesis line below the confidence percentage. Correlation with confidence in the government is shown in the second parenthesis line below the confidence percentage. Percentage Expressing Confidence is individuals who express some or much confidence in the institution. For the 1998 and 1999 U.S. samples, confidence is the citizens who respond that the institution is doing an excellent or good job. NA 5 Not asked. Data sources: a ´metro; d1998 and 1998 Hewlett Survey; b2000 World Value Survey; c2003 Latinobaro 1999 University of New Mexico Public Policy Survey.
shows the Pearson correlation between perceptions of the police and perceptions of the other institutional structures. Three of the surveys examine attitudes in Mexico. Mexicans’ confidence in the government increases from 30 percent in 1998 to 36 percent in 2000 and then declines severely to 23.5 percent in 2003, while their confidence in the police declines in a more linear but equally dramatic fashion for the same time period (33, 28, and 16 percent, respectively).14 Moreover, governmental confidence correlates most highly with confidence in the police in 2000 (r 5 0.542), while the Mexican police/government correlations for 1998 and 2003 are significantly lower (r 5 0.326 and 0.222, respectively). Mexico held an historic presidential election in 2000, and this correlation surge in 2000 offers some support for Hypothesis 2. A similar correlation pattern occurs among the three U.S. distributions. While U.S. citizens express far greater and more consistent 14 For differences in police percentages, Chi-square 5 8.08 and p-value 5 0.00448. For differences in government percentages, Chi-square 5 10.18 and p-value 5 0.001420.
Walker & Waterman
351
confidence in the police (73 percent in 1998, 71 percent in 1999, and 71 percent in 2000) than they do in the government (45 percent in 1998, 36 percent in 1999, and 38 percent in 2000), U.S. governmental confidence correlates with confidence in the police most highly in the 2000 presidential year (r 5 0.320). The governmental/ police correlation reaches only 0.154 in 1998 and 0.210 in 1999. The Costa Rican distribution also provides evidence for this possible presidential year surge. Costa Ricans’ confidence in the government is correlated with confidence in the police at 0.397 in the presidential year of 1998, but declines to 0.304 during the nonpresidential year of 2003.15 Bivariate findings suggest that public attitudes toward the police are responsive to national presidential elections. A more rigorous test of this thesis is presented in the next section.
Analytical Model of Citizen Evaluation of Police In this section, we more fully test Hypotheses 1 and 2 by creating a multiple regression model that approximates the theoretical model that we presented earlier in this discussion. To account for our hypotheses and the previous acknowledged determinants of attitudes toward the police, we posit confidence in the police to be a function of these variables: government (national) support, support for other governmental institutions, age, race, partisanship, and political ideology. Confidence in the police is our dichotomous outcome variable, with citizens who express some or much confidence coded as confidence (1) and all other responses coded as no confidence (0).16 For the 1998 and 1999 U.S. samples, confidence is the response that the police are doing an ‘‘excellent’’ or ‘‘good’’ job. The University of New Mexico questionnaire does not use the confidence-in-police question to evaluate citizens’ attitudes toward the police. Instead this survey asks respondents to assess the performance of the police with responses of poor, fair, good, and excellent. While this assessment question differs from the confidence question, we believe our equating of these two questions is justified because both measures are predominantly measures of short-term satisfaction with the performance of the police.17 Moreover, as 15 Government confidence declines from 40 percent in 1998 to 23 percent in 2003, while police confidence is consistent (39 and 36 percent [w2 5 1.47 and p-value 5 0.22534]). 16 We model police support as a dichotomous rather than an ordered response variable for comparability. Few persons in Mexico express ‘‘much’’ or ‘‘a lot’’ of confidence in the police. Indeed, of the 1,200 Mexicans who respond to the police confidence question for 2003, only eight express confidence in the police. Given these few categorical observations, we choose to dichotomize the Mexican data. For comparability, we do the same for the U.S. and Costa Rica data. 17
See Gibson et al. (2003) for a discussion of institutional confidence.
352
Elections as Focusing Events
evidence of face validity, Table 2 shows that confidence-in-the-police and evaluation-of-police questions produce similar results for U.S. respondents.
The Explanatory Model Given the dichotomous outcome variable (confidence in the police), we use logistic regression to examine the determinants of citizens’ confidence in the police and posit the following analytical model: police confidence 5 government support 1 party identification 1 local government support 1 state government support 1 military support 1 race 1 age 1 attitude about crime 1 political ideology 1 attitude about order 1 city size.
Government support (confidence) is the primary independent variable. This variable is an ordinal variable with values of 1 to 4: 1 5 no confidence, 2 5 little confidence, 3 5 some confidence, and 4 5 much confidence. Once again, the University of New Mexico questionnaire does not use the confidence-in-government question to evaluate citizens’ attitudes toward the national government. This survey asks respondents to assess the performance of the federal government with responses of poor, fair, good, and excellent. As we mentioned for the police, Table 2 shows that confidence-in-thegovernment and evaluation-of-government questions produce similar results for U.S. respondents. Moreover, the 1998 and 1999 IPP use of these evaluative questions is consistent across the police and governmental variables. We include party identification as our second important independent variable and to test Hypothesis 3. We specify primary party-dummy variables in the following manner: Costa RicaFfirst party 5 PUSC, second party 5 PLN; MexicoF first party 5 PRI, second party 5 PAN, third party 5 PRD; United StatesFfirst party 5 Republican, second party 5 Democrat. Party codes are explained in fuller detail in the Appendix. We include other independent variables as statistical controls and as representations of the acknowledged determinants of police attitudes. To account for support for other governmental institutions, we include measures of local government, state government, and military support. Unfortunately, we lack measures of attitudes toward local government and the military for all eight distributions of the study.18 Nevertheless, we can control for the effect of attitudes toward the military on police confidence in four of our eight 18 There are three reasons for this deficiency: Costa Rica does not have a military; the World Values Survey does not ask the local confidence-in-government question; the University of New Mexico survey does not ask a military assessment question.
Walker & Waterman
353
distributions (Mexico 1998, 2000, and 2003; and United States 2000). Military confidence is measured on a four-point scale with 1 5 none, 2 5 little, 3 5 some, and 4 5 much. We can also control for attitudes toward local and state government in four of our eight distributions (United States 1998, 1999; Costa Rica 2003; and Mexico 2003). We use respondents’ evaluations of the mayor or city manager to model local government evaluation and respondents’ evaluations of the governor to model state government evaluation in the United States.19 For the 2003 Costa Rican and Mexican distributions, we use confidence in local officials as a measure of local government evaluation.20 Age is included in the model because it is a proven and resilient predictor of police attitudes in the United States, with older respondents more supportive of the police. Because we use national data, we lack measurements of neighborhood demographic effects. We do, however, include attitude about crime, which captures the concern that some individuals have about the local crime problem. While attitudes about crime may be related to local concerns, national candidates also seek to nationalize the issue. Recall that McCann and Lawson show that Mexicans’ attitudes toward crime control were responsive to campaign effects from the 2000 Mexican presidential campaign (2003:69). Interestingly, this panel study shows that attitudes moved from the ‘‘tough-on-crime’’ position to the ‘‘fight-crime-with-job-creation’’ position across the five months of the study. We also include attitude about order and size-of-respondents’-city (city size) in our model. Attitude about order taps into general attitudes about the democratic regime’s function and concerns about civil and political liberties. Moreover, Cao and Hou (2001) find a significant negative relationship between the flip side of order (what they term a deviant subculture) and confidence in the police. We believe that municipality size should be negatively related to confidence in the police because citizens in smaller municipalities have a greater opportunity to have a personal relationship with police officers (Cao & Zhao 2005:408). In addition, we control for attitude about corruption, political ideology, gender, education level, income, and political efficacy (see Appendix for a description of measurements). Because the variance is a function of the mean for the binomial distribution and susceptible to over-dispersion, we simulate the standard errors and use the z-score for the government support 19 Both measures are ordinal values with a range of 1 to 4: 1 5 poor, 2 5 fair, 3 5 good, 4 5 excellent. 20 The measure is ordinal with a range of 1 to 5: 1 5 none, 2 5 little, 3 5 no opinion, 4 5 some, 5 5 much.
354
Elections as Focusing Events
coefficient estimate as our primary test statistic and measure of the strength of the government/police support relationship.21 The z-score is the MLE coefficient estimate for the variable of interest divided by the simulated standard error estimate of the MLE coefficient estimate. The z-score has an approximate standard normal distribution, and a z-score value of 2 is fairly strong evidence of an effect. We are also interested in the overall strength of association of the police confidence models. The Pseudo R2, while not an actual measure of model fit in the generalized linear model (GLM) context, is a good comparative measure of the strength of association that approximates the function that the R2 plays as a measure of linear association in the ordinary least squares model. We use Nagelkerke’s (1991) version, which allows the Pseudo R2 to achieve a maximum value of 1. This approach to the R2 creates an R2 measure that is comparable across the eight logistic models of this study.
Findings Table 3 shows that even when we control for a variety of other factors, confidence in the national government (government support) is a significant factor in citizens’ assessment of the police in the three countries and all eight samples; that is, even as we control for attitudes toward other governmental institutions, race, age, attitude about crime, and attitude about order. Since past research finds that these factors are related to citizens’ attitudes toward the police, this finding offers solid evidence for our first hypothesis. Citizens positively and significantly link their support for the police to their support of the national government. The conditional effect of government support on police support is strongest for the Mexico 2000 distribution (z-score 5 19.4). Meanwhile, the strength of the government/police support relationship for Mexico 1998 and Mexico 2003 are similar (z-scores 5 8.08 and 6.31, respectively), but significantly lower, when we control for these other factors. Recall that 2000 is a presidential year. The Mexican distribution therefore offers support for our second hypothesis. The focusing event provided by the presidential election intensifies the connection between the national government and evaluation of the police when we control for other factors. Figure 1 plots the predicted probability of police support for centrist male individuals who identify with the first political party and have median support for other governmental institutions, 21 MLE estimation produces consistent coefficient estimates even when the variance is misspecified (Agresti 1996).
0.055 (0.144) [0.38] 0.053 (0.144) [ 0.37] 0.101 (0.088) [1.15] 1.857 (0.462) 205.58 o2e-16 0.253 980
0.771 (0.069) [11.17]
CR 1998 Pres coef (se) [z]
0.156 (0.063) [2.48] 2.890 (0.465) 135.21 o2e-16 0.168 1004
0.123 (0.136) [ 0.90]
0.227 (0.047) [4.83]
0.482 (0.063) [7.65]
CR 2003 coef (se) [z]
0.024 (0.129) [0.19] 0.437 (0.130) [ 3.36] 0.060 (0.079) [ 0.76] 0.646 (0.438) 171.10 o2e-16 0.183 1179
0.476 (0.059) [8.03] 0.327 (0.062) [5.27]
Mexico 1998 coef (se) [z]
0.065 (0.134) [0.49] 0.211 (0.134) [ 1.57] 0.099 (0.040) [ 2.48] 5.208 (0.483) 648.84 o2e-16 0.473 1485
1.452 (0.075) [19.36] 0.450 (0.062) [7.26]
Mexico 2000 Pres coef (se) [z]
0.030 (0.060) [ 0.50] 2.920 (0.453) 198.76 o2e-16 0.211 1200
0.730 (0.134) [ 5.45]
0.385 (0.061) [6.31] 0.247 (0.047) [5.26] 0.239 (0.048) [4.98]
Mexico 2003 coef (se) [z]
0.166 (0.045) [3.69] 2.639 (0.528) 283.86 o2e-16 0.329 1039
0.367 (0.054) [6.80] 0.541 (0.057) [9.49] 0.080 (0.152) [0.53]
0.314 (0.068) [ 4.62]
US 1998 coef (se) [z]
0.180 (0.046) [3.91] 3.405 (0.569) 254.28 o2e-16 0.338 873
0.339 (0.056) [6.05] 0.216 (0.058) [3.72] 0.252 (0.160) [1.56]
0.578 (0.075) [ 7.71]
US 1999 coef (se) [z]
0.061 (0.155) [0.39] 0.201 (0.156) [1.29] 0.042 (0.045) [0.93] 3.491 (0.575) 561.61 o2e-16 0.511 1177
1.163 (0.086) [13.52] 0.832 (0.074) [11.24]
US 2000 Pres coef (se) [z]
Pres 5 Presidential Year Model. Estimates for party identification, race and ethnicity, corruption and citysize are shown in Table 4. Model also controls for political ideology, gender, education, income, and efficacy. These variables are not included due to lack of space (see Appendix). Coef 5 Coefficient. Standard errors of coefficient estimates are simulated in the R statistical environment using Design Librarys lrm (logistic regression model) function in the R statistical environment. (se) 5 are simulated standard errors of MLE estimates. Data sources: 1998 Hewlett Survey; 2000 World Value Survey; 2003 ´metro; and 1998 and 1999 University of New Mexico Institute for Public Policy Survey. Latinobaro
LR-Chisq Prob4Chisq 2 Nagelkerke R Observations
Intercept
Age
Order
Crime
State support
Local support
Military support
Government support
Variable
Table 3. Logistic Regression Model of Police Support: Institutional Support Variables
Walker & Waterman
355
356
Elections as Focusing Events
Figure 1. Predicted Police Support by Government Support. The figure plots predicted probability of police support for a centrist male individual who identifies with the first political party and has median local support (3 in the United States and 2 in Costa Rica and Mexico), median state support (3 in the United States), median military support (3 in the United States and 2 in Mexico), median education, income, and efficacy by government support in the three respective countries.
median education, median income, and median efficacy by government support for the three respective countries. The first political party is the PRI in Mexico, the PUSC in Costa Rica, and the Republican Party in the United States. The ‘‘Mexico’’ plot of Figure 1 shows graphically the effect government support has on support for police and represents the predicted probabilities produced by the police confidence models. In 1998 (the solid line), centrist-PRI partisans’ support for the police is linearly related to their support for the government. In the 2000 presidential year (the dashed line), centrist-PRI partisans’ support for the police is more strongly related to their support for the government. By 2003 (the dotted line), centrist-PRI partisans’ support for the police is linearly related to their support for the government as it was in 1998. Note that the two lines are roughly parallel. Still, centrist-PRI partisans are far less likely to support the police than they were in 1998. This
Walker & Waterman
357
is convincing evidence that the intensity of the 2000 presidential election affected the government/police support relationship. Yet the 2000 Mexican presidential election was historic, and the increased connection between the government and the police could simply be a manifestation of the importance of this particular presidential election. On the other hand, the 2000 U.S. and 1998 Costa Rican models corroborate the thesis that presidential elections in general influence attitudes toward the police. For U.S. citizens, the governmental confidence z-score value increases from 4.62 in 1998 to 13.52 by the 2000 presidential election year. The ‘‘United States’’ plot of Figure 1 shows that in 1998, the probability that centrist U.S. male Republicans support the police increases linearly based on their level of support for the government (the solid line).22 By 2000, the probability that these same centrist U.S. male Republicans support the police increases more strongly as their level of support for the government increases (the dotted line). For Costa Rican citizens, the governmental confidence z-score value decreases from 11.17 in the 1998 presidential year to 7.65 in the nonpresidential year of 2003.23 The ‘‘Costa Rica’’ plot of Figure 1 shows the increase in the government support slope in the 1998 presidential year (the solid line). In both the U.S. and Costa Rican cases, the increase of the presidential year slope is consistent with the more dramatic Mexican presidential year increase. In addition, the d tuning ratio, which measures the strength of the conditional government support/police support relationship in the presidential year divided by the strength of the average conditional government support/police support relationship in nonpresidential years, is 1.3 for Costa Rica, 1.8 for the United States, and 2.3 for Mexico.24 As we stipulated in our articulation of the presidential-year police support model, this d ratio (the presidential-year z-score/the average nonpresidential year z-score) exceeds 1. In addition, the 2000 U.S., 2000 Mexican, and 1998 Costa Rican models explain 51, 47, and 25 percent of the variation in police confidence probabilities, respectively. This is significantly greater than the variation explained by the nonpresidential-year models, which include a local support variable. Even with the local support variable included in the 2003 Costa Rican model (an additional 22 The graphical linear effect is slight because ‘‘state support’’ is held constant at the median value of 3. State support has a stronger effect in 1998. This was a year when 36 states held gubernatorial elections, suggesting a gubernatorial election focusing event. 23 For comparability, we exclude the ‘‘local support’’ control from the 2003 Costa Rican model and generate a z-score 5 8.84 (b 5 0.539, se 5 0.063). 24 We generate these estimates using the z-score from police confidence models that exclude measures for military support, local government support, and state government support. These estimates are more conservative. Using the z-scores in Table 3, the estimates are 1.46 for Costa Rica, 2.2 for the United States, and 2.7 for Mexico.
358
Elections as Focusing Events
variable), the 1998 presidential-year Costa Rican model explains far more of the variation in police attitudes. Taken together, this is solid evidence in favor of Hypothesis 2. In Hypothesis 3, we hypothesize that partisans’ attitudes toward the police in developing democracies are more likely to shift due to change in partisan control of the national government. Given Mexico’s developing democracy status, we find support for our thesis in the ‘‘Mexico’’ plot of Figure 1. Note that PRI partisans are far less likely to support the police in 2003 than they were in 1998. In addition, Table 4 shows that none of the Mexican politicalparty dummies is related significantly to police confidence in 2000. By 2003, all other partisan identifiers, including the PRI, are less likely than PAN partisans (the new party of the government) to have confidence in the police. The change in the party of the government affects the attitudes that the PRI and other partisans have toward the police. In 2003, the PRI still had significant political power at the state and local levels. Nevertheless, PRI partisans have negative attitudes toward the police because the PAN controls the national-political government. The ‘‘Mexico’’ plot of Figure 1 shows the dramatic decline in PRI partisans’ support for the police from 1998 (the solid line) to 2003 (the dotted line). The police confidence probability of a PRI partisan with ‘‘much’’ support for the government declines from 0.67 in 1998 to 0.21 by 2003. By contrast, U.S. Republicans are more likely than their Democratic (the party of the president in 1998, 1999, and 2000) counterparts to support the police. In addition, we find no significant difference in the police support probabilities of Costa Rican PLN and PUSC partisans even when the party of the government changed from PLN in 1998 to PUSC in 2003. It is also worth noting that the PUSC candidate won the presidency in 2002. Nevertheless, there is little difference in the mean probability of PUSC support for the police across this fiveyear period. This is solid evidence in support of Hypothesis 3. Why do the attitudes of Mexican partisans shift so dramatically, while the attitudes of U.S. and Costa Rican partisans remain in a fixed configuration? By fixed, we mean that Republicans have consistently stronger support for the police than do Democrats, even when Democrats controlled the national government. Wilson (1963) points out that political institutions at the local level must attain a certain (nonpartisan) character for the professional police force to exist. Our partisanship findings suggest that a similar type of system relationship must develop at the national level. The police in Costa Rica and the United States are a part of a usable state bureaucracy that operates within professional norms (Linz & Stepan 1996:55). The unconsolidated nature of the set of Mexican national civil and political institutions prompts Mexican partisans
2.24 (0.053) [ 4.23] 0.213 (0.064) [ 3.33]
0.539 (0.178) [3.03]
0.450 (0.177) [2.54] 0.000
0.132 (0.048) [2.75] 0.073 (0.033) [ 2.21]
0.210 (0.195) [ 1.08] 0.257 (0.194) [ 1.32]
0.191 (0.187) [1.02]
0.070 (0.178) [ 0.39]
0.132 (0.174) [ 0.76]
0.000
C R 2003 coef (se) [z]
0.000
CR 1998 Pres coef (se) [z]
0.312 (0.179) [1.74] 0.766 (0.198) [ 3.87] 0.140 (0.046) [ 3.04] 0.160 (0.046) [ 3.48]
0.005 (0.186) [0.03] 0.000
0.123 (0.183) [ 0.67] 0.090 (0.177) [ 0.51] 0.438 (0.182) [ 2.41]
0.000
Mexico 1998 coef (se) [z]
0.089 (0.029) [ 3.07]
0.058 (0.163) [0.36] 0.348 (0.168) [0.36]
0.000
0.062 (0.211) [0.29] 0.084 (0.213) [ 0.39] 0.094 (0.215) [0.44] 0.252 (0.213) [0.90]
0.000
Mexico 2000 Pres coef (se) [z]
0.123 (0.047) [2.62] 0.013 (0.030) [0.43]
0.653 (0.217) [ 3.01] 0.495 (0.208) [ 2.38] 0.801 (0.209) [ 3.83]
0.878 (0.208) [ 4.22] 0.000
Mexico 2003 coef (se) [z]
1.451 (0.190) [ 7.64] 0.571 (0.167) [ 3.42]
0.000
0.374 (0.179) [ 2.99]
0.601 (0.189) [ 3.18]
0.000
US 1998 coef (se) [z]
1.234 (0.200) [ 6.17] 0.189 (0.190) [0.99]
0.000
0.612 (0.197) [ 3.11]
0.456 (0.195) [ 2.34]
0.000
US 1999 coef (se) [z]
0.032 (0.040) [ 0.80]
0.606 (0.154) [ 3.94]
0.000
0.948 (0.188) [ 5.04]
1.462 (0.202) [ 7.24]
0.000
US 2000 Pres coef (se) [z]
Notes: Party Codes: Costa RicaFFirst Party 5 PUSC; Second Party 5 PLN; MexicoFFirst Party 5 PRI; Second Party 5 PAN; Third Party 5 PRD; USF First Party 5 Republican; Second Party 5 Democrat. These variables are not included due to lack of space (see Appendix). Coef 5 Coefficient. Standard errors of coefficient estimates are simulated in the R Statistical environment using Design Library’s lrm (logistic regression model) function in the R Statistical environment. (se) 5 are simulated standard errors of MLE estimates. Data sources: 1998 Hewlett Survey; 2000 World Value Survey; 2003 ´metro; and University of New Mexico Institute for Public Policy Survey. Latinobaro
City size
Corruption
Other/Hispanic
Negro/Black
Mestizo
Race and Ethnicity Blanco/White
No party
Other party
Third party
Second party
Party Identification First party
Variable
Table 4. Police Support Models: Party Identification and Race and Ethnicity
Walker & Waterman
359
360
Elections as Focusing Events
to change their level of support for the police based on the partisan composition of the national political government. Control of the executive branch infers partisan control of the coercive force of the democratic state. In short, many Mexican citizens believe that the legal system of the state (including the police) is a tool of the dominant political party. In the aftermath of the loss of control of the national government, PRI partisans become less likely to support the police than do PAN (the new party of the government) partisans. Our partisan findings suggest that police attitudes reflect on the character of the democratic regime and support Hypothesis 3.
Conclusion We tend to think of the police as a local institution, and generally also as a nonpartisan one, which performs a specific function: to serve and protect its citizenry. We find strong cross-national attitudinal support for this local support thesis. Other research indicates that attitudes toward national governmental structures also influence attitudes toward the police. We extend this research and show that perceptions of the police are responsive to systematic fluctuations in attitudes toward the national government. We show that presidential elections, with their greater focus on crime-related issues, provide a focusing event at the national level that influences citizen perceptions of the police at the local level. The determinants of public attitudes toward the police shift during the presidential election year because the attitude is salient, provides a partisan advantage for a political party, and relates to the voting decision. In short, the attitude toward the police is politicized by the national presidential election. A national presidential election is a periodic change in the character of the national government that impacts attitudes toward the police. The election as a focusing event intensifies the police support/government support relationship for citizens in both the developing Mexican democracy and the full democracies of Costa Rica and the United States. Still, we find an important difference in the police attitudes of citizens in these two types of democracy. Mexicans’ support for the police shifts based on changes in the partisan composition of the national political government, while Costa Ricans and U.S. citizens do not shift their attitudes about the police based on changes in the partisan composition of the national government. Nevertheless, candidates in full democracies believe that ‘‘borrowing’’ the legitimacy of the police offers an electoral advantage. Indeed, we see signs of an emphasis on law-and-order themes in the 2008 U.S. presidential election. The Republican presidential
Walker & Waterman
361
candidate Rudolph Giuliani adopted a ‘‘top cop’’ persona as the centerpiece of his primary election campaign. His emphasis was not only his experience in the aftermath of the September 11, 2001, terrorist attack, but also the idea that he alone of the candidates could keep America safe. In addition, former Tennessee Senator Fred Thompson ascended to the upper tier of candidates in the Republican primary election campaign partly on his notoriety as the ‘‘tough-on-crime’’ district attorney on the popular television series Law & Order. On the Democratic side, New York Senator Hillary Clinton also emphasized experience and toughness, in both foreign and domestic policy. Although tough rhetoric on foreign policy is at the forefront of campaigns in a nation at war, law-and-order themes at home remain the essential bread and butter of the campaign. Merely by earning the endorsement of a police organization or standing for a photo opportunity with a group of police officers provides support for the message; this candidate is tough on crime and is tough enough to be president. Therefore, the 2008 presidential race emphasizes the same themes, and we expect the same sort of bump in public support for the police, as a result. For developing democracies, and building on Cao and Zhao’s (2005) findings, our findings suggest that police attitudes are an important barometer of attitudes about the development of state apparatus autonomy. Moreover, we speculate that a similar effect should exist for attitudes about other areas of the bureaucratic state, most particularly the judiciary. While our study involves three countries, it suggests that a broader study that could include other Latin American nations, Canada, and Europe, for example, would provide a more thorough test of our focusing event thesis and its relation to state autonomy. In the meantime, the results here provide strong support for the idea that we should conceptualize the police not merely as local actors operating on a limited stage, but as political actors who perform on a broad national stage.
Appendix: Independent Variables Partisanship (dummies): Costa Rica: PUSC (Partido Unidad ´n Nacional); MexSocialcristiana), PLN (Partido Liberacio ico: PRI (Partido Revolucionario Institucional), PAN (Part´n Nacional), PRD (Partido de la Revolucio ´n ido Accio Democra´tica). Political ideology: dummy variables from 10-point left/right ideology scale. 1–3 5 left, 4–6 5 center, 7–10 5 right (base category), 11 (none, not sure) 5 no ideology, 12–13 (no response) 5 no ideology response.
362
Elections as Focusing Events
Corruption: for Costa Rica and Mexico 1998: 1 5 almost no one, 2 5 few, 3 5 not sure, 4 5 many, 5 5 almost everyone. For Costa Rica and Mexico 2003: 1 5 no progress, 2 5 little, 3 5 don’t know, 4 5 some, 5 5 much. Crime: dummy variableFcombating crime is the most important function of government. Order: dummy variableFmaintaining order is the most important function of government. Efficacy: for 2000 data, efficacy is a count variable measured from 0 to 3. 1 for each positive response to three questions: sign a petition, attend a demonstration, or join in boycott. 1998 data: dummy for individuals who say they are willing to personally do something to hold government accountable. For 1998 and 1999 U.S. data: measured 0 to 7. City size: 1 5 fewer than 5,000; 2 5 5,000 to 10,000; 3 5 10,000 to 20,000; 4 5 20,000 to 40,000; 5 5 40,000 to 50,000; 6 5 50,000 to 100,000; 7 5 capital. Race: dummy variables. Age: for 2003 data, age is measured as follows: 1 5 16–25, 2 5 26–40, 3 5 41–60, 4 5 611. For 1998: 1 5 15–34, 2 5 35–54, 3 5 551. For 2000 World Values Data: 1 5 15–24, 2 5 25–34, 3 5 35–44, 4 5 45–54, 5 5 55–64, 6 5 651. For 1998 and 1999 U.S. data: 1 5 18–29, 2 5 30– 39, 3 5 40–49, 4 5 50–64, 5 5 651. Education: for 2003 data: 1 5 illiterate, 2 5 primary incomplete, 3 5 primary complete, 4 5 secondary incomplete, 5 5 secondary complete, 6 5 superior incomplete, 7 5 superior complete. For 2000 World Values data: 1 5 lower, 2 5 middle, 3 5 upper. For 1998 data: 1 5 no school, 2 5 primary, 3 5 secondary, 4 5 superior. For 1998 and 1999 U.S. data: 1 5 low, 6 5 high. Income: for 2003 data: 1 5 very bad, 2 5 bad, 3 5 average, 4 5 good, 5 5 very good. For 2000 World Values data: 0 5 no response, 1 5 lower, 2 5 middle, 3 5 upper. For 1998 Costa Rica and Mexico data and 1998 and 1999 U.S. data: 0 5 no response, 1 5 low, 2 5 middle low, 3 5 middle high, 4 5 high.
References Agresti, Alan (1996) An Introduction to Categorical Data Analysis. New York: John Wiley. Albrecht, Stan L., & Miles Green (1977) ‘‘Attitudes toward the Police and the Larger Attitude Complex: Implications for Police-Community Relationships,’’ 5 Criminology 67–86.
Walker & Waterman
363
Benson, Paul R. (1981) ‘‘Political Alienation and Public Satisfaction with Police Services,’’ 24 Pacific Sociological Rev. 24–64. Birkland, Thomas A. (1997) After Disaster: Agenda Setting, Public Policy and Focusing Events. Washington, DC: Georgetown Univ. Press. Bordua, David J., & Larry L. Tifft (1971) ‘‘Citizens Interviews, Organizational Feedback, and Police-Community Relations Decisions,’’ 6 Law & Society Rev. 155–82. Brown, Ben, & Wm Reed Benedict (2002) ‘‘Perceptions of the Police: Past Findings, Methodological Issues, Conceptual Issues and Policy Implications,’’ 25 Policing-An International J. of Police Strategies and Management 543–80. Brown, Ben, et al. (2006) ‘‘Public Perceptions of the Police in Mexico: A Case Study,’’ 29 Policing-An International J. of Police Strategies and Management 158–75. Call, Charles T. (2002) ‘‘War Transitions and the New Civilian Security in Latin America,’’ 35 Comparative Politics 1–32. Camp, Roderic Ai (1999) Politics in Mexico: The Decline of Authoritarianism. New York: Oxford Univ. Press. Canache, Damarys, et al. (2001) ‘‘Meaning and Measurement in Cross-National Research on Satisfaction with Democracy,’’ 65 Public Opinion Q. 506–28. Cao, Liqun, & Charles Hou (2001) ‘‘A Comparison of Confidence in the Police in China and in the United States,’’ 29 J. of Criminal Justice 87–99. Cao, Liqun, & Jihong Solomon Zhao (1998) ‘‘Public Confidence in the Police: A Comparative Study between Japan and America,’’ 26 J. of Criminal Justice 279–89. FFF (2005) ‘‘Confidence in the Police in Latin America,’’ 33 J. of Criminal Justice 430–51. Collier, David, & Steven Levitsky (1997) ‘‘Democracy with Adjectives: Conceptual Innovation in Comparative Research,’’ 49 World Politics 403–12. Correia, Mark E., et al. (1996) ‘‘Public Perceptions of State Police: An Analysis of Individual-Level and Contextual Variables,’’ 24 J. of Criminal Justice 17–28. Dahl, Robert (1989) Democracy and Its Critics. New Haven, CT: Yale Univ. Press. Gibson, James L., et al. (2003) ‘‘Measuring Attitudes toward the Supreme Court,’’ 47 American J. of Political Science 354–67. Godwin, Marcia L., & Jean Reith Schroedel (2000) ‘‘Policy Diffusion and Strategies for Promoting Policy Change: Evidence from California Local Gun Control Ordinances,’’ 28 Policy Studies J. 760–76. Howell, Susan E., et al. (2004) ‘‘Black Cities/White Cities: Evaluating the Police,’’ 26 Political Behavior 45–68. Kessel, John (1980) Presidential Campaign Politics: Coalition Strategies and Citizen Response. Homewood, IL: The Dorsey Press. Kingdon, John W. (1995) Agendas, Alternatives, and Public Policies. New York: Longman. Lasley, James R. (1994) ‘‘Ethnicity, Gender, and Police-Community Attitudes,’’ 75 Social Science Q. 85–97. Levitt, Steven D. (1997) ‘‘Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime,’’ 87 The American Economic Rev. 270–90. Linz, Juan J., & Alfred Stepan (1996) Problems of Democratic Transitions and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: The John Hopkins Univ. Press. ´pez-Montiel, Angel Gustavo (2000) ‘‘The Military, Political Power, and Police RelaLo tions in Mexico City,’’ 111 Latin American Perspectives 79–94. McCann, James A., & Chappell Lawson (2003) ‘‘An Electorate Adrift? Public Opinion and the Quality of Democracy in Mexico,’’ 38 Latin American Research Rev. 60–95. Nagelkerke, Nico J. D. (1991) ‘‘Miscellanea: A Note on a General Definition of the Coefficient of Determination,’’ 78 Biometrika 691–2. O’Donnell, Guillermo (2001) ‘‘Democracy, Law, and Comparative Politics,’’ 36 Studies in Comparative International Development 7–36.
364
Elections as Focusing Events
Olivero, J. Michael, & Rodrigo Murataya (1998) ‘‘Citizen Satisfaction with Police Services in Guadalajara, Mexico,’’ 22 International J. of Comparative and Applied Criminal Justice 303–10. Pereira, Anthony W. (2001) ‘‘Virtual Legality: Authoritarian Legacies and the Reform of Military Justice in Brazil, the Southern Cone, and Mexico,’’ 34 Comparative Political Studies 555–74. Przeworski, Adam, & Henry Teune (1970) The Logic of Comparative Social Inquiry. New York: John Wiley. Reisig, Michael D., & Roger B. Parks (2000) ‘‘Experience, Quality of Life, and Neighborhood Context: A Hierarchical Analysis of Satisfaction with Police,’’ 17 Justice Q. 607–30. Reiss, Albert (1967) Studies in Crime and Law Enforcement in Major Metropolitan Areas. Washington, DC: U.S. Government Printing Office. Sampson, Robert J., & Dawn Jeglum Bartusch (1998) ‘‘Legal Cynicism and Subcultural Tolerance of Deviance: The Neighborhood Context of Racial Differences,’’ 32 Law & Society Rev. 777–804. Schumpeter, Joseph (1975) Capitalism, Socialism, and Democracy. New York: Harper. Seron, Carroll, et al. (2004) ‘‘Judging Police Misconduct: ‘Street-Level’ versus Professional Policing,’’ 38 Law & Society Rev. 665–710. Shaw, Greg M., et al. (1998) ‘‘The Polls Trends: Crime, the Police, and Civil Liberties,’’ 62 Public Opinion Q. 405–26. Shipan, Charles R., & Craig Volden (2006) ‘‘Bottom-Up Federalism: The Diffusion of Antismoking Policies from US Cities to States,’’ 50 American J. of Political Science 825–43. Sigelman, Lee, et al. (1997) ‘‘Police Brutality and Public Perceptions of Racial Discrimination: A Tale of Two Beatings,’’ 50 Political Research Q. 777–91. Smith, Douglas A. (1986) ‘‘The Neighborhood Context of Police Behavior,’’ in A. J. Reiss & M. Tonry, eds., Communities and Crime. Chicago: Univ. of Chicago Press. Sunshine, Jason, & Tom R. Tyler (2003) ‘‘The Role of Procedural Justice and Legitimacy in Shaping Public Support for Policing,’’ 37 Law & Society Rev. 513–47. Terrill, William, & Michael D. Reisig (2003) ‘‘Neighborhood Context and Police Use of Force,’’ 40 J. of Research in Crime Delinquency 291–321. Tuch, Steven A., & Ronald Weitzer (1997) ‘‘The Polls Trends: Racial Differences in Attitudes Toward the Police,’’ 61 Political Opinion Q. 642–63. Tyler, Tom R. (1988) ‘‘What Is Procedural Justice? Criteria Used by Citizens to Assess the Fairness of Legal Procedures,’’ 22 Law & Society Rev. 103–35. Weitzer, Ronald (1995) Policing Under Fire: Ethnic Conflict and Police-Community Relations in Northern Ireland. Albany, NY: SUNY Press. FFF (2000) ‘‘Racialized Policing: Residents Perceptions in Three Neighborhoods,’’ 34 Law & Society Rev. 129–55. Wilson, James Q. (1963) ‘‘The Police and Their Problems: A Theory,’’ 12 Public Policy 189–216. Whyte, William Foote (1943) Street Corner Society: The Social Structure of an Italian Slum. Chicago: Univ. of Chicago Press. Zamble, Edward, & Phyllis Annesley (1987) ‘‘Some Determinants of Public Attitudes toward the Police,’’ 15 J. of Police Science and Administration 285–90.
Lee Demetrius Walker is an Assistant Professor in the Department of Political Science at the University of South Carolina. Before joining the University of South Carolina, he was an Assistant Professor of political science at the University of Kentucky (2003 to
Walker & Waterman
365
2006) and Visiting Scholar at the Institute of Quantitative Social Science at Harvard University (2006). He has published articles in the Journal of Politics, Party Politics, and Latin American Research Review. Richard W. Waterman is a Professor of political science at the University of Kentucky. He has published eight books and published articles in such journals as the American Political Science Review, the American Journal of Political Science, and the Journal of Politics. His research interests include the presidency, the bureaucracy, the environment, and the courts.
366
Elections as Focusing Events
367
Popular Constitutionalism’s Hard When You’re Not Very Popular: Why the ACLU Turned to Courts
Emily Zackin
Through a case study of the early American Civil Liberties Union (ACLU), this article examines the empirical ramifications of constitutional scholars’ recent exhortations to ‘‘take the Constitution away from the courts’’ in order to promote democratic deliberation about constitutional meaning. While it is now one of the most prominent examples of a litigation-based interest group, the ACLU began its existence demonstrating a commitment to constitutionalism outside the courts. Through coding a decade’s worth of meeting minutes and examining archival sources, I demonstrate that the ACLU’s mounting unpopularity rendered extrajudicial politics impossible, precipitating the ACLU’s shift toward litigation. The ACLU’s move toward litigation, despite its early devotion to political activism outside the courts, suggests that it is not always possible for political actors to make constitutional arguments without courts. Furthermore, the ACLU’s use of courts to publicize and dramatize its constitutional arguments demonstrates that litigation may actually promote popular deliberation about constitutional meaning. These political realities both highlight and contradict two empirical assumptions underlying arguments about the normative desirability of restricting courts’ involvement in constitutional politics. First, the state is not a neutral arena in which all political actors are equally free to pursue their constitutional visions through majoritarian processes. Second, courts may facilitate (rather than hinder) popular deliberation about constitutional questions.
I
n recent years, prominent constitutional law scholars have made arguments for greater popular participation in determinations about the U.S. Constitution’s meaning. Kramer and Tushnet have offered two of the most prominent arguments for restricting the court’s role in determining the meaning of the Constitution. Kramer emphasizes popular conclusions about the Constitution, while Tushnet stresses congressional deliberation. Yet these authors share the central conviction that courts’ active involvement in
I am grateful to Keith Whittington, Paul Frymer, Hendrik Hartog, and Kenneth Kersch for their advice and direction on this article. Daniel Linke and the entire staff of the Seeley Mudd Manuscript Library were also very generous with their time and assistance. Carroll Seron and anonymous reviewers provided helpful feedback and support. I also owe many thanks to Justin Crowe, Roban Kramer, and Tiffany Lennon for their careful readings and thoughtful comments. Please address correspondence to Emily J. Zackin, Princeton University Department of Politics, 130 Corwin Hall, Princeton, NJ 08544-1013; e-mail: ezackin @princeton.edu.
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
368
Popular Constitutionalism’s Hard
constitutional politics discourages and distorts the participation of nonjudicial actors (Tushnet 1999; Kramer 2004). The main thrust of both arguments is that American politics would be more democratic, and therefore better, if courts played an extremely limited role in determining how Americans understand and apply their fundamental law. Thus these authors argue for the normative desirability not only of constitutional debate outside the courts, but also of constitutionalism without the courts. This article does not participate directly in the normative debate about the value of popular and judicial constitutional interpretations. Instead, it is one of the growing number of studies that examines how nonjudicial actors, and even nonlawyers, make arguments about the meaning of the Constitution and make these arguments outside of courts. However, by focusing on why the American Civil Liberties Union (ACLU) might seek to use courts, this study uncovers (and challenges) some of the empirical assumptions upon which Tushnet and Kramer’s normative arguments rest. For instance, Tushnet advocates that interest groups engage in the politics of pluralism, rather than litigation, and seems to assume that it is possible for even unpopular or subversive groups to participate. Both Kramer and Tushnet also suppose that litigation can only hamper popular deliberation about the Constitution, not stimulate it. I do not wish to argue that popular deliberation or pluralist politics are normatively undesirable, or even empirically impossible. Instead, I want to examine organizations’ move to litigation in light of the historical and institutional realities that these groups may face. Historical circumstances may place groups squarely on the wrong side of public opinion and, as a consequence, they may find that political institutions other than courts are effectively unavailable as avenues for advancing their political arguments. The ACLU’s history is instructive in comparing such political realities to the empirical assumptions of scholars such as Tushnet and Kramer. While it is now one of the most prominent examples of a litigation-based interest group, it began its existence demonstrating a significant commitment to constitutionalism outside the courts. In fact, during the first decade of its existence, the ACLU began to litigate largely in response to its mounting unpopularity both within and without the government. Thus this case study suggests that, while constitutional activism without courts may prove successful some of the time for some groups, others will find themselves unable to pursue their constitutional visions without using courts. It is certainly possible that the normative dangers of judicial review may outweigh the problems associated with the suppression of unpopular arguments, but an empirical assessment of the challenges that unpopular minorities face when they pursue
Zackin
369
constitutional change without courts as well as the role courts may play in popularizing constitutional debate provide a useful caveat to arguments opposing constitutional litigation.
Constitutionalism Outside the Courts The existence of constitutional debate outside the courts has garnered increasing scholarly attention. Some studies demonstrate that formal political institutions other than courts make decisions about the Constitution’s meaning and its implications for politics. Some, such as Pickerill, describe the way that Congress grapples with the Constitution’s meaning, which it does largely in response to Supreme Court decisions about the constitutionality of federal legislation (Pickerill 2004). Though courts are largely irrelevant to the production of constitutional meaning that Whittington (1999) describes, his study of constitutional politics outside the courts, like Pickerill’s, focuses on debates among elected officials and members of formal political institutions. Whittington notes that, while courts are primarily concerned with interpreting the text of the Constitution, nonjudicial actors often engage in a different endeavor, constructing meaning from parts of the text that are otherwise indeterminate (Whittington 1999). Though both of these scholars note the existence of constitutional debates outside of courts, neither argues that courts should cease to engage in these debates as well. Tushnet (1999) differs on this point. Like Pickerell and Whittington, Tushnet sees Congress as a potential site of constitutional deliberation. However, unlike these scholars, Tushnet advances an argument about the normative desirability of constitutional interpretation inside institutions such as Congress and the consequent undesirability of judicial review. In Tushnet’s view, the active exercise of judicial review hinders elected representatives’ abilities to deliberate about the meaning of the Constitution. He is particularly concerned that democratic bodies interpret the Constitution’s sweeping and ambiguous statements of principle, which he terms the ‘‘thin Constitution.’’ While the ‘‘thick Constitution’’ contains detailed, mundane, or self-enforcing provisions, the ‘‘thin Constitution’’ contains ‘‘fundamental guarantees of equality, freedom of expression, and liberty’’ (Tushnet 1999:11). Tushnet argues that because judicial review is exercised so actively and justices’ views are treated as the final word on constitutional meaning, the practice of judicial review (at least relating to matters of fundamental principle) denies legislators the political incentive to think seriously about the Constitution’s content. He reasons that legislators have not assumed their proper roles as the arbiters of constitutional
370
Popular Constitutionalism’s Hard
meaning because the Supreme Court’s prominence in determining that meaning has convinced voters that they ought not hold their congressional representatives accountable for doing so (Tushnet 1999:66). Furthermore, he fears that Congress may not even understand what the Court has ruled or may design legislation primarily to preempt unfavorable Court decisions. He speculates that such dynamics might result in the production of distorted or inferior legislation (Tushnet 1999:58–9). In the ‘‘thought experiment’’ through which he advances his argument about the desirability of eliminating judicial review, Tushnet suggests that the national project to make constitutional meaning may well be better served if courts were deprived of the authority to overturn the constitutionality of legislation. At the very least, he argues, ‘‘Doing away with judicial review would have one clear effect: It would return all constitutional decision-making to the people acting politically. It would make populist constitutional law the only constitutional law there is’’ (Tushnet 1999:154). Although Tushnet’s earlier work focused on the National Association for the Advancement of Colored People’s (NAACP) litigation campaign on behalf of minority rights (Tushnet 1987), he nonetheless suggests that minority groups, such as the NAACP, both can and should advance their agendas through legislative bargaining, rather than litigative, means (Tushnet 1999:158–9). Other studies have focused on the constitutional interpretations offered by social movements and constitutional conclusions drawn by the general electorate. For instance, Ackerman argues that in extraordinary circumstances the citizenry as a whole will consider and decide constitutional questions, registering their decisions primarily through the election of political representatives (Ackerman 1991). Similarly, Siegel demonstrates that social movements have had significant effects on judicial interpretations of the Constitution. Siegel even sees the exchange between judges and advocacy groups as a necessary and productive mechanism through which constitutional meaning is shaped (Siegel 2001). Like Ackerman and Siegel, Kramer emphasizes popular capacity to interpret the Constitution. Yet unlike these scholars, Kramer argues that the active exercise of judicial review is destructive to the people’s capacity to interpret the Constitution. Like Tushnet, Kramer is particularly disturbed by the assertion that court rulings ought to trump populist or popular constitutional interpretations. Instead of emphasizing the role of the legislature, he argues that ordinary citizens must claim the primary responsibility for interpreting their highest law. Furthermore, Kramer argues that because American citizens now accept judicial interpretations as binding, they have abdicated responsibility for interpreting the Constitution and have therefore ceased to consider its
Zackin
371
meaning. He calls for a renewed effort on the part of the citizenry to curb the court in order for citizens to reclaim their capacity and fulfill their obligation to determine the Constitution’s meaning. Though Kramer does not advocate the abolition of judicial review, he does argue that people should resist constitutional interpretations that they dislike, not only through their own interpretations, but also through political means. He writes, ‘‘Justices can be impeached, the Court’s budget can be slashed, the president can ignore its mandates, Congress can strip it of jurisdiction or shrink its size or pack it with new members or give it burdensome new responsibilities or revise its procedures’’ (Kramer 2004:249). Kramer emphasizes the people’s interpretations of the Constitution, while Tushnet stresses congressional interpretations. Yet these authors share the central conviction that courts’ active involvement in politics is damaging to democracy, and therefore normatively unappealing. The main thrust of both arguments is that American politics would be more democratic, and therefore better, if courts played a much more limited role in constitutional politics. Thus they call on ‘‘the people’’ to both consider themselves the proper arbiters of constitutional meaning and to ‘‘take the Constitution away from the courts.’’ In effect, they are advocating for constitutional politics to be conducted not only outside courts, but largely without the courts. In the article that follows, I argue that the normative arguments of both Tushnet and Kramer rely on empirical assumptions that are often unfounded. Tushnet’s argument for constitutionalism among elected representatives assumes that these representatives constitute a neutral, pluralist state, but the existence of such a state is belied by the ACLU’s experience. In addition, both Tushnet’s and Kramer’s calls for purely popular constitutionalism assume that litigation stifles popular debate. However, the ACLU’s early use of litigation suggests that litigation may actually stimulate such debate. Thus by studying an early instance of popular constitutionalism, this article highlights the flawed empirical assumptions of those who would severely restrict the courts’ role in constitutional interpretation.
The ACLU and Constitutionalism Outside the Courts Despite its now-famous tradition of litigation, the ACLU was founded predominantly by activists who professed and demonstrated a strong commitment to both institutional and grassroots forms of constitutionalism outside the courts. The story of the ACLU’s founding begins in 1914, when Americans began to debate the wisdom of entry into the European war. Those who opposed participation formed a host of antiwar organizations. The most
372
Popular Constitutionalism’s Hard
prominent of these was founded at the end of 1915 by the prominent Internationalist and lawyer Crystal Eastman, the Settlement House founder Lillian Wald, and the nationally known Progressive leader Paul Kellogg. This group was also associated with wellknown Progressives such as Hull House founder Jane Addams and, later, Roger Baldwin, a young social worker making a name for himself in St. Louis. It adopted the name American Union Against Militarism (AUAM). Thanks to its illustrious members, the AUAM was well-respected nationally and wielded considerable political clout. Despite the efforts of the AUAM, Congress declared war in April 1917 and instituted a draft in May. The day after Congress passed the Selective Service Act, Baldwin began to organize a special bureau of the AUAM to defend those who would choose not to fight (D. Johnson 1963:18). The bureau began to refer to its charges as ‘‘conscientious objectors’’ and to itself as the Conscientious Objectors’ Bureau. It prepared to defend men from the legal consequences of refusing to comply with the Selective Service Act. From the outset, the Conscientious Objectors’ Bureau argued that draft laws were unconstitutional and, in July 1917, it expanded its mission to the wartime defense of all personal liberties, renaming itself the National Civil Liberties Bureau (NCLB) (D. Johnson 1963:21). The NCLB became independent of the AUAM in October 1917 and began to distance itself from its pacifist and internationalist roots (Witt 2004:46–8). Under Baldwin’s leadership, the organization adopted a new focus on the rights of labor rather than on conscientious objectors and renamed itself the American Civil Liberties Union (ACLU). Though it began to represent labor leaders with more frequency after 1920, the organization maintained the same general character throughout its name change (D. Johnson 1963:146). Consequently, this analysis treats the NCLB and ACLU as a single organization. These commitments of the early ACLU were strongly influenced by the aims and experiences of the Progressive movement, which emphasized the need for government regulation to protect society from the moral and physical consequences of industrialization and capitalism. The progressive intellectuals of the early twentieth century were generally skeptical of individual rights and tended to associate the protection of civil liberties such as free speech with their more salient concerns about economic and social justice (Graber 1991). For instance, when the AUAM’s Executive Committee first discussed establishing a bureau ‘‘for the maintenance of American liberties in wartime,’’ in June 1917, it tied these liberties directly to the importance of improved working conditions. An AUAM press release explained, ‘‘Part of the maintenance of liberty is the preservation of . . . labor laws . . . An eight-hour law,
Zackin
373
a full crew law, a child labor law, a Sherman ActFthese things are more than fine phrases about freedom. They are the fruit of self government,FThey are democracy made real in the lives of the people’’ (AUAM Executive Committee 1917: n.p.). Despite this rhetoric, the NCLB devoted most of its wartime energies to the novel tasks of advocacy for and defense of conscientious objectors. In this work, the organization’s tactics reflected its leaders’ orientation toward constitutional deliberation outside of the courts. The minutes of the ACLU’s executive committee describe its efforts to coordinate letter-writing campaigns and petitions, stage mass meetings and protests, and sponsor lecture tours by prominent liberal intellectuals. The NCLB employed many of these tactics in an effort to secure amnesty for conscientious objectors who had been jailed during the war and for those who had been jailed under the Espionage and Sedition Acts for the content of their speech. Throughout its early years, the ACLU also published a large volume of pamphlets, aimed at creating popular support for conscientious objection and the exercise of other civil liberties. This approach can be seen as an outgrowth of Progressive Era calls for popular constitutionalism (Kramer 2004:215). Many of the Progressive Era’s left-leaning political figures defined their views on courts in direct opposition to the Supreme Court’s famous Lochner decision,1 concluding that democratic majorities ought to determine the Constitution’s meaning in light of societal needs rather than individual freedoms. In addition to promoting popular activism, the ACLU also made constitutional arguments to elected representatives. This version of constitutionalism outside the courts can also be understood in light of Progressive politics. Progressives argued that democratically elected governments should play a strong role in regulating industry and improving the living conditions of the urban poor. Consequently, they stressed the social benefits of strong and interventionist government and attempted to work alongside government officials in their social improvement projects. While Progressives are known for their admiration of social science and expertise, the general antipathy of judges to the protective legislation that Progressives valued made some of the organization’s founders quite leery of courts and judges. Accordingly, the ACLU attempted to make constitutional arguments in Congress 1 In Lochner (Lochner v. New York 1905), the Court struck down a New York statute that capped the workday of bakers at 10 hours. The Court ruled that the statute violated the individual’s freedom of contract and that the state’s right to legislate could not supersede the individual’s right to contract freely. The statute in question was typical of a host of legislation passed at the turn of the century to protect workers from the dangers of industrialization, and the ruling typical of a host of judicial decisions striking down such legislation.
374
Popular Constitutionalism’s Hard
and to work alongside government agencies as counselors and allies. For instance, when the Espionage Act was being debated in Congress, the organization put a great deal of effort into Congressional lobbying, sending three lobbyists from its small full-time staff to Washington (Walker 1990:19). Even in the work of defending people whom the government had forced into military service, the ACLU described itself as a friend and advisor of President Woodrow Wilson’s administration. Eastman explains the plan to ‘‘endeavor at every point (through advising the War Department at Washington) . . . to aid the President in so administering the law that it may become in effect merely an efficient organizing of the fighting forces of the nation, not as a means of forcing men into the army against their conscience. Thus we might really help, and not hinder, the administration’’ (Eastman 1917:3). Baldwin wrote a letter to the Secretary of War, Newton Baker, stating, ‘‘We are entirely at the Service of the War Department,’’ and told Baker’s assistant that ‘‘We don’t want to make a move without consulting you’’ (as quoted in Walker 1990:18). Throughout its early years, the ACLU continued negotiating with government officials about the meaning of civil liberties (Baldwin 1920a). From its inception, the NCLB was also active in the courts. Affiliated attorneys defended conscientious objectors at their trials and before their draft boards. Yet simultaneously, the members of the New York staff worked behind the scenes with government officials to alleviate their clients’ sentences and secure them pardons. The ACLU’s initial relationship to courts and the law can also be understood in terms of its roots in the Progressive Era. As historian Michael Willrich describes, Progressive lawyers and social workers convinced municipal legislatures to establish a plethora of new courts specifically designed to remedy social ills (Willrich 2003). In his youth as a reformer, ACLU founder Baldwin was heavily involved in this movement, even coauthoring a book about juvenile courts (Walker 1990:33). Deriding procedural and individual rights, Baldwin viewed these courts primarily as a site of contact and engagement with the juveniles whom he intended to help.2 Like Baldwin, many members were quite skeptical about the usefulness of courts. Consequently, the NCLB/ACLU did most of its work on behalf of conscientious objectors by raising awareness of their plight and by leveraging its members’ status as political insiders. But the early Executive Committee also contained moderate 2 In his writings about juvenile courts, Baldwin argued that formal legal protections, such as due process, had no place in a system of juvenile justice, consequently, he believed ‘‘that it is rarely necessary or desirable for counsel to appear in the interest of children’’ (quoted in Walker 1990:33). Like Baldwin, many Progressives believed that, instead of legal arguments, social science (administered by trained social workers) could best protect the interests and meet the needs of delinquent citizens.
Zackin
375
and even conservative lawyers who believed that legal arguments could be made against conscription and in favor of safeguarding other individual liberties. In spite of these members, it is clear that, in the first years of its existence, the NCLB/ACLU placed far greater emphasis on extrajudicial methods of interpreting and applying the Constitution than it did on judicial methods. In his proposal for reorganization of the NCLB, Baldwin summarized the activities of this organization by saying, ‘‘The methods the [NCLB] has followed have been almost exclusively those of a protesting and petitioning group. Together with the legal aid work in particular cases’’ (Baldwin 1920b: n.p.).
The ACLU’s Embrace of Litigation Despite its early commitment to both institutional and grassroots forms of constitutionalism outside the courts, over the course of the 1920s, the ACLU placed increasing emphasis on the value of the same legal, procedural protections and judicial interpretations that its founders had once derided. Scholars of the civil liberties movement, Murphy (1979) in particular, have long noted this transformation of liberal ideas about free expression and have attributed the shift to the unprecedented power of the centralized, administrative state during World War I as well as its new role in regulating political expression. The transformation of ACLU tactics reflects this phenomenon. However, this study of the ACLU’s tactics not only illustrates a shift in thinking about liberties, but also describes a liberal shift in thinking about, and seeking support from, courts and judges. The domestic politics of World War I not only convinced liberals of the importance of individual liberties, but also convinced some liberals of the importance of courts and judges in interpreting and enforcing constitutional guarantees of liberty. Another influence on the ACLU may have been that other leftwing organizations of this period were also beginning to employ test-case litigation, most notably the NAACP.3 In fact, almost from its inception in 1909, the NAACP used courts as one avenue to promote its political agenda. Upon its founding, it established a national legal committee to review relevant cases and recommend promising ones for the organization’s involvement (Carle 2002). By 1915, the NAACP even achieved a significant legal victory at the 3 As Carle details in ‘‘Race, Class, and Legal Ethics in the Early NAACP (1910–1920)’’ (2002), several other civil rights groups also adopted the strategy of test-case litigation in this period. These included Booker T. Washington’s Afro-American Council, W. E. B. Du Bois’s Niagara Movement, and the Constitution League, founded by John Milholland (Carle 2002:4).
376
Popular Constitutionalism’s Hard
Supreme Court level through its participation in Guinn v. United States. The case declared Oklahoma’s ‘‘grandfather clause’’ unconstitutional, thereby addressing electoral disenfranchisement, one of the organization’s earliest concerns (Kellogg 1967:206). Not only did the NAACP meet with early success in the courts, but it also had several founding and/or influential members in common with the ACLU.4 This network of contacts between these organizations may have facilitated the ACLU’s development of its litigative strategies. It is difficult to chart the ACLU’s shift in thinking about and participating in litigation. However, I attempt to demonstrate the shift in two ways. First, I employ the minutes of the Executive Committee’s weekly meetings, and second, I point to evidence of the shift in archival documents. The NCLB/ACLU’s Executive Committee made all decisions regarding the activities of the organization and served as a forum for discussing current tactics as well as proposing future plans. From the minutes, it appears that these meetings consisted of reports on the various activities of the organization as well as discussions and decisions about how to proceed with each activity. A typical meeting seems to have comprised between 10 and 15 items for discussion and addressed tactical and administrative concerns. For instance, a typical entry from the minutes of a meeting held in January 1923 reads: The committee discussed the campaign proposed against criminal syndicalism laws in the legislatures [sic] meeting in January. Mr. Baldwin recommended that a short intensive campaign be organized through the office with the publication of a pamphlet, circulars, etc. to other interested organizations, in an effort to stir up the issue, on the assumption that these laws cannot be repealed but can be made unpopular enough to stop further prosecutions. The alternative of a more extensive nation-wide campaign on the issue was also discussed by the committee, but it was agreed that a small intensive campaign was preferable. An estimate of its cost as part of the office work, not to exceed $300.00, was authorized. (ACLU Executive Committee 1923: n.p.)
I sampled the minutes from weekly meetings between 1918 and 1928, examining an entire month’s worth of meetings at six-month intervals with the idea that the topics discussed in these meetings 4 The early NAACP and ACLU shared several of the same prominent pacifists as influential members, including Jane Addams, Lillian Wald, and Oswald Garrison Villard (Kellogg 1967:249). Though Wald resigned from the early ACLU when it began to litigate, the similar staffs of these organizations suggest that the early leaders of both groups traveled in the same social and political circles. In addition, the Garland Fund, which financed some of the NAACP’s test case litigation in the 1930s, was primarily administrated by Baldwin and had both NAACP and ACLU members serving on its board. James Weldon Johnson, who became the general secretary of the NAACP in 1920, also served on the national board of the ACLU as well as on the board of the Garland Fund.
Zackin
377
(as captured by the minutes) reflect the allotment of the organization’s time and resources. I classified each entry in the minutes according to the tactic it described (e.g., publish a pamphlet, hold a march, hire a lawyer, etc.). I then characterized the tactics as focused either outside the courts, focused inside the courts, or attending to administrative details. It is important to note that each entry does not represent a discrete ACLU activity, so that the particular number of minutes in each category is not meaningful. However, it seems reasonable to suppose that the percentage of total minutes from a given meeting (or a given year) that are devoted to a particular type of tactic ought to reflect the relative amount of time and energy that the organization devoted to that type of tactic. Thus I present a time-series of the fraction of courtbased or non-court-based strategies described in the minutes from each year. Many of the entries in the minutes describe tactics designed to publicize the ACLU’s activities. The question of how to characterize these publicity-oriented tactics was a difficult one. On one analysis, publicity about civil liberties violations could be construed as a tactic aimed at generating democratic interpretation of the Constitution outside of the courts. However, publicity about particular court cases could be considered part of a litigation strategy, in which the ACLU used courts to frame and publicize the constitutional questions at issue. On this understanding, publicity-related activities are coded as court-based tactics when they relate to cases and as extrajudicial when they do not (for a complete description of this coding scheme, see Appendix A). The results of this analysis, presented in Figure 1 below, suggest that over the course of the 1920s the ACLU’s Executive Committee used an increasing percentage of meeting time to discuss court-based tactics. The figure indicates a corresponding decrease in the percentage of the meeting time used to discuss tactics that involved constitutional interpretation by actors outside the courts. Another view of ACLU activities might lead one to distinguish all publicity-related activities from either court-based or non-courtbased tactics. It seems reasonable to imagine that the ACLU publicized its activities and concerns regardless of the content of those activities. When minutes discussing publicity are distinguished from those that discuss court-based and non-court-based strategies, the picture of the ACLU’s energies, shown in Figure 2, remains largely unchanged. The light gray line near the bottom of the graph represents the publicity-oriented activities discussed during meetings of the Executive Committee. This line remains fairly flat, suggesting that publicity was a reasonably consistent part of the early ACLU’s activities, regardless of its degree of involvement with courts. More
378
Popular Constitutionalism’s Hard
% court-based % outside courts Percentage of Total Minutes
60 50 40 30 20 10
1928
1927
1926
1925
1924
1923
1922
1921
1920
1919
1918
0
Year
Figure 1. ACLU Strategy: 1918–1928.
important, treating publicity as a separate category seems to have little effect on the committee’s clear trend away from tactics outside the courts and toward court-based strategies. In his history of the ACLU, Walker (1990) notes that in 1920, only three of the 20 Executive Committee members were lawyers and that lawyers had little influence on the activities of the ACLU. Walker highlights the appointment of Arthur Garfield Hayes and Morris Ernst as the ACLU’s co-general counsels in 1929, arguing that the appointment %court-based %outside courts %publicity 60 50 40 30 20 10
19 18 19 19 19 20 19 21 19 22 19 23 19 24 19 25 19 26 19 27 19 28
0
Year
Figure 2. ACLU Strategy With Publicity Distinguished: 1918–1928.
Zackin
379
of two general counsels ‘‘signaled a decisive shift in the ACLU’s priorities’’ (Walker 1990:69). While Walker singles out the 1929 appointments as a decisive turning point in ACLU strategy, the analysis presented above suggests that these appointments actually represent the culmination of a gradual transformation that occurred over the course of the previous decade. Archival evidence also indicates that a shift away from popular constitutionalism (and toward court-centered tactics) was under way well before 1929. This transformation is evident most clearly in ACLU members’ opposition to it. Opposition surfaced within the AUAM immediately upon the founding of a branch of the organization devoted to defending the rights of conscientious objectors. In response to the suggestion that the AUAM should found such a bureau, prominent Progressive social worker Lillian Wald threatened to resign, arguing that the defense of conscientious objectors and expression rights more generally would force the bureau into an unacceptably confrontational stance with respect to the government (Eastman 1917). Three years later, another founding member of the AUAM, Zona Gale, leveled a different criticism at the ACLU. Gale’s primary concern was with the ACLU’s defense of individual liberties at the expense of social harmony. When the end of World War I allowed the ACLU to spend less and less time defending conscientious objectors, it turned its attention to the defense of other forms of suppressed, political speech. In particular, the ACLU intensified its defense of Communist rhetoric, some of which urged the need for a violent, proletarian revolution. For the antiwar Progressive Gale, the ACLU’s defense of the individual right to advocate violent revolution was simply beyond the pale. Interestingly, Gale seems to have agreed with the ACLU’s interpretation of the Constitution, but in a moving letter to Baldwin, she wrote, ‘‘[t]he constitutionality is not enough. It is the extra-constitutional, extra-legal right that I mean. The moral right, if you will. The moral right to stand for the right to ‘violent and obnoxious speech’! I can’t square it’’ (Gale 1920a: n.p.). Gale’s distinction between moral rights and legal rights went hand in hand with her disdain for judicial interpretations of the Constitution. Not surprisingly, then, in her final letter of resignation from the ACLU’s board, Gale revealed that she had tolerated, but disliked, the ACLU’s move toward the use of court-based tactics. She wrote, ‘‘I have even been willing to pass over the need to use the courtsF which I do not use, and by which I feel one loses more than one gains, even when anything is gained’’ (Gale 1920b: n.p.). Though both Wald and Gale resigned early in the organization’s history, and in response to particular policy decisions, the ACLU’s larger shift away from popular constitutionalism was not the result of an intentional or decisive switch. In fact, the ACLU
380
Popular Constitutionalism’s Hard
had always been involved with the courts, offering legal advice and defense to conscientious objectors who found themselves on trial. In 1917, the New York Times quoted from the organization’s first press release: ‘‘Its chief purpose will be to give legal aid and advice through attorneys and committees of citizens in all parts of the United States to persons whose rights are invaded under pressure of war’’ (‘‘Bureau to Defend Lovers of Peace,’’ The New York Times, 3 July 1917, p. 5). The same article quoted Baldwin saying, ‘‘It will be the object of the Civil Liberties Bureau to act as a clearing house for complaints of injustice’’ (‘‘Bureau to Defend Lovers of Peace,’’ The New York Times, 3 July 1917, p. 5). In 1920, Baldwin suggested that the ACLU affiliates not only defend people who happen to find themselves in court, but also that they start getting themselves dragged into court on purpose. Baldwin noted near the end of the document, ‘‘Of course we should retain our cooperating attorneys and such legal aid as is necessary’’ (Baldwin 1920b: n.p.). From this description, it appears that while defendants would no longer be on trial against their will, their ACLU-affiliated lawyers would play largely the same role as they had previously, offering the best defense possible to defendants who were not of their choosing. By 1923, the ACLU had still not identified courts as the primary venue for its work. Yet the annual report for that year reflected the possibility of pursuing legal change through litigation. In discussing a case of police interference with Worker’s Party meetings, the report stated, ‘‘[t]he case is important as involving the right to hold meetings on private property without interference . . . The case presents a clear issue and may be carried to the Supreme Court of the state’’ (ACLU 1923: n.p.). By the end of 1924, however, the ACLU began to intentionally generate test cases. For instance, Baldwin explained his arrest to The New York Times by telling the paper, ‘‘[i]t [is] part of the tactics of [my] organization to defy efforts to prevent meetings and thus force such cases into Court’’ (‘‘Defends Parading for Civil Liberties,’’ The New York Times, 18 Dec. 1924, p. 15). By 1927, 10 years after the NCLB was founded as a ‘‘clearinghouse for complaints of injustice,’’ the ACLU’s annual report stated, ‘‘It should be remembered that we do not take up all cases involving civil rights. We are not a general defense organization . . . The Union tackles test cases involving laws and regulations, demonstrations in places of conflict, proceedings against lawless officials, public protests and propaganda against repression and intolerance in any form’’ (ACLU 1927: n.p.; emphasis added). This description of tactics reveals the ACLU’s eventual reliance not only on legal defense, but also on the intentional challenge of laws through the careful selection of test cases. Thus by the late 1920s, not only was the ACLU still active in courts,
Zackin
381
but this activity had also become central to its mission, which had become the realization of legal change.
Why Litigate? Though some of the literature on interest group litigation suggests that groups begin to litigate when they begin to believe judges will rule in their favor (Hansford 2004:173), the ACLU’s leadership shifted the organization’s resources to the courts before they believed it would win cases. Despite the fact that the ACLU did not accomplish its first real Supreme Court victories until 1931, it had already devoted most of its energies to a strategy of litigation.5 Though the ACLU’s shift in tactics cannot be attributed to a change in the courts, it can be explained in large part by the failure of its original strategies. Throughout the 1920s, the ACLU was primarily committed to defending the speech of decidedly unpopular figures, such as labor leaders, socialists, and communists. These defendants did not engender broad public support for their civil liberties and made it extremely difficult for the ACLU to pursue grassroots, activist strategies. During World War I, Baldwin attempted to organize a series of mass meetings to rally public opinion behind freedom of conscience, but he found some auditorium 5 The ACLU first began making arguments about freedom of expression as part of its opposition to the draft. Such arguments did not fare well in court. In the Selective Draft Law cases of 1919, the Supreme Court upheld the draft law as a legitimate exercise of Congress’s power to declare war. The ACLU then extended its freedom of expression arguments to criticism of the Espionage Act but met with an equal lack of judicial sympathy. When the first crop of Espionage Act cases (Schenck v. United States, Frohwerk v. United States, and Debs v. United States) reached the Supreme Court in 1919, the Court unanimously upheld the Espionage Act as well as all three convictions. Few lower courts proved much more sympathetic to the ACLU’s First Amendment arguments against the Act (Rabban 1997:256). The Court’s Abrams decision, five months after Schenck, also upheld the Espionage Act, as well as Abrams’s conviction under it (Abrams v. United States 1919). However, in this case, Justices Oliver Wendell Holmes and Louis Brandeis began their famous tradition of pro-speech dissents, arguing that speech was protected by the First Amendment unless it created a clear and present danger. The Supreme Court upheld the Espionage Act in three more cases in 1920 and 1921, though Justices Holmes and Brandeis dissented in these as well, again arguing in favor of the clear-and-present-danger test. The ACLU won its first partial victory in 1925 in the case of Gitlow v. New York, which incorporated the First Amendment’s free speech guarantee against the states and established freedom of speech as a fundamental liberty. However, the Court upheld Gitlow’s conviction under a New York statute prohibiting the advocacy of anarchy, thereby convincing many ACLU members that the case represented yet another defeat (Walker 1990:80). The Supreme Court also upheld California’s criminal syndicalism laws and the conviction of a Communist Labor Party founder in the 1927 case Whitney v. California (Lewis 1991:65–91). Despite the encouraging nature of Justices Holmes’s and Brandeis’s dissents, the ACLU did not win an unmitigated victory before the Supreme Court until the 1931 Stromberg and Near cases (Stromberg v. California, Near v. Minnesota), in which the Court overturned a California law outlawing the display of a red flag and a Minnesota law authorizing prior restraint of the press (Walker 1990:90). These two decisions represented a shift in First Amendment doctrine and signaled a newly sympathetic Court.
382
Popular Constitutionalism’s Hard
managers unwilling to rent space to him when they discovered the topic of his meetings (Walker 1990:23). On occasion, the ACLU was also denied the use of public venues because of its association with radicals (‘‘Sanctions Bar Talks for Pupils: Board of Education Refuses Use of School Hall to Civil Liberties Union,’’ Los Angeles Times, 21 Aug. 1923, p. I17; ‘‘Beach Talk on Debs Refused,’’ Los Angeles Times, 20 Sept. 1923, p. II10). When ACLU leaders were allowed to speak, their affiliations with extremists were often held up as their sole cause for existence. One particularly expressive Washington Post columnist wrote, ‘‘[The ACLU] is a part of the malignant movement which has been extant for some years for sapping the spirit of American patriotism and poisoning our national life with Sovietism and international Communism . . . They clamor like howling dervishes for the right to promote every anarchistic fad that has oozed out of the ferment of old world corruption’’ (‘‘Against Teaching Patriotism,’’ The Washington Post, 6 Nov. 1924, p. 6).6 Even more discouraging to the practice of direct, popular constitutionalism were the extremely violent responses that radicals faced while trying to enact their vision of the Constitution through public meetings, speeches, and strikes. They often found themselves targets of vigilante justice perpetrated by groups such as the Ku Klux Klan and the American Legion. According to the ACLU’s estimates, more than 800 alleged radicals fell victim to mob violence between September 1920 and June 1922 alone (‘‘More than 800 Victims of Mobs in 7 Months,’’ The Washington Post, 22 Nov. 1922, p. 2). In light of this extraordinary hostility, it also seems foreseeable that the ACLU would begin to place less emphasis on popular interpretations of the Constitution while stressing the need for judges to enforce due process protections and individual rights.7 The ACLU’s shift to the courts was also driven by the failure of its attempts to promote its vision of the Constitution among 6 Several decades later, the NAACP also found its attempts to organize rallies thwarted by its association with communist groups. In fact, because communist groups continued to pursue the tactic of the mass rally in order to protest police brutality, the NAACP began to file civil suits instead (M. Johnson 2003:219). This shift in tactics is yet another example of the way that that an interest group’s political context often determines whether popular mobilization or litigation will render the members of that group as more effective or appealing in a particular period. 7 AUAM, Proposed Announcement for Press, September 24, 1917, AUAM Collected Records, Swarthmore College Peace Collection. Even as early as 1917, Eastman expressed her sense that the mainstream public, which had been so supportive of many of the AUAM’s early activities, might not be dependable in safeguarding the expression rights of pacifists. She explained, ‘‘It takes an exceedingly large-minded liberal to fight for the right of another man to say exactly what he himself does not want said. He may stand for free speech, but he won’t really fight for free speech, so long as what he wants to say goes. With rare exceptions the minority must depend upon itself and its own unaided efforts to maintain its right to exist’’ (Eastman 1917: n.p.).
Zackin
383
members of the government. Though the ACLU began its existence hoping to serve as a friend and advisor to the Wilson administration during wartime, it soon found that most members of the administration desired neither the organization’s friendship nor its advice. The NCLB was only granted admittance to the White House once during the entire decade after the war (Walker 1990:56). Furthermore, although President Wilson had appointed a liberal staff to the Department of War, this staff became much less friendly to the NCLB after military intelligence agents investigated the NCLB and determined that it was encouraging draft-dodging. Frederick Keppel, a senior staff member of the War Department and close personal friend of several NCLB members, formally severed relations with the NCLB in 1918, writing that further contact with the organization would be ‘‘embarrassing’’ (D. Johnson 1963:37). Citing the Espionage Act, the Post Office Department began to suppress the socialist newspaper The Masses. When Baldwin responded as an ally of the government, seeking guidelines about what content would be considered legal, the Post Office Department flatly refused to provide any. Postmaster General Albert Burleson responded to requests for guidance by suggesting that the NCLB was free to sue over any decisions with which it disagreed (D. Johnson 1963:58–9). Shortly thereafter, the Post Office Department seized several of the NCLB’s own mailings (D. Johnson 1963:60–1). The FBI raided the ACLU’s offices in September 1918, which prompted the ACLU to hire outside counsel (Phillips 1954:67). In his oral history, Baldwin explained, ‘‘In the First World War Years, the savage repression of dissent alienated us from government’’ (quoted in Phillips 1954:180). In 1920, Baldwin wrote to Gale, ‘‘To my mind, there is no violence greater than that of governments’’ (Baldwin 1920c: n.p.). Given the antagonism that the Wilson administration displayed toward the ACLU, it is not surprising that the ACLU was forced to abandon its strategy of cooperating with the government.
The Empirical Assumptions Underlying Calls for Constitutionalism Without Courts In order to understand the ACLU’s turn to the courts, it is necessary to understand the state as the ACLU leaders came to understand itFas an active agent with its own agenda and coercive power (Evans et al. 1985). Yet this view of the state is notably absent from calls to abolish judicial review. For instance, Tushnet (1999) declares that, despite the conventional wisdom that disadvantaged and insular minorities require judicial protection, he sees no reason why such minorities should not simply use conventional political
384
Popular Constitutionalism’s Hard
bargaining to overcome their disadvantages. He suggests that disadvantaged minorities ought to identify an issue on which the majority is closely divided and offer its entire vote share to the side that agrees to meet its terms. He writes, ‘‘It is pretty easy to see how such a minority actually can get quite a bit of what it cares about . . . say to both sides ‘we will deliver our votes on that issue to whichever side votes for our issues’’’ (Tushnet 1999:159). The ACLU’s early history speaks to this suggestion. Though ACLU leaders professed and demonstrated devotion to the principle of popular constitutionalism, they found themselves unable to bargain in the way Tushnet suggests. Tushnet’s image of bargaining assumes that unpopular minorities function as well-organized political units. However, in the case of the ACLU’s early clients, the formation of such political units was exactly what the state and federal governments had outlawed through the Espionage Act and criminal syndicalism laws and punished through mail stoppages, surveillance, and office raids. Not only had the government banned the political expression of unpopular groups, but the high levels of private violence perpetrated against these groups made it difficult for them assemble, let alone trade votes. This fly in Tushnet’s bargaining ointment highlights his assumptions about the state, as well as the empirical problems with those assumptions. The bargaining that Tushnet describes takes place within a state that appears to be an arena, one that sets the rules for political competition but never enters the fray. However, the state that the ACLU and its radical clients encountered had its own agenda, largely independent of the citizens it governed. It sought to eliminate criticism of its methods, actions, and existence by outlawing such criticism. While an outraged citizenry might have been able to overcome this agenda, the ACLU’s clients were generally so unpopular in this period that private actors also made it difficult for them to argue their cases outside the courts. In short, it became difficult for the ACLU to motivate constitutional deliberation outside of the courts without using the courts. The ACLU discovered that litigation could facilitate the kind of grassroots movement around constitutional meanings that it was trying to spark. Advocates of court-free constitutionalism assume that constitutional litigation precludes popular interpretation of the Constitution. They are particularly rankled by ACLU-style litigation, which often relies upon courts to nullify laws passed by elected legislatures or to interpret the law in order to forbid a widespread practice.8 Yet many leaders of the early ACLU 8 In order to be effective, this type of litigation relies on the consensus that judges’ interpretations of the Constitution are legally binding, even if people disagree with the ruling or dislike its implications for public policy. Kramer asserts that the idea that the
Zackin
385
conceived of their court-based tactics as serving the goal of stimulating constitutional deliberation among the general public. This vision is particularly evident in many of Baldwin’s remarks, including his 1920 plan for the reorganization of the ACLU. Baldwin argued that, to get people to understand the importance of protecting civil liberties, the ACLU had to exercise those civil liberties in places where they were threatened. In response to prohibitions on labor meetings, he wrote, ‘‘[this] denial of civil liberties can only be dramatized by defying it through the exercise of constitutional rights. In addition, a procession of strikers’ wives, for instance, could be organized with well-known sympathizers at its head to go to the Sheriff demanding the repeal of his order. The fight could be carried by meetings and demonstrations to the Governor’s office’’ (Baldwin 1920b: n.p.). These test meetings, as suggested by Baldwin, became a staple of ACLU strategy. Yet as this quotation makes clear, the ACLU began to employ test meetings and cases as part of a larger strategy of popular constitutionalism that included more recognizably democratic tactics, such as demonstrations and the petitioning of public officials. ACLU attorney Hayes also saw court proceedings as a platform for political and philosophical arguments. He did not see much potential in changing legal precedents, but he did believe that trials provided valuable opportunities to educate both judges and the broader public (Walker 1990:53). Another ACLU attorney, Walter Nelles, did see promise in bringing test cases before the court in order to obtain legal victories. Yet Nelles also believed that the courts could play a positive role in popular deliberation about constitutional questions. Though Nelles agreed with the basic notion of dramatization, he felt that a magazine, explaining the issues involved in the ACLU’s test cases, was necessary to promote effective deliberation among members of the public. Nelles believed that ‘‘as dramatizations of the issues of civil liberty, sedition trials ought to be effective’’ but that the press generally focused on sensationalist details, rather than pertinent questions of law (Nelles 1919:2). Yet Nelles clearly believed that trials had a role to play in popular constitutionalism. He wrote, ‘‘The job of getting the real nature of these cases before the court of public opinion ought to be done by a lawyer; it cannot professionally be done by a lawyer who is counsel in the case. That job I should like to take on. The Abrams public should accept judicial rulings even when they dislike them ‘‘is an ideological tenet whose whole purpose is to persuade ordinary citizens that, whatever they may think about the Justices’ constitutional rulings, it is not their place to gainsay the Court. It is a device to deflect and dampen the energy of popular constitutionalism’’ (Kramer 2004:233). He states that the object of this idea is ‘‘to maximize the Court’s authority by inculcating an attitude of deference and submission to its judgments’’ (Kramer 2004:233). Thus Kramer argues that policy-oriented litigation is wholly at odds with popular deliberation about the Constitution’s meaning.
386
Popular Constitutionalism’s Hard
case leaves the status of civil liberty hopeless so far as it is the concern of the courts of law’’ (Nelles 1919:2). This letter raises several important points. The first is that, even when the courts did not seem likely to rule in the ACLU’s favor, Nelles still saw value in litigating. Litigation under these circumstances struck Nelles as valuable because it reached the members of, and may even have created, a ‘‘court of public opinion.’’ While judges were determining the meaning of the Constitution in a court of law, informed citizens were simultaneously determining the meaning of the Constitution in the court of public opinion. This image of dual courts suggests that litigation was not merely aimed at generating publicity, but that the publicity it generated was in the service of public deliberation and interpretation about constitutional cases. As demonstrated by Francis in her description of the NAACP’s campaign against lynching and mob violence, the early NAACP also treated court cases as part of a larger strategy to publicize the repression it opposed. Like the ACLU, the NAACP used courts in part to supplement its other efforts to change the public understanding of lynching and mob violence (Francis n.d.). This use of courts suggests that litigation may not stunt popular constitutionalism, as Kramer and Tushnet argue, but on the contrary, it may promote the public consideration of politics through a constitutional lens. This notion is echoed by modern legal scholars, such as Post and Siegel, who also identify a reciprocal relationship between popular and judicial interpretations of the Constitution. They challenge arguments such as those of Tushnet and Kramer, declaring, ‘‘We are capable of prizing law by denigrating politics, or of prizing politics by denigrating law, but we rarely imagine law and politics as respectfully coexisting, as they often do’’ (Post & Siegel 2003:20). Just as Post and Siegel object to the Supreme Court’s assertion of total supremacy over the Constitution, so too do they object to notions of a Constitution to which political majorities alone give meaning. Instead, they recall Corwin’s trenchant description of the Constitution as a double symbol, representing both a check on tyranny (of the majority and its state) as well as the establishment of the right of that majority to political self-determination (Corwin 1936). These dual functions may exist in tension with one another; however, we need not imagine that they are mutually exclusive. By asking judges to interpret the Constitution, in hopes of encouraging private citizens to do the same, ACLU leaders recognized and utilized this dual symbolism of the Constitution. In using the courts for public relations purposes, the ACLU also recognized the theatrical nature, and therefore educative potential, of court cases. Both Baldwin and Nelles used the word dramatize in their description of litigation’s role in popular constitutionalism.
Zackin
387
This term is quite suggestive. One component of dramatization is clearly the generation of publicity. However, the term also suggests the creation of a drama or story line, intended for spectatorship. Litigation packages an abstract rights claim into a particular narrative, centered on a single incident and defendant. Such cases are made even more educative because the proceedings offer defendants (and their lawyers) an opportunity to argue for the justness of their claims. These arguments are protected when made inside a courtroom and are amplified by the media attention that court cases, particularly those before high courts, often draw. These properties help explain why the ACLU turned to litigation well before it was likely to win any notable judicial victories. Even when a judicial victory is outside the realm of possibility, test cases may secure a victory ‘‘in the court of public opinion’’ by dramatizing civil liberties issues. The ACLU’s awareness of the way test cases could stimulate public discussion and influence public opinion was reflected by the types of test cases it engineered as well as its policy of notifying the press before staging disobedience to repressive laws.9 In a display that typified the ACLU’s tactics, Upton Sinclair responded to a ban on labor meetings by alerting the police that he planned to violate the ban and then by reading the First Amendment in the hopes of getting arrested. Though Sinclair personally hoped that the case would be dropped, he wrote many articles and gave many speeches publicizing and criticizing his arrest and prosecution (Zanger 1969:392–7). The publicity generated by this demonstration was instrumental in forming the Southern California chapter of the ACLU. From the perspective of dramatization, however, the Scopes trial was perhaps the most effective test case in which the early ACLU participated. As a result of the case’s famous attorneys (William Jennings Bryan and Clarence Darrow) and of the battle between science and religion that the case seemed to represent, it generated an enormous amount of media attention. Most of this attention presented the ACLU in a favorable light. Scopes was a sympathetic defendant; a young, white, middle-class native of Tennessee, he was popular around town and with his students. Furthermore, Tennessee’s anti-evolution statute represented something of a regional outlier. To much of the country, the case seemed to pit progress and freedom of thought against superstition and the narrow-minded repression of ideas. As far as the Scopes case was concerned, the ACLU seemed to represent progress (Garey 1998:81–5). As a result, the Scopes trial bolstered the national image of the ACLU. In connection with the case, it received its first wholehearted endorsement from a national newspaper, and the 9
Notification of the press is discussed in Philips (1954:175).
388
Popular Constitutionalism’s Hard
American Academy for the Advancement of Science as well as the American Bar Association made positive public statements about the organization (Walker 1990:75–6). However, most of the ACLU’s early clients remained decidedly unpopular figures. The ACLU’s transition to litigation is interesting not only because of what it can tell us about theories of constitutionalism without courts, but also because it illustrates why courts may be useful to unpopular minorities. It is important to note that the courts are by no means guaranteed to protect the rights of unpopular minorities. Yet as the ACLU discovered, even when judges decide with the majority, litigation may still aid unpopular minorities. Because courts had the power of judicial review, they provided the ACLU with an avenue for political participation when the organization found its more accustomed avenues inaccessible. Baldwin explained, ‘‘Just to make a fight, even hopeless, testified to a courage and determination which built up the power of resistance to injustice, good anyhow, as we saw it, for the future’’ (Phillips 1954:152). No matter how unpopular it got, the ACLU could engineer test cases, and at least lower-level courts could not keep them out. This property of litigation has been noted in other contexts, most explicitly in McCann’s study (1994) of women’s pay equity struggles of the 1970s and 1980s. McCann explains that even when the courts do not serve as countermajoritarian heroes, litigation provides activists with a vehicle for challenging a firmly entrenched status quo.
Conclusion The legal strategies of groups such as the NAACP and ACLU are often described as examples of ‘‘disadvantaged constituencies [practicing] a minoritarian politics that allow[s] them to reach beyond the limits of the political process by making direct appeals to constitutional principles through litigation’’ (Ivers 1995:9). The idea that courts ought to protect unpopular minorities, by validating their direct appeals to constitutional principles, is famously expressed in Ely’s influential theory of judicial review (Ely 1980). It has certainly become the conventional wisdom. Kramer and Tushnet are part of a larger scholarly movement aimed at discouraging interest groups, particularly those on the left, from believing this conventional wisdom and continuing to seek social change through litigation. For instance, Rosenberg’s The Hollow Hope famously argues that the Supreme Court is like flypaper, attracting and then ensnaring unsuspecting activists (Rosenberg 1991:336). Much of this literature implies that today’s political organizations have been confused into continued litigation by the legacy of the Warren
Zackin
389
Court. This article suggests that well before the Warren Court existed, minority groups felt the need to litigate. Consequently, it serves as a reminder that the idea of courts protecting minorities pre-dates the 1960s. In fact, the idea that courts could defend minorities can be traced back to the Federalists’ concern for the necessity of checking majoritarian politics. The minorities imagined in this scenario were economic minoritiesFin other words, economic elites. At the turn of the twentieth century, business groups with this frame of reference turned to the courts for protection against threatening trade unionism and regulatory legislation. One of the earliest such groups, the American Anti-Boycott Association (AABA), was founded in 1902 with the express purpose of litigating and lobbying against trade unionism (Ernst 1995:5). From 1903 to 1925, it conducted a sustained program of litigation seeking, through its lawsuits, to eradicate the practice of closed union shops (Ernst 1995:91). Litigation seemed like an obvious strategy for this group because, for the previous half-century, judges had denounced unions as anticompetitive, conspiratorial, and illegal cartels (Ernst 1995:6). Thus the AABA’s founders believed that precedent lay squarely on their side. AABA historian Ernst explains, ‘‘The goal of making new law, so readily acknowledged by today’s legal defense funds was no part of the employers’ plans for the AABA. To adopt such a goal would go against their deep conviction that existing law already recognized the justice of their cause’’ (Ernst 1995:22). This kind of litigation was designed to convince courts that prevailing constitutional norms should be applied in defense of the status quo. Progressive Era interest groups on the political left also found themselves concerned that judicial interpretations of the Constitution would uphold the status quo. In fact, they often feared precisely this outcome. For instance, the National Consumers’ League found it necessary to attend to the judicial branch in order to defend the protective labor legislation for which it had successfully lobbied from judges who consistently declared such legislation unconstitutional (Vose 1957). This organization found that it needed to develop the capacity to litigate, not because courts were relatively sympathetic to its aims or even a last resort when majoritarian politics became impossible. On the contrary, the National Consumers’ League was interested in courts because they were so disruptive to its political program. It litigated in order to defend majoritarian legislative outcomes from the courts, rather than to defend minority rights from legislation. This litigation was quite different from that of groups such as the ACLU and NAACP. At a time when courts were strongly associated with conservative causes, these left-wing organizations began to develop the understanding that courts might be useful to them. Though the decisions of the Warren Court undoubtedly
390
Popular Constitutionalism’s Hard
strengthened the modern-day perception that courts are the guardians of social (rather than economic) minorities, groups such as the ACLU and NAACP began to model the view that courts could facilitate social change decades before the Warren era. In fact, in the 1940s, Jewish advocacy organizations, most notably the American Jewish Congress, emulated this model, adopting litigation strategies like those of the NAACP and ACLU (Ivers 1995:51). The interest group literature has traditionally focused on the properties, costs, and benefits of large-scale litigation campaigns. By focusing on these groups’ transition to litigation, this study offers insights about extrajudicial strategies aimed at changing constitutional norms as well as the political challenges that accompany those strategies. The ACLU’s experience demonstrates that groups may be driven to courts as a result of the empirical problems with court-free constitutionalism. Tushnet’s and Kramer’s normative arguments do not seem to account for such problems. For instance, Tushnet’s vision of constitutionalism seems to be that legislatures and executives will allow unpopular and even threatening minorities to develop the political organizations necessary to participate in pluralist politics. As the Espionage and Sedition Acts demonstrate, however, institutional actors pursue their own interests, which often dictate that such groups are not allowed to even organize, let alone participate. Another empirical flaw in the theory of court-free constitutionalism is that it poses litigation as being in total opposition to popular deliberation about the meaning of the Constitution. As the ACLU’s test cases demonstrate, however, litigation may actually help stimulate popular discussion of constitutional questions. This article does not argue that the ACLU should have won its cases or that the real meaning of the Constitution has been validated through court decisions. Furthermore, it does not even claim that the ACLU was ultimately successful or that its litigation accounted for any success it may have had. Instead, the ACLU serves as an example of an organization that tried popular constitutionalism but found that empirical realities made it nearly impossible to continue the pursuit of this strategy. The scholars who call for constitutionalism without courts do not discuss these obstacles. They seem to imagine that groups must choose to either litigate or participate in democratic politics and that either option will always be available. As this case demonstrates, these assumptions do not always hold. It is not clear whether those who call for restrictions on judicial review are willing to accept the consequence that some groups will find popular constitutionalism beyond their reach. However, it is worth considering whether such an outcome presents a normative cost and, if so, whether that cost offsets the benefit that scholars hope to achieve by restricting the role of courts in constitutional interpretation.
Zackin
391
Appendix A: Coding Scheme Used to Generate Figure 1 Topics Coded as ‘‘Court-Based’’
informational report about a court ruling or case made to meeting publicize a case write/make remarks in court civil disobedience/test meeting test case discussion of whether to accept an individual case consult or retain a lawyer application for injunction write of a legal brief summary of cases ACLU was handling at time of meeting
Topics Coded as ‘‘Outside Courts’’
letter-writing campaign public or press statement public meeting public protest coordinate speaking tour write letter to the editor publish a pamphlet or book conference with member of administration ask a congressman to make a statement or introduce a bill arrangement with War Department memo to congressman or state legislator letter to government official advertisement in newspaper miscellaneous political activities
Topics Not Coded in Either Category
hold a benefit informational report given during meeting gather information/stage an investigation bail fund sue or threaten to sue a newspaper for libelous statements about ACLU administrative details letter to members/friends of organization unspecified publicity
392
Popular Constitutionalism’s Hard
Table 1. Data for Figure 1
Date
Court-Based Number
Outside Courts Number
Neutral Number
Total Minutes for Year
% Tactics CourtBased
% Tactics Outside Courts
% Tactics Neutral
1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928
4 1 32 12 35 36 50 38 45 31 58
12 41 53 27 45 34 20 15 13 14 7
22 38 66 31 45 60 35 54 27 25 37
38 83 153 71 129 138 108 111 87 72 105
10.53 1.20 20.92 16.90 27.13 26.09 46.30 34.23 51.72 43.06 55.24
31.58 49.40 34.64 38.03 34.88 24.64 18.52 13.51 14.94 19.44 6.67
57.89 45.78 43.14 43.66 34.88 43.48 32.41 48.65 31.03 34.72 35.24
Percentages do not sum to 100 because some minutes could not be classified.
Appendix B: Coding Scheme Used to Generate Figure 2 Topics Coded as ‘‘Court-Based’’
informational report about a court ruling or case made to meeting write/make remarks in court civil disobedience/test meeting test case discussion of whether to accept an individual case consult or retain a lawyer application for injunction write of a legal brief summary of cases ACLU was handling at time of meeting
Topics Coded as ‘‘Outside Courts’’
letter-writing campaign public meeting public protest coordinate speaking tour conference with member of administration ask a congressman to make a statement or introduce a bill arrangement with War Department memo to congressman or state legislator letter to government official miscellaneous political activities
Zackin
393
Topics Coded as ‘‘Publicity’’
publicize a case public or press statement write letter to the editor publish a pamphlet or book unspecified publicity advertisement in newspaper
Topics Not Coded in Any Category
hold a benefit informational report given during meeting gather information/stage an investigation bail fund sue or threaten to sue a newspaper for libelous statements about ACLU administrative details letter to members/friends of organization Table 2. Data for Figure 2 Court- Outside Total % Tactics % Tactics Based Courts Neutral Publicity Minutes Court- Outside % Tactics % Tactics Date Number Number Number Number for Year Based Courts Publicity Neutral 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928
4 1 31 12 35 36 49 34 41 31 56
9 32 39 24 36 23 17 11 11 10 5
21 34 55 22 40 50 30 50 25 22 35
4 13 26 12 14 21 9 12 8 7 6
38 83 153 71 129 138 108 111 87 72 105
10.53 1.20 20.26 16.90 27.13 26.09 45.37 30.63 47.13 43.06 53.33
23.68 38.55 25.49 33.80 27.91 16.67 15.74 9.91 12.64 13.89 4.76
10.53 15.66 16.99 16.90 10.85 15.22 8.33 10.81 9.20 9.72 5.71
55.26 40.96 35.95 30.99 31.01 36.23 27.78 45.05 28.74 30.56 33.33
Percentages do not sum to 100 because some minutes could not be classified.
References ACLU (1923) Annual Report. ACLU Printed Material Box 1879, Seeley Mudd Library, Princeton University, Princeton, NJ. FFF (1927) Annual Report. ACLU Printed Material Box 1879, Seeley Mudd Library, Princeton University, Princeton, NJ. ACLU Executive Committee (1923) Minutes, 8 Jan, Series 18, Volume 229, Reel 31, Seeley Mudd Library, Princeton University, Princeton, NJ. Ackerman, Bruce (1991) We The People. Cambridge, MA: Harvard Univ. Press. AUAM Executive Committee (1917) Minutes, 15 June, AUAM Collected Records, Swarthmore College Peace Collection, Swarthmore, PA.
394
Popular Constitutionalism’s Hard
Baldwin, Roger (1920a) ‘‘The Fight for Civil Liberty During the War,’’ ACLU Papers, Volume 108, Seeley Mudd Library, Princeton University, Princeton, NJ. FFF (1920b) ‘‘Suggestions for Reorganization of the National Civil Liberties Bureau,’’ Directing Committee Minutes, ACLU Papers, Volume 120, Seeley Mudd Library, Princeton University, Princeton, NJ. FFF (1920c) Letter to Zona Gale, 23 Aug., Directing Committee Correspondence, ACLU Papers, Volume 120, Seeley Mudd Library, Princeton University, Princeton, NJ. Carle, Susan D. (2002) ‘‘Race, Class, and Legal Ethics in the Early NAACP (1910–1920),’’ 20 Law and History Rev. 97–146. Corwin, Edward S. (1936) ‘‘The Constitution as Instrument and as Symbol,’’ 30 The American Political Science Rev. 1071–85. Eastman, Crystal (1917) Letter to AUAM Executive Committee, 14 June, AUAM Collected Records, Swarthmore College Peace Collection, Swarthmore, PA. Ely, John Hart (1980) Democracy and Distrust: A Theory of Judicial Review. Cambridge, MA: Harvard Univ. Press. Ernst, Daniel R. (1995) Lawyers Against Labor: From Individual Rights to Corporate Liberalism. Urbana: Univ. of Illinois Press. Evans, Peter B., et al. (1985) Bringing the State Back In. Cambridge, United Kingdom: Cambridge Univ.ersity Press. Francis, Megan (n.d.) ‘‘Mobilizing the Supreme Court Against Injustice: The Untold Story of the NAACP and the Revolution in Criminal Procedure.’’ Ph.D. dissertation, Princeton University, Princeton, NJ. Forthcoming. Gale, Zona (1920a) Letter to Roger Baldwin, 25 Aug., Directing Committee Correspondence, ACLU Papers, Volume 120, Seeley Mudd Library, Princeton University, Princeton, NJ. FFF (1920b) Letter to Roger Baldwin, 7 Sept., Directing Committee Correspondence, ACLU Papers, Volume 120, Seeley Mudd Library, Princeton University, Princeton, NJ. Garey, Diane (1998) Defending Everybody: A History of the ACLU. New York: TV Books. Graber, Mark (1991) Transforming Free Speech: The Ambiguous Legacy of Civil Libertarianism. Berkeley: Univ. of California Press. Hansford, Thomas (2004) ‘‘Lobbying Strategies, Venue Selection, and Organized Interest Involvement at the US Supreme Court,’’ 32 American Politics Research 170–97. Ivers, Gregg (1995) To Build A Wall: American Jews and the Separation of Church and State. Charlottesville: Univ. Press of Virginia. Johnson, Donald (1963) The Challenge to American Freedoms: WWI and the Rise of the American Civil Liberties Union. Lexington, KY: Univ. of Kentucky Press. Johnson, Marilynn (2003) Street Justice: A History of Police Violence in New York City. Boston: Beacon Press. Kellogg, Charles Flint (1967) NAACP: A History of the National Association for the Advancement of Colored People, Vol. 1 (1909–1920). Baltimore: Johns Hopkins Univ. Press. Kramer, Larry (2004) The People Themselves: Popular Constitutionalism and Judicial Review. Oxford, United Kingdom: Oxford Univ. Press. Lewis, Anthony (1991) Make No Law: The Sullivan Case and the First Amendment. New York: Vintage Books. McCann, Michael W. (1994) Rights at Work: Pay Equity Reform and the Politics of Legal Mobilization. Chicago: Univ. of Chicago Press. Murphy, Paul (1979) World War I and the Origin of Civil Liberties in the United States. New York: W. W. Norton. Nelles, Walter (1919) Letter to Roger Baldwin, 11 Dec., Directing Committee Correspondence, ACLU Papers, Volume 120, Seeley Mudd Library, Princeton University Princeton, NJ. Phillips, Harlan B. (1954) Reminiscences of Roger Nash Baldwin. Oral history, Columbia University Oral History Research Office Collection, Columbia University, New York.
Zackin
395
Pickerill, J. Mitchell (2004) Constitutional Deliberation in Congress: The Impact of Judicial Review in a Separated System. Durham, NC: Duke Univ. Press. Post, Robert, & Reva Siegel (2003) ‘‘Protecting the Constitution from the People: Juricentric Restrictions in Section Five Power,’’ 78 Indiana Law J. 1–45. Rabban, David (1997) Free Speech in Its Forgotten Years. Cambridge, United Kingdom: Cambridge Univ. Press. Rosenberg, Gerald (1991) The Hollow Hope. Chicago: Univ. of Chicago Press. Siegel, Reva (2001) ‘‘Text in Contest: Gender and the Constitution from a Social Movement Perspective,’’ 150 University of Pennsylvania Law Rev. 297–35. Tushnet, Mark (1987) The NAACP’s Legal Strategy Against Segregated Education, 1925. Chapel Hill: Univ. of North Carolina Press. FFF (1999) Taking the Constitution Away from the Courts. Princeton, NJ: Princeton Univ. Press. Vose, Clement E. (1957) ‘‘The National Consumers’ League and the Brandeis Brief,’’ 1 Midwest J. of Political Science 267–90. Walker, Samuel (1990) In Defense of American Liberties: A History of the ACLU, 2d ed. Carbondale: Southern Illinois Univ. Press. Whittington, Keith (1999) Constitutional Construction: Divided Powers and Constitutional Meaning. Cambridge, MA: Harvard Univ. Press. Willrich, Michael (2003) City of Courts: Socializing Justice in Progressive Era Chicago. Cambridge, United Kingdom: Cambridge Univ. Press. Witt, John (2004) ‘‘The Internationalist Beginnings of American Civil Liberties,’’ http:// ssrn.com/abstract=539802. Zanger, Martin (1969) ‘‘Politics of Confrontation: Upton Sinclair and the Launching of the ACLU in Southern California,’’ 38 Pacific Historical Rev 386–406.
Statutes Cited Espionage Act, 40 Stat. 219 (1917). Sedition Act, 40 Stat. 553 (1918). Selective Service Act, 40 Stat. 76 (1917).
Cases Cited Abrams v. United States, 250 U.S. 616 (1919). Debs v. United States, 249 U.S. 211 (1919). Frohwerk v. United States, 249 U.S. 204 (1919). Gitlow v. New York, 268 U.S. 652 (1925). Guinn v. United States, 238 U.S. 347 (1915). Lochner v. New York, 198 U.S. 45 (1905). Near v. Minnesota, 238 U.S. 697 (1931). Schenck v. United States, 249 U.S. 47 (1919). Scopes Stromberg v. California, 283 U.S. 359 (1931). Whitney v. California, 247 U.S. 357 (1927).
Emily Zackin is a Ph.D. candidate in politics at Princeton University. Her dissertation focuses on positive rights in U.S. state constitutions.
396
Popular Constitutionalism’s Hard
397
Think of the Hippopotamus: Rights Consciousness in the Fat Acceptance Movement
Anna Kirkland
All the recent attention to the so-called obesity epidemic provides a fascinating context for understanding interactions between civil rights consciousness and the ordinary lives of fat people, who both deploy and resist the ideological formations that make up our most basic presumptions about who deserves rights protections. This study of fat acceptance advocates asks how stigmatized people who are excluded from legal protections muster descriptions of themselves as deserving inclusion in antidiscrimination laws. Analysis of in-depth interviews with fat acceptance advocates from around the United States reveals elaborate techniques for managing social life and enacting legality that coexist with more narrowly framed and contradictory arguments for rights. Culturally dominant logics for reasoning about what persons deserve prefigure what is possible to say in defense of fat people, in many ways even for fat advocates themselves. And yet in their struggles to overcome the limitations of the presumptions they are given, fat advocates reveal deep tensions in our antidiscrimination ethics and hint at a new way to think about difference.
I do a lot of swimming and I get in the water and I just feel like a total ballerina in the water. I’m very buoyant and graceful and amazing in the water. But then when I’m on land, I feel very clumsy and large and awkward. I feel just the opposite in the water. I absolutely love being in the water. On the Discovery Channel, I always think of the hippopotamuses. (Vicky, a 47-year-old homemaker from Massachusetts and fat acceptance group member) This work is only possible because many people were willing to take the time to be interviewed, and I thank each of them for their time and willingness to open up about sometimes painful experiences. I would like to gratefully acknowledge generous funding for this project from the University of Michigan Institute for Research on Women and Gender (IRWG). My graduate research assistant, Carla Pfeffer, did extensive interviewing as well as transcribing and site observation. I am grateful for those many hours and for her thoughtful feedback on the initial interview protocol and on previous drafts of this article. Kathy Wood also provided excellent transcription services. Tom Burke, Paul Campos, Elizabeth Cole, Kjerstin Elmen-Gruys, William Haltom, Don Herzog, Ann Lin, Abigail Saguy, Tobin Siebers, Jeffrey Sobal, Peggy Somers, Miriam Ticktin, Marilyn Wann, Elizabeth Wingrove, and several anonymous reviewers at Law & Society Review offered helpful feedback at various stages of this project, improving it very much. Please direct correspondence to Anna Kirkland, 204 South State Street, Ann Arbor, MI 48109-1290; e-mail:
[email protected].
Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
398
Think of the Hippopotamus
Why Study Fat Acceptance?
E
veryone is talking about fat people.1 Contemporary American society is experiencing a period of intense media attention and general cultural anxiety over the so-called obesity epidemic. News reporting on Americans’ increasing weights spiked dramatically by the late 1990s (from only a few dozen stories early in the decade to thousands) and tend to dramatize, moralize, and individualize body weight, particularly when referring to higher weights among minority populations (Saguy & Almeling 2008).2 Being fat is highly stigmatized (Brownell et al. 2005) and, following the tone of mass media coverage, most people attribute it to individual choices and behaviors (Oliver & Lee 2005). Editorials have blamed obesity on ‘‘feminist careers’’ because women prepare fewer home-cooked meals these days (Saguy & Almeling 2008:68), and increasing weights have been used as a metaphor in decrying the ‘‘cultural decline from sturdy thrift to flabby self-gratification’’ as consumer indebtedness drives up bankruptcy rates (Lears 2006:13). Attention to obesity supports narratives about the decline of American society from across the political spectrum. On the right, decrying obesity leads naturally to calls for greater personal responsibility and aids in dramatizing the dangerous expenses of national health care. On the left, critics target Big Food, corn subsidies, and junk food advertising to children for making people fat, linking the trend to corporate greed and overconsumption generally. Scholars from many disciplines have taken note of the political, historical, legal, moral, and disciplinary components of our national focus on fat (Gard & Wright 2005; LeBesco 2004; Oliver 2006; Saguy & Riley 2005; Sobal 1995; Stearns 1997). Critical social scientists maintain that the public is being systematically misled by news reporting about obesity as well as by misinformation about the safety and efficacy of diets and weight loss pharmaceuticals. In this view, the supposed epidemic is certainly not worth all the frenzy and looks much more like a moral panic (Campos 2004; Gaesser 2002; Saguy & Riley 2005). Rhetorics of individual blame and the blend of moralism and attention to self-care termed ‘‘healthism’’ (Crawford 1980; Greenhalgh & Wessely 2004) work
1 I use the word fat as a descriptor following many of my interviewees, who want the word to become just an ordinary term such as tall or dark-haired. Medical researchers and the media overwhelmingly use obese or overweight to indicate a medicalized conception of fatness and its undesirability, respectively. I still use the term obesity now and then to match these mainstream contexts. 2 Gina Kolata’s reporting in The New York Times is a notable exception (2002, 2003, 2004a, 2004b, 2007a).
Kirkland
399
together with media representations to constitute ‘‘common sense’’ about obesity trends, to echo Haltom and McCann’s use of the term (2004:10). Everyone simply knows that Americans are fat and getting fatter and that the trend portends disaster. Concern over obesity trends (and the trends themselves) are worth our careful study for many reasons, but sociolegal scholars have not yet paid adequate attention to the ways that fat works as a sociolegal problem to naturalize and reinforce particular formulations of rights-claiming. The crisis of fat is a compelling contemporary example of a ‘‘macro-contextual factor’’ shaping mass consciousness (McCann 2006:xxv). Attention to obesity instructs us in proper citizenship, stewardship of our bodies, and in what we can rightfully expect from law and the state (as well as the opposite: what we ought to do for ourselves by willpower, habituation, or good character). Law’s durability and ideological power, I suggest, comes in part from its widely accepted accounts of the deserving person as a properly functioning and responsible individual, free of disease, excess, or disablement. This is the person who can work, who looks after him- or herself so that others do not have to, who is health-conscious, and whose body and movement in the world conform to a normalized standard. Mass cultural production of the unhealthy and morally decrepit fat person undergirds an account of what kind of person is undeserving of rights protections. Despite all the attention, the voices of fat people themselves are rarely heard. Fat men and women are presumed to be in pursuit of weight loss and literally hoping to disappear as fat people. What if scholars reimagined them as citizens with claims to justice based on their status as fat? How would they make arguments for rights? Is resistance to the ‘‘lore’’ about fatness possible, and if so, in what terms?3 What do the ways fat acceptance advocates defend themselves tell us about our regime of civil rights protections? My aim here is first to understand more about how these stigmatized peopleFwhose status has been the subject of so much intense public attention but who are not included in any major antidiscrimination law, receiving only scattered local protectionsFexplain themselves and what justice looks like to them. To that end, this article analyzes in-depth interviews with fat acceptance group members from across the country and also draws upon participant-observation and the study of primary source documents from the National Association to Advance Fat Acceptance, or NAAFA. Second, 3 I borrow the term lore from Haltom and McCann (2004), along with common sense, to emphasize the heavily mediated and constructed nature of our mass cultural knowledge about obesity. My perspective is a skeptical one, though for the purposes of this research the reader need only consider that obesity does not just exist out there in the world as a property of bodies, unmediated. It has a history, a story, a causal narrative, and a moral edge in addition to being the site of much scientific knowledge production.
400
Think of the Hippopotamus
studying fat advocates is a way to ask about the ideological power of the law. How do those who defend fat people confront such a dominating cultural story about fat identity? Does the lore about obesity set the terms for advocates’ arguments that they are deserving citizens, and if so, how? Is there a way to move from a context in which, as Vicky puts it, fat people are awkward and illfitting, to a world in which they can be graceful and amazing? Are there other terms for advocates’ arguments, perhaps mustered from the fat rights movement itself or from comparisons with other antidiscrimination categories? This study combines a recognition of the ways that mass attention to obesity has mattered for legal consciousness with a close study of the terms available for fat citizens to defend themselves against the lore that they are unhealthy, costly, and gluttonous.
Civil Rights Consciousness: Does Fat Fit? As I noted, fatness is a highly stigmatized trait that is widely regarded as the product of individual fault (Brownell et al. 2005). The idea that fat people should be protected under antidiscrimination laws is most often met with derision. An editorial by management-side employment lawyers critical of a Massachusetts bill to add height and weight to the state’s antidiscrimination law scoffs at the idea of legal protection ‘‘based merely on ‘weight’Fas if weight were immutable and worthy of protected status on par with an individual’s race or sex’’ (Feldman & Ashton 2007:A11). Another critic likens the proposal to opening the door to all kinds of silly and burdensome rules that have lost any connection to real oppression: ‘‘We might as well add colorblind, left-handed, allergicto-cashews, and get it over with’’ (Maguire 2007:B13). National annual spending on weight loss products and services reached $46 billion in 2005 and has been growing at about 6 percent per year (Terlep 2005), evidence that many Americans are very keen on getting rid of this trait (and persist in hoping they can) rather than on transforming its social meaning or making life easier or fairer for fat people. Even some of my interviewees had only recently thought of themselves as victims of discrimination. As one woman put it, ‘‘I didn’t think of myself as being discriminated against when I was being made fun of. It was just being made fun of.’’ When defenders of fat people evoke civil rights rhetoric, it is often to bemoan fat’s exclusion from the pantheon of protected traitsFit’s ‘‘the last legally acceptable form of discrimination’’Fshowing that even they understand it primarily by its exclusion from the standard list of appropriately protected traits (e.g., Flanagan 1996).
Kirkland
401
Fat is considered quite unlike the traits usually protected in civil rights laws: race, religion, sex, national origin, age, and disability. Protected traits are classically those that bear a recognized history of oppression and are understood to be outside the realm of personal choice, irrelevant to one’s merit and capacities, and in the case of disability, a lamentable affliction to be overcome with gumption and equal opportunity. Fatness, by contrast, is understood as either analogous to smoking (i.e., an unhealthy and deadly condition brought on by behaviors but, once put in place, very difficult to change) or simply as physical evidence of overeating and laziness (primarily an aesthetic problem on this interpretation, but the health critique follows quickly behind). People who see a genetic link to obesity are more likely to think that civil rights protections for fat people would be appropriate (Oliver & Lee 2005:943) because according to that view it is an immutable trait one cannot help. Otherwise, the way to avoid discrimination is to just lose weight. Antidiscrimination law conceptualizes unfairness as rooted in governmental classifications that single out and burden groups of people without sufficiently good reason (in the case of equal protection) or in employment decisions based on protected traits (under Title VII of the 1964 Civil Rights Act). The dominant conception of what is gained is the right to be judged as an individual based on what is really relevant and important to the context, not based on stigmatizing, distracting, or irrelevant traits (Post et al. 2001). This person is what I call the functional individual, whom I have argued elsewhere encapsulates the dominant logic of the person in American antidiscrimination law (Kirkland 2008).4 The proper way to evaluate her is to measure her capabilities (that is, her functioning in a job or role) and to keep group-based stereotypes at bay (hence the emphasis on the distinctness of the individual). I purposely use the feminine pronouns throughout because my sample was overwhelmingly female and all the advocates quoted here are women. I devised this follow-up study to chart how ordinary fat people would navigate these dominant legal logics, such as functional individualism, which are also very important for ordinary understandings of justice. As I show, many of my interviewees turned out to be deeply invested in this vision of just treatment in which the body and its abilities are dissociated completely. As one interviewee put it (echoing nearly all the 4 Scholars of legal consciousness have more frequently used the term schemas to describe the resources ordinary people use to enact legality (Ewick & Silbey 1998; Kostiner 2003). McCann regularly invokes the term logics to refer to forms of legal consciousness through which activists make arguments, which is also the sense in which I use it (e.g., McCann 1994).
402
Think of the Hippopotamus
others), ‘‘Body size has nothing to do with competence or value, ability to do a job. So I think it should be illegal to discriminate.’’ The fact that ordinary peopleFfat acceptance advocates who have given a great deal of thought to why they are deserving citizensFinvoke a logic that is dominant in both the formal legal realm and the realm of citizen-to-citizen discussion is not itself surprising. What is fascinating to watch here is how the advocates must struggle with the contradictions inherent in trying to separate one’s body and how it looks from doing a job. The interviewee’s statement is, after all, both exceedingly common and deeply perplexing. Of course one cannot predict intelligence by examining the dimensions of someone’s body or whether she possesses the level and type of training or experience to do a job. Yet the body is how and where we work (and function) in the most fundamental sense. Many skills are dependent on certain operations of the body (strength, reach, dexterity, endurance, and capacities such as typing, walking, speaking, or fitting into a certain space). Some jobs require that a person’s body project an embodied message, such as sensuality, humor, neatness, fitness, perkiness, youth, masculinity, or femininity. Is body size more like eye color, which almost never matters, or more like gender, which is ‘‘just the way a person looks’’ but also so much more? One cannot just be an individual, evaluated as if these social messages were not always constitutive of the personFthat is, not if we think that traits such as gender have real meaning attached to them. Employers are almost always permitted to enforce these bodily requirements under Title VII through dress and grooming codes or by arguing that presentation is a necessary part of the job (Kirkland 2006b; Post et al. 2001). That is because we do not really mean that we want to ignore these supposedly irrelevant traits all the time. What we really mean is that we want to transform their social meaning, stripping them of some implications (for moral character, for instance) while retaining others (where we feel there is a legitimate link to an aspect of personhood we want to elevate, such as functional capacity). Disability law, by contrast, captures discrimination against people whose bodies are different in certain medically defined ways, and thus employment protections based on disability do not begin with the pretense of ignoring a person’s embodied appearance. A disability rights frame acknowledges that the rules about proper functioning are not themselves neutral, ahistorical, or nonpolitical. A disability is then not something that is just wrong with a person, but rather it is a site of difference that exposes hegemony and injustice in the normal workings of the world. The problem is the stairs, not the legs of the person who uses a wheelchair to get around. This view, termed the social model of disability by scholars in the field, has not filtered into the general public consciousness but
Kirkland
403
remains a founding insight of disability studies (though as I explain below, it is subject to critique for losing sight of the body in the ‘‘social’’) (Shakespeare 2006). Could there be a new source of civil rights consciousness for fat identity there? Could fat be a trait that prompts us to think more deeply about what it is to both notice something different about a person and at the same time to assess that person fairly? At the moment, however, fat does not fit in the antidiscrimination pantheon because it cannot be squared with functional individualism if fat is unhealthy and folks just bring it on themselves. Fat does not fit into the Americans with Disabilities Act (ADA) framework if it is not an impairment on its own (and the trend seems to be courts holding that it is not [Equal Employment Opportunity Commission v. Watkins Motor Lines 2006], though there is one well-known case saying it can be [Cook v. State of Rhode Island Department of Mental Health, Retardation, and Hospitals 1993]). Currently, then, those who try to come up with legal arguments for fat rights must argue that it is irrelevant and without meaningful implications if they want Title VII-type protections, but that it is a medicalized impairment if they want disability accommodation (Solovay 2000). It is a fairly impossible position to be in. Fat advocates know this only too well. Fat citizens’ lack of fit into antidiscrimination laws provides a rich perspective on what is required to fit, or at least, what citizens think is required to fit. For those who are quite publicly excluded from legal rights, the bare fact of legal exclusion requires explicit positioning. Hull shows that gay and lesbian couples enact legality in their relationships through ritual and the use of terminology such as spouse despite being excluded, while at the same time being painfully aware of the cultural power of legal recognition they lack (2006:142–9). Legality, as Ewick and Silbey explain, is what people both draw from and contribute to as they ‘‘participa[te]Fthrough words and deedsFin the construction of legal meanings, actions, practices, and institutions’’ (1998:247). Critically, citizens are ‘‘constrained by what is available, by legality as it has been previously enacted by others’’ (Ewick & Silbey 1998:247). Fat acceptance advocates must enact legality bodily, as they move about the public sphere and go to work, and also discursively, as they describe their personhood in terms that help them fit into a pregiven description of a deserving person. Their identities push up against deep ambiguities in our understandings of what civil rights protections ought to be for and for whom. Fat advocates lack the cultural and discursive resources that come from national legislation, as the men and women with disabilities in Engel and Munger’s study of the interactions of rights and identity had (2003). Despite never filing a lawsuit, the people they interviewed experienced identity transformations as legal change opened up new opportunities and offered new discourses to understand what they deserved out of life and work. So far Washington, D.C.; San Francisco;
404
Think of the Hippopotamus
Santa Cruz, California; and the state of Michigan have added height and weight or (in D.C.) appearance to their list of legally protected traits (District of Columbia Human Rights Act 2007; Compliance Guidelines to Prohibit Height and Weight Discrimination 2007; Santa Cruz Municipal Code 2007; Elliot-Larsen Civil Rights Act 2007). There are only a few recorded cases invoking the Michigan law, which dates to a 1978 legislative session in which a Democratic coalition put as many new traits into the state law as possible. At that time, the concern was for minimum height and weight restrictions that kept women out of male-dominated jobs (later taken care of nationally through disparate impact lawsuits under Title VII of the 1964 Civil Rights Act). There was no thought to fat rights at the time, and very little discussion of the new language. The San Francisco ordinance, by contrast, was passed in May 2000, after fat activists protested a fitness company billboard with an antifat message and secured the crucial support of municipal supervisor Tom Ammiano. As Marilyn Wann, the activist who organized the protests, explained to me, the time was right politically and the elements came together for an explicit act in defense of fat rights. So while the few Bay Area residents in my sample share a sense of mobilization, the Michigan fat activists have only a distant knowledge that a law exists, which none have used formally and only a few have invoked informally. Fat acceptance group members are emblematic of outsiders looking into the law. They are highly conscious of their exclusion from the formal law nearly everywhere in the United States; they are the subject of high-pitched national media coverage that reinforces the reasons for that exclusion; and they construct their identities using many of the same ideological commitments and meaningful practices that seem to denigrate them, finding little purchase for alternative arguments. They must wrestle with the question of whether being fat is the result of personal choices, and they must contend with the fact that their bodies differ from the norm (without much hope of ‘‘passing’’). Fat, in other words, interrogates the same issues of choice and bodily difference that we see in disputes over gay rights and disability rights. Fat advocates, as I show, exemplify a highly constrained and defensive enactment of legality, but one that must be enacted publicly and with as much dignity as one can muster, again and again, every day, in and through a body that no one will ignore.
Methodology: What Is the Fat Acceptance Movement and Who Joins It? The targeted population for sampling was defined as members of fat acceptance organizations, operationalized as anyone who had been a member or leader in NAAFA. Founded in 1969 and bearing
Kirkland
405
an acronym reminiscent of the NAACP, the organization balances its civil rights orientation with concern for members’ social lives. The official account on the NAAFA Web site is capacious: it is a ‘‘non-profit human rights organization’’ working to ‘‘eliminate discrimination based on body size’’ and provide ‘‘self-empowerment’’ (National Association to Advance Fat Acceptance 2008a: n.p.). The preamble to the NAAFA constitution contains explicit civil rights language: ‘‘Millions of fat Americans . . . constitute a minority group with many of the attributes of other minority groups: poor self-image, guilt feelings, employment discrimination, exploitation by commercial interests, and being the subject of ridicule’’ (National Association to Advance Fat Acceptance 2008b: n.p.). The conference program from the 2005 NAAFA convention illustrates the organization’s different aims in interesting detail. The greatest number of workshops offered (17) focused on activism and political change (including a feminist caucus and a workshop on changing the corporate workplace).5 The next most popular category was health and fitness, with 11 workshops including water aerobics, dance, and diabetes management. There were eight arts and crafts workshops, six personal care sessions on topics such as travel and ‘‘reclaiming our power to love ourselves,’’ and four sessions on sexuality (with titles such as ‘‘men who love fat women’’ and ‘‘everything you ever wanted to know about fat sex but were afraid to ask’’). Conferences also featured a private pool party, a fashion show, and a dinner dance.6 More politically minded leaders such as Wann have been trying to push the organization in a more activist direction in recent years, and the abundance of overtly political workshops shows that they have had considerable success. So while I do not claim that NAAFA members are necessarily devoted to rights-claiming (and in fact they may attend conferences for all the other fun things), I nonetheless stipulate that it is the most important and most prominent organization pushing against the dominant conceptions of fat people. That it does so in multifarious and complex ways should not surprise us. I do not go to the lengths of Luker (1984) to include all major leaders of the movement, so like Kostiner (2003) I do not make any claims of representativeness or comprehensiveness in the sample of members. Leaders of local NAAFA chapters were identified and 5 These are my characterizations of the workshops’ content, not groupings in the program itself. 6 The conflict within the organization between fulfilling members’ desires for meeting each other for sex and others’ interests in political organizing played out in fascinating and concrete ways at the convention. Because of the multiple purposes of the meetings, attendees wore a name badge upon which one could affix different-colored dots that revealed sexual orientation, whether one was looking to meet people (possibly sexually), or whether one was in a relationship (i.e., not interested in liaisons).
406
Think of the Hippopotamus
contacted through chapter Web sites. This recruitment method was probably more likely to capture more politically minded leaders in NAAFA chapters than more casual ‘‘social’’ members. Snowball sampling yielded additional interviews from members of similar groups. Of course, by virtue of joining a fat acceptance organization, these men and women are preselected for positive views about being fat. I do not therefore investigate whether or not they support fat rights. I am interested in the logics they use as they describe their fat acceptance, and what those logics reveal about what is necessary for rights claiming. Interviews took place between May 2005 and January 2006. Because of the distances involved, most interviews took place by phone.7 The final sample contains 35 interviews of fat acceptance group members from nine U.S. states (California, Illinois, Michigan, and New York were well represented because these states have active NAAFA chapters), and one from Canada. Four are men and thirty-one are women. They are overwhelmingly white, educated, middle-aged and middle-class, though the sample includes one female fat admirer (FA, a person who desires a fat partner) working as a janitor, one African American male FA working as a truck driver, one Latina, and one selfidentified white lesbian. It is unlikely that a larger sample would have been much more diverse, as the tendency of fat acceptance groups to contain mostly white fat female members and thinner white male FAs (all assumed to be heterosexual) has been well documented (Gimlin 2001; Goode 2002; Millman 1981).8 Fat acceptance group members are not the classic powerless subjects of rights featured in work on the welfare poor or the homeless (Cowan 2004; Gilliom 2001; Sarat 1990); they are more usefully compared to unionized workers or social justice activists (Kostiner 2003; McCann 1994; Polletta 2000). Because, as Ewick and Silbey point out, ‘‘social marginality is related to counterhegemonic consciousness’’ and my interviewees are not socially marginal (aside from their fatness, which many experienced as very marginalizing), to select this group to articulate the views of fat people is likely to preselect for more mainstream views (1998:234). Their integration into mainstream American culture makes their struggles with rights claims all the more compelling, however. They are dedicated 7 I am not considered fat, but my research assistant, Carla Pfeffer, who conducted half the interviews and attended the NAAFA conference, identifies as fat and would be perceived as fat. Very few interviewees ever inquired about our own body sizes. Like Sturges and Hanrahan (2004), I was not able to determine any differences between the quality of the interviews obtained by phone and the few we each obtained in person, nor could we tell any difference between the quality of interview obtained by either one of us. 8 The few men interviewed here tended to be FAs rather than fat-identified themselves. Since my aim here is to understand fat citizens’ conceptions of rights, I focus on the women.
Kirkland
407
to the same sets of norms that are so dominant in antidiscrimination consciousness, yet cannot help but expose the rickety features and contradictions that sustain it. The absence of racial diversity in particular is important, given evidence that nonwhite men and women experience both body size norms (Kemper et al. 1994) and relationships to the law (Nielsen 2000) differently than middle-class whites do. Scholarship detailing the exact kinds of differences that come with minority racial identities is not consistent on either axis, however (Cachelin et al. 2002; Engel & Munger 2003; Fleury-Steiner 2004), so I cannot presume to know how my results might differ if the sample were more racially and ethnically diverse. This group of interviewees also cannot throw much light on the dynamics of fat identity in a nonheterosexual context, though it seems likely that sexual minority communities might have different views (and different from each other, particularly between gay men and lesbians). The lack of solidarity my subjects exhibited with disabled people is also likely related to the study design. Most people interviewed here reported themselves to be in the range of 250 to 400 pounds. My interview subjects would be recognized in most contexts as quite fat but for the most part were not housebound or users of scooters. Other research on NAAFA members with a wider range of sizes and abilities has found greater willingness to invoke disability as an analogy (Saguy & Riley 2005). Members were initially told that I was interested in hearing their reflections about being in the fat acceptance group without any mention of law or civil rights. Interviews consisted of first, general questions about how the person came to join the organization; second, questions about the person’s ‘‘experiences of unfair treatment’’9 and third, prompts to discuss the desirability of antidiscrimination laws protecting fat people and to explain why fat people should be included, the grounds for such laws, the effect of such laws on the person’s life, and any experiences with using such laws. The interview questions did not explicitly refer to law or discrimination until about halfway through, making it possible for me to watch for the interviewee’s invocation of law (or its absence) and to see what he or she would first discuss under the category of ‘‘unfair treatment’’ (Kostiner 2003; Nielsen 2000). Interviews lasted about an hour and were digitally recorded and transcribed verbatim. All names are pseudonyms (many selected by the
9 Ill treatment in health care settings was a major topic that arose when I asked about ‘‘unfair treatment.’’ I analyze those responses in a subsequent article (Anna Kirkland, ‘‘Revisiting What Rights Do: Fat, Health, and Antidiscrimination,’’ Studies in Law, Politics and Society (forthcoming 2008).
408
Think of the Hippopotamus
interviewees, some unusual). Quotes have been lightly edited for length and clarity.
Managing Fat Identity in Everyday Life This section describes in detail how the people I spoke with manage their spoiled identity (Goffman 1986). No one had ever filed a formal complaint over a weight discrimination issue (and of course, very few of them could). Yet the fat acceptance group members described a regular set of distinct techniques they used to constitute an identity that was consonant with being a good worker and an individual rather than a health statistic. They described a richly detailed construction of a public self through which they anticipated, managed, and settled eruptions of discrimination and harassment. The techniques described here are consistent with previous research about this group (Maurer & Sobal 1995; Sobal & Maurer 1999), though my aim here is to explain their connection to rights consciousness. The work on the self and the development of culturally shared techniques for being a fat person in public are interdependent, showing how a person who understands herself as deserving yet excluded can still muster a form of legal personhood. This legal personhood is constituted by its exclusion but simultaneously by its commitments to the basic ideological underpinnings central to our antidiscrimination regime. It is a public identity for fatness that is suffused with legality but also guided by, as LeBesco puts it, ‘‘pursuit of an inhabitable subject position for fat people’’ (2004:124). I call their techniques (1) moral instruction, (2) redirecting shame, (3) scanning, (4) positive self-presentation, and (5) ignoring the mistreatment.10 Moral Instruction
Moral instruction is when a fat person delivers a didactic comeuppance to someone who has discriminated against her or otherwise rebuked her in the past. Respondents mostly described moral instruction as moments when they showed a former adversary that he or she had misjudged a fat person. Moments for moral instruction function in much the same way as civil rights protections do: that is, instructing citizens that treating people badly or differently because of a trait they bear is wrong. A common context for moral instruction was after rejection in employment, as with Jacqueline, a white woman in her fifties from Georgia who is a plus-size beauty pageant winner and works as an office manager: 10
I thank Carla Pfeffer for suggesting some of these names for the techniques.
Kirkland
409
Three weeks later, they called me and asked me if I was still interested [in the job]. I was mad at first. ‘‘They had the audacity to call me back?’’ But I did take the job. And then as I showed what I could do and everything, the president said, ‘‘You know, you’re the best person that I hired.’’ And I said, ‘‘Then why didn’t you take me in the beginning? Was it because of my weight and how tall I was?’’ And he looked at me and he said, ‘‘Yes.’’ And I said, ‘‘You don’t judge a book by its cover.’’ And he said, ‘‘I learned my lesson.’’ And then he didn’t hire other people just because they were beautiful or skinny or whatever. He hired them on their capability of handling the job.
Jacqueline’s invocation of moral instruction (‘‘You don’t judge a book by its cover’’) explicitly echoes antidiscrimination requirements: hire capable people without prejudice based on irrelevant traits. Other instances of moral instruction took place in situations also recognizable as employment discrimination but were handled much more indirectly. Kristin, a 43-year-old sales consultant from the Chicago area, also used moral instruction, but in a very attenuated way reminiscent of Bumiller’s profile of victims of job discrimination who could not square the demands to level complaints with their own psychological resources (1988). Here she thought back on a job where the boss invited her back to his apartment for drinks and, when she refused, later fired her, saying she was too fat: They asked me to work for them again in a few years and I turned them down. So I guess maybe that was my way, you know, turning down going back to work for them. I guess I should have [reported him for the harassment]. I probably should have done something about that. But I was too young in my acceptance of myself at that point to probably really do something. Today I would probably call a lawyer. I don’t like to be sue-happy or anything like that, but I think I would have called a lawyer just to bring it out into the open.
This is fairly blatant illegal sexual harassment, but Kristin did not do anything formal about it. She attributed not having called a lawyer to insufficient self-acceptance and was quick to distance herself from those who are ‘‘sue-happy.’’ Her imagining of rights was entirely prospective (‘‘Today I would probably call a lawyer’’) and somewhat regretful (‘‘I probably should have done something about that’’). Her refusal to go back to work there was one of the most subtle accounts of moral instruction my interviewees described. She was aware that invoking discrimination law would have been the formal option (to ‘‘bring it out into the open’’), but she decided instead to disdain contact with the discriminators.
410
Think of the Hippopotamus
Redirecting Shame
Redirecting shame is what I call verbal responses to ill treatment. The technique is similar to moral instruction in that it is situational and interactive, but it is less ‘‘legal’’ in the sense of being less about teaching an antagonist the proper way to evaluate another person and more about showing off a new-found confidence in what NAAFA members called ‘‘snappy comebacks.’’ (Members reported talking quite a bit about how to handle these situations, and developing shared responses was an important goal for group meetings.) This technique involves moving shame and social disapproval back onto the person who initiated the situation and may bring in onlookers on the fat person’s side. Carol, a 53-year-old applications analyst from Indiana, has been both active and passive in humiliating situations (once having had a woman take food out of her grocery cart as she wordlessly looked on). She described one of the first times she dealt with humiliation differently: ‘‘I pulled out in front of a guy and he got out of the car, berating me calling me a fat pig and whatever else. And it was one of the first times I stood up for myself. I told him I’d rather be the fattest person in the world than be an asshole like him.’’ Janice, a 42-year-old Latina working for the Los Angeles County government, redirected shame at a fast food restaurant: A woman tried to cut in line in front of me at a McDonalds and I said, ‘‘I’m sorry, I was here first.’’ She just let out a barrage of fat insults at me. ‘‘You have such a fat ass.’’ And I said, ‘‘Yeah, I do have a fat ass, don’t I?’’ And what I found was that I was fine with it, the gentleman behind the counter was smiling ear-to-ear and she just got more and more frustrated because she couldn’t shame me for being fat.
Lillian, a 47-year-old fitness trainer from New York, also relished support from onlookers in a confrontation on the subway: Some kids in the morning, they stand in the subway, and they don’t want to let you out of the door. I saw a kid make a comment about me. I didn’t exactly hear what he said, was clearly a comment about my weight. And he wasn’t going to turn and let me out the door, like joking, let’s see if she can pass through this. And I slammed on his foot. And I said, ‘‘I guess I’m too fat to get through.’’ And I got applause behind me, because who the hell wants these kids? Nobody wants that.
Redirecting shame shows that there are resources for resisting the stigma of being fat in the most ordinary public situations. At best, the smiling McDonald’s employee and the applauding subway riders show that they do not agree that fat is shameful, that fat people should not eat fast food, or that fat people should not take public
Kirkland
411
transportation, and at least they come down against bullying fat people. Maybe some of them were fat themselves, or had fat friends or family members. After all, when 65 percent of us are overweight or obese according to government standards and when well over 90 percent of diets fail to keep weight off (Kolata 2007b), it is not so surprising when survey evidence shows a greater willingness in recent years to find overweight people attractive (Balzer 2005). These moments of solidarity find little reflection in mass media representations of fat. Nonetheless, they are clearly one component of interaction that, like the cultural and discursive shifts that created new possibilities for employees with disabilities (Engel & Munger 2003), may assist fat advocates in constructing a more habitable identity. Scanning
Scanning is a technique for assessing, surveilling, and planning one’s movements through the world to avoid discomfort and humiliation. I mean to use the term scan both in its literal sense, to cast a glance over a situation or a place quickly, but also more broadly to mean the kind of assessments and observations that one learns to make about how one will be received in new situations that then constitute expectations and behaviors in the ongoing present. The situation being scanned can be anything from a room with chairs to a new job opportunity. Alice, a 54-year-old teacher from the greater Chicago area, explained how it works: When I walk in a room, I automatically scan it for difficult areas, chairs that look sturdy, chairs that don’t look sturdy, chairs that look ample to fit in, chairs that look like they might be a squeeze, any possible physical parameters in the room that might cause me embarrassment or anything else. It’s just a split-second automatic scan. And I do it without even thinking. And thin people don’t do that, OK?
Renee, a 36-year-old human resources manager from Ohio, had a particularly well-honed plan for scanning. She described avoiding joining a health club and any other social setting in which she might ‘‘stick out or feel uncomfortable’’ because of her size: If I ever had a concern about would a seat be wide enough or would I be able to fit, I learned to call and say, ‘‘Um, I have some concerns about your seating. Can you measure how wide it is?’’ I kind of just said, you know what? I’m not going to be uncomfortable. I’m going to find out in advance. And then if they can’t accommodate me, then I’m not going to go there.
Scanning in the employment context can mean not applying for a new job or having to switch jobs to avoid an unpleasant person.
412
Think of the Hippopotamus
Scanning is then about assessing how hard it will be to overcome bad first impressions. Bianca, a 45-year-old project manager from Illinois who is trained as an engineer, explained how scanning had caused her to self-limit her career options: ‘‘I can’t decide if I’m maybe passed over or looked at differently because I’m female or because I’m fat. Career-wise [being fat has] made a difference. It makes it more difficult to change jobs because there’s definitely an initial impression that you have to overcome. So I’ve probably stayed with the company I’m at maybe longer than I would have.’’ Sometimes the lessons incorporated into a lifelong practice of scanning are extremely limiting and come from just one or two isolated experiences of humiliation. For Vicky, the homemaker from Massachusetts who loves to swim, an incident 20 years ago made her quit her job altogether: There were six or seven teenage boys who picked on me. I was absolutely humiliated. There were all these people around me and nobody did anything or said anything. I was absolutely horrified. Very soon after that, I left my job. Because I just couldn’t bear riding the [public transit] anymore. I do not anymore ride [public transportation]. Today, I’m a lot stronger. Today, I would probably stand up for myself. And I don’t know if speaking out or saying something would make any difference, but I definitely could stand up for myself better. Definitely.
For Vicky, scanning has eliminated one entire context of social life Fpublic transportationFwhere she expects continued humiliation. This implementation of the technique has become a form of bare self-preservation. Positive Self-Presentation
Positive self-presentation is a technique most clearly articulated by the fat beauty pageant contestants in the sample, but it is connected to the NAAFA teaching that self-acceptance is the first step toward empowerment. Positive self-presentation is a state of nurtured self-confidence in the face of fat prejudice. It was often discussed in therapeutic terms, though its performance was social and interactive. Its purpose is to present a self to the world that will provoke nondiscriminatory treatment from others. Jacqueline articulated this theme again and again in her interview to explain how she keeps weight discrimination from happening to her. ‘‘I think my positiveness, the way I carry myself, the way I act and my professionalism really got me to where I am today. I didn’t think of myself as a plus-size person. I thought of myself as a human being.’’ For Jacqueline, positive self-presentation prevents discrimination: ‘‘To me, I don’t care if you’re 10 feet tall or if you’re two feet tall or if you’re 5,000 pounds or 100 pounds, it’s how you carry yourself
Kirkland
413
and how you come across that has a lot to do with it. And if you have low self-esteem and you’re negative about yourself, it’s gonna showFwhether you’re black, white, purple, or green, it’s gonna show.’’ The intermixing of actual traits (100 pounds) with imaginary ones (being purple or green or 10 feet tall, or weighing 5,000 pounds) emphasizes Jacqueline’s commitment to individualism over any social meanings of traits in the world, which she tries to empty of meaning and render fanciful. She characterized those who complain about their ill treatment as using a ‘‘crutch’’: ‘‘You don’t have to use [weight] as a crutch because if you’re positive then you can do anything you want to do. As long as you stay positive, then you can do the job.’’ Alicia, a 32-year-old from Ontario working at a technical call center, described self-love as a personal choice: ‘‘I think that just the fact that I wanted to love myself no matter what that made me just stay strong in that department. I think [self-acceptance] is really a choice of the person.’’ Anna, a 43year-old sales consultant and pageant competitor from the Chicago area, explained how she was able to construct positive self-presentation even after being badly treated in her family: My dad said to me as a young teenager that I was so fat I’d never have anybody marry me or have sex with me. Very embarrassingly, he referred to me as a brick shithouse. My brothers and sister tended to be smaller. I was kind of isolated in terms of the family. I think that if there was job discrimination I think it was mostly in my own head. And I say that because where I am now with it is that I’m bright, I have the talent, I’ll show you that I have the talent and I’m bright. And if you don’t want to hire me, well I don’t want to be there. Which is pretty empowered.
Ignoring Mistreatment
Some respondents simply try to ignore ill treatment. When she could tell things were going badly in a job interview because of her weight, Foxglove, a woman in her sixties from Michigan who had spent her career as a civil servant, reported handling the experience ‘‘with disappointment, but I still kept a smile on my face. I still kept trying.’’ Frannie, a 62-year-old fundraising purchaser from the Bay Area, recalled ignoring discrimination at job interviews before the San Francisco ordinance was enacted (which now governs her workplace): And I will tell you that every time I’ve experienced discrimination, I have had to ignore it because either I’ve wanted another job in the place, wanted to continue my job, or there wasn’t a way I could fight it. And, frankly, at the time these interviews happened, I didn’t have the regulation in San Francisco. I had something
414
Think of the Hippopotamus
behind me to back me up, if I’d had that regulation behind me, I think I would have confronted. But I had nothing behind me.
Although Frannie described herself as nonconfrontational, the San Francisco ordinance is something she imagined stiffening her spine. Many other interviewees spoke of law as something ‘‘backing [them] up,’’ as a support ‘‘behind’’ them. It is a distant and instrumental conception of law, but they imagined that its presence would have transformed their sense of entitlement and self-worth. The San Francisco ordinance is the only formal law that was enacted with sufficient publicity to provoke the kind of cultural and discursive changes for fat identity that Engel and Munger (2003) describe happening after the enactment of the ADA; it will be worth keeping an eye on. These techniques present ways to both enact legality and evade mistreatment and shift tactics in recognition of the absence of a formal law as a means of protection. Nearly every interviewee described using at least one of these techniques, and most used them in combination on a daily basis. Moral instruction is a way to secure a just outcome that feels like retribution or compensation and may even involve penitence from the discriminator. Positive selfpresentation and scanning are compliance rules for avoiding confrontation before it begins. Redirecting shame is a dispute resolution response that leaves the fat person feeling that she has won and, in the best of circumstances, harnesses some localized solidarity to make the perpetrator into the deviant one rather than the fat person. Ignoring the mistreatment settles the dispute through capitulation (and may also be used to avoid confrontation), and it also reflects awareness that one is outside even the shadow of the law. Moral instruction most clearly enacts the ethical norms of antidiscrimination law but ultimately depends on the discriminator making some move to woo back the rejected fat person (hardly a scenario to be counted on). Positive self-presentation keeps the burden on fat people to deflect discrimination against them. Scanning limits fat people’s lives in ways that are both invisible and difficult to quantify. Redirecting shame signifies the attainment of a fierce pride that is a precondition to feeling entitled to rights, but it is psychologically costly to shyer fat people and unlikely to provoke widespread acceptance or understanding of fat troubles in those who would harass fat people. Simply ignoring mistreatment is a time-honored response to injustice that sociolegal research discovers over and over again; it is no more empowering here than it has ever been (Bumiller 1988). Because antidiscrimination protections are so sparse for this group, formal use of the law is not much of a possibility. But Ashley, a 52-year-old white woman, lives in Michigan and has referred to
Kirkland
415
her state’s height and weight antidiscrimination provision to convince people to provide accommodations such as armless chairs. She described how she has used the Michigan law instrumentally in advocacy, even looking it up to see if it could really be true: Some time after I was working in Michigan I looked to see, you know, is it really in the statement? There it is, how cool! I brought it up, for instance, with the [armless] chair issue. I’ve done workshops at youth programs about size acceptance kinds of things and I’ll bring it up there. I don’t know that I’ve ever met anybody who knew it. [So you’ve used the Michigan law for leverage in some of your own advocacy for armless chairs?] Yeah. But not in a way I wouldn’t wanna say, ‘‘Hey, there’s a law.’’ It’s more in it’s that legitimacy and not, ‘‘That’s [Ashley] the advocate. Always bringing up weird stuff.’’ You know? It’s like, ‘‘No, it’s not me. Look at, there’s a whole law that addresses it.’’
For Ashley, the existence of the law on the books is a way to move from the illegitimate (‘‘weird stuff ’’) to the legitimate (‘‘there’s a whole law that addresses it’’). She was the only interviewee to describe encounters in which she called upon an available law. Interestingly, the Michigan law prohibits only animus and stereotyping of fat employees who can otherwise do their jobs, and it does not require any accommodations (and, as I noted, fat people were not the primary target of protections when it passed). So without realizing it, Ashley is being doubly inventive: using a law that no one thought would apply to fat people to gain accommodations that the law does not even require. So while it would be wrong to say there is no use of or longing for the formal law among fat acceptance advocates, in the absence of legal protections, they have evolved these techniques for getting by in a hostile world. These techniques work as conduits for reimagining the selfFnot a self-hating fat person, but a confident woman whom no one even considers discriminating against. Perhaps these outsiders are showing us another side of the psychological self: not just the new person of late-modern liberal governmentality (e.g., Rose 1990), but also a site of small resistances. Again, it is critical that the focus of my analysis is a highly visible and much-stigmatized group that is presently the focus of an international moral panic. They must act on their own to punish, teach, and organize against discrimination.
Mustering Legality: Using the Master’s Tools Defensively Is Being Fat the Result of Personal Choices?
The next phase of the interviews probed explicitly for advocates’ reasons for thinking that fat people should not be
416
Think of the Hippopotamus
discriminated against. Most immediately took up the ‘‘fat is unhealthy’’ argument. Fat had to be transformed from evidence of bad habits into a protectable trait. As Frannie put it: ‘‘We have to address the issue of health, because so often people will say to you, ‘Well, I can agree with you about discrimination against fat people is wrong when they’re denying jobs.’ But what about your health? And I really want to say, ‘Well, what about it?’ That’s usually a way to discriminate against us.’’ This process is a fundamentally defensive one, in which fat advocates must use the ‘‘master’s tools’’ to construct their own plausible inclusion. Are people fat by choice, or is it more like something you are born to be? Can a fat person just lose weight and thereby avoid discrimination? Janice made the argument that in fact many fat people pursue the same ideal lifestyle supposedly reserved for the thin and the upper-class: ‘‘A lot of fat people are not suffering from any illness. They’re able to get up and live life actively and healthfully. They go out and do things, they eat decent, healthy foods, vegetarian lifestyles that are accepted among thin people as a healthy lifestyles. Simply because they’re fat, they’re considered unhealthy.’’ Nearly all respondents conceded that fatness was linked to voluntary behaviors such as eating but also insisted that most people are predetermined to fall within a certain range of weight that cannot be changed without herculean effort. Many people readily acknowledged that losing weight is certainly possible. They had done it themselves many times. But they always gained the weight back eventually. Most women used the phrase ‘‘dieted myself up to [present weight]’’ to capture personal experience with the welldocumented effects of weight cycling (also called yo-yo dieting) (Ernsberger & Koletsky 1995). Most then reported that when they found self-acceptance, stopped dieting, and simply tried to be healthy, their weight stabilized. Macskat, a 41-year-old self-employed massage therapist from Michigan, went beyond the ‘‘set point’’ theory of useless dieting and simply said that the loss of relationships and the effort required to lose weight just was not worth it: I used to think people could lose weight. One time, I lost 90 pounds and another time I lost 55 pounds. And both times, when I got where I was, I was like, ‘‘This is not worth it.’’ I mean it was life-consuming both times. That was a year of not going out with my daughter. You know, we used to have dates and we’d go to the movies and dinner. I wouldn’t eat out because I didn’t want to eat anything off my little program. I mean it was sick. It was a whole year of not being social, of not having friends, of not, you know, doing anything other than exercising. I had my tennis buddies, but I didn’t really do any socializing that involved, you know, sharing meals with anybody.
Kirkland
417
Others drew explicit civil rights parallels that did not turn on proving the futility of dieting. As Monique put it, ‘‘I think that being made to purchase two seats for an airline is comparable to not allowing someone to sit at the front of the bus.’’ Many interviewees used analogies to other traits, explaining how fatness was or was not like being black or being gay. (‘‘Black’’ and ‘‘gay’’ were invoked most commonly by far; other identity traits received almost no mention.) Vicky’s argument was typical, first acknowledging the public perception of fat people as fat by choice, then drawing a race and sexual orientation comparison: I think that because people think that people are fat because they’re lazy and because they have a choice, that it’s different than any of the other reasons that people are discriminated against. In other words, you don’t choose to be black. To me, [being fat] is not a choice. Just like I believe that being gay or lesbian is not a choice. And it doesn’t matter whether it’s a choice or not. People still need to be treated with respect and dignity and to have their rights.
Michelle, a 63-year-old nurse who works in a cardiac rehab center, echoed the Protestant work ethic and the imperative of self-improvement in a comparison of fatness to poverty: ‘‘So if you think it’s a choice, you could think that fat people could be thin if they would just try. I guess the poor, you could say the same thing, you know? If they just worked harder, they’d be rich.’’ Foxglove was unusual in her quick reference to disability as a point of comparison: ‘‘It’s just as illegal to discriminate against fat people as it is against people with a broken arm, or a disease. HIV, you cannot discriminate against people for that.’’ As I discuss below, Foxglove’s experience with being listed as disabled and experiencing employment benefits she attributed to affirmative action for disabled people probably explains her unique perspective. Disability was not considered an applicable analogy among most of my respondents. Most were keen on pushing that label away from fat people because it complicated their arguments that fat people are fully functional and healthy. They wanted to be seen as functional individuals first and foremost, but after first invoking an account of their personhood in those terms, they often picked up other more capacious language that more deeply challenged dominant notions. Vicky’s turn against the very concept of non-choice as the sine qua non of antidiscrimination (‘‘It doesn’t matter whether it’s a choice or not’’) was a common polyvocalism among many interviews. So while many interviewees made these analogies to identity traits understood to be beyond choice (‘‘You don’t choose to be black’’), most also moved beyond the choice issue to stress the misery and disutility of dieting and then to describe a vision of
418
Think of the Hippopotamus
nondiscrimination that did not really turn on the absence of choice. My interviewees were certainly in a difficult rhetorical position. How can rights protections rest on an identity that is highly variable on a scale, not an ‘‘either/or’’ (as we often understand race and sex to be), which is both changeable in some sense yet often stable, which is related to both voluntary behaviors and to genetics, and which is constructed and defined within different and sometimes competing communities and in different dimensions (pharmaceutical advertising and media presentations, but also in ethnic and racial enclaves)? Is fat an unusually ill-fitting category for antidiscrimination law, or does it simply illuminate the ways that other identity categories must be understood as stable, binary, or imbued with a singular meaning when actually they are not, either? The women struggled to describe their bodies as simultaneously mutable and stable, as deserving of protection from discrimination despite the awkwardness of the ‘‘like race’’ comparison. An overwhelmingly white and middle-class group, they were tentative with ‘‘like race’’ comparisons and usually explicitly differentiated fat hatred from racial hatred even as they often mentioned the analogy. Nor was there much consensus that fat people are an identity group. Most people pointed out that while fat people may share common experiences, there are gradations of fatness and, moreover, most fat people want to get out of the group rather than celebrate it as a site of culture and knowledge (as disability rights advocates describe disability, by contrast). Interviewees were reluctant to speak of fat people as a coherent group that should be the subject of rights. Their facility with the web of theories, analogies, and assumptions that make up our society’s dominant ideology about antidiscrimination was quite impressive. The way they talkedFhesitations, arguments, quick comparisons, more pained comparisons, negotiationsFshowed how difficult it is to describe an identity that has not been adequately ‘‘made up’’ for antidiscrimination coverage. Their descriptions of themselves were channeled into a defensive posture in which they had to rely on formulations that did not wholly capture what they meant but needed to be invoked anyway. As I explain next, that defensive posture forecloses much engagement with disability, which, depending on how it is understood, could be the most interesting possibility for alliance. Functional Individualism and the Tension With Disability Rights
The predominant way of justifying nondiscrimination against fat people was to use the logic of functional individualism. As I have described, functional individualism is a way of reasoning about who
Kirkland
419
deserves rights that defines the deserving person as a font of capacities and talents who should be evaluated on those alone, not on any feature of appearance. It is a foundational idea in our antidiscrimination laws and larger culture, so I was not surprised to see its emergence in these interviews. Jacqueline, the plus-size beauty pageant winner with the positive attitude, typifies the general ethical impulse behind functional individualism: I don’t think you should evaluate or determine a person by their size or their weight or their height or anything. I think you should take the person for what they are and think of them as a personFeven if you are African American or Indian or French or Chinese. To me, everyone’s a human being and you treat ‘em as that personFas as a personFnot as, ‘‘Oh, you’re black or you’re Chinese or you’re fat or you’re tall or you’re small or you’re short.’’ You know, I don’t look at it. And I feel that everyone else should look at the person as a personFnot as a size.
Alice’s view of employment nondiscrimination exemplifies the way functional individualism works in a job context: ‘‘A person should be judged on their abilities and if they are able to do the job that is set before them. And those are the only considerations that should be brought into play: what are their abilities, what is their training, what is their background, what can they do?’’ Functional individualism is incompatible with disability rights, since if a person cannot do a job because he or she cannot physically fit in the work space, for example, then there is no reason within functional individualism itself to make any changes (Kirkland 2003). As I noted earlier, the social model of disability would turn this individualism upside down, asking why the constructed world makes it possible for some to function and not others. In this view, disability is politically and socially created and is not some feature of abnormal bodies. Without this inversion, disability will be regarded through the lens of healthism or functional individualism, and ‘‘abnormal’’ or costly differences will not deserve to be ignored. Yet fat advocates and disability studies scholars are both deeply engaged with the actual movements, pangs, pains, and what Siebers calls ‘‘blunt, crude realities’’ of bodies that are different (2006:178). The fat people I spoke with cannot ignore their bodies, and as the previous sections described, they actually devote a great deal of agency toward techniques for living in them. What about those people we call disabled (fat or not) who have a lot of pain, who will die young, or who seek medical intervention or rehabilitation? What if a fat person wants to use a scooter for greater mobility, needs to adapt her home for greater comfort and safety in
420
Think of the Hippopotamus
bathing, or has joint pain?11 Disability studies scholars have sharply criticized the social model for moving disability too hastily to the social and the political realms. ‘‘In practice,’’ Shakespeare points out, ‘‘it is the interaction of individual bodies and social environments which produces disability’’ (2006:201). We should understand disability in a more complex and varied way, he argues, as both environmental and individual, embodied and social. Again, like the women’s multifaceted reflections on whether being fat is a choice, their articulations about disablement and bodily difference straddle this fault line: how can people talk about themselves as deserving both the dignities of functional individualism, which recommends ignoring their appearance, as well as accommodations for bodily difference, which confront it? And can they permit variable accounts of being fatFsometimes one’s fault, sometimes not, sometimes unhealthy, sometimes not, sometimes disabling, sometimes not? Given the tendency of discourses of disability to lapse into endless individualization, such a move might dissolve their collective political critique. But it would more accurately capture the full range of feelings my interviewees expressed, as well as draw them back from endorsing health and ability as the takenfor-granted starting point for deserving rights. I was interested in how my interview subjects would consider the tension between removing attention to the ‘‘abnormal’’ body or giving it needed attention, so I asked then about whether they agreed that fat people should receive accommodations at work such as armless chairs.12 Reactions varied, but most interviewees were either sharply negative about being considered disabled (even if it would secure more rights) or highly pragmatic about using the label disabled. As Marilyn Wann put it, ‘‘In the dark times, you use whatever you have.’’ While resistance to the disability label prevailed, there was wide agreement that businesses should provide armless seating and other provisions that would constitute disability accommodations. This inconsistency may look like a simple dodge or like bias against disabled people (eschewing the label but wanting the benefits), but I argue that we should consider why it was so difficult to expand the concept of disability rather than to push it away. Michelle, the cardiac rehab nurse, articulated a common-sense view of disability rights in which they are a matter of fairness to everyone that should be provided without regard to the medical status of obesity: 11 Incidentally, these kinds of personal care and mobility accommodations would not be required under the ADA because they are not job-related, and social welfare provisions for people with disabilities can be fairly stingy when it comes to assistive devices such as power chairs or scooters (Bagenstos 2004). 12 In general, obesity does not count as a disability on its own, but there are some interesting cases and trends suggesting otherwise (Kirkland 2006a).
Kirkland
421
Well, I don’t know should it legally be a disability or not. I know a lot of people feel strongly one way or the other about that. But I think accommodations should be made. I think we should have CAT scan machines big enough to deal with large people, we should have chairs in waiting rooms wide enough. So I think just as we have curbs where a stroller and a wheelchair can roll up them, why not have things that fit everybody?
Macskat had no patience with the disability label and, in a typical move, referred to what real disability is to distinguish it from just being fat. ‘‘I’m not disabled. And a lot of fat people I know aren’t disabled. We just have a bigger butt. We need a bigger chair. That’s just common sense and common courtesy. Not like my mother, who is a thin disabled woman who gets around in a wheelchair. That’s disabled.’’ Foxglove was unique in experiencing disability as a practical tool in her employment history, and also in her view that it would be ‘‘wonderful’’ to see fat people making ADA claims. (She worked in various civil servant positions in the Michigan state government, including social services and education.) A few times, I felt discriminated against because, when I would go to interview people would laugh at me when I walked in, or you know, make some disparaging remarks. But ironically, I was not able to advance until I actually got my name on the handicapper list for the State of Michigan. So at that point, I was put into interview with other handicappers that were in wheelchairs, and you know, crutches and malformed. And then I got a promotion [laughs] because I was the most functional of all of them that was interviewed. And so, it worked for me in a sense.
She explained that she had gotten on the ‘‘handicapper’s list’’ not for being fat but for a back problem. She had also successfully made a disability claim herself, initiated not in court but with a union grievance. ‘‘I had migraines. Heat and light would bring them on. I had hot, really strong fluorescent lights over [my work station]. I asked for them to be turned off because of the pain in my eye. They refused to do it. So through the American Disabilities Act [sic], I made a claim, and the union backed me up and within two weeks, they came and unscrewed the light bulbs [laughs].’’ Disability studies scholar Rosemarie Garland-Thomson points out how well fat and disability can fit together. ‘‘The fat body is disabled,’’ she observes, ‘‘because it is discriminated against in two ways: first, fat bodies are subordinated by a built environment that excludes them; second, fat bodies are seen as unfortunate and contemptible’’ (2005:1582). Even though fat rights attorney Sondra Solovay argues that disability law fits fat people too (2000:
422
Think of the Hippopotamus
145–9), a real alliance is quite far off.13 The scholarly critique of disability as stigmatized abnormality has not filtered out to the ordinary sensibilities of most fat acceptance activists. But Michelle is clearly advocating what’s known as universal designF‘‘why not make things that fit everybody?’’ She blends parents pushing strollers, users of wheelchairs, and people with large bodies together as simple variations on human needs, not as a medically demarcated class of impaired people who need help living in the world of ‘‘normals.’’ Macskat, by contrast, sees attention to different bodies (‘‘We just have a bigger butt’’) as normal and regulated by ‘‘common sense and common courtesy,’’ while those with wellrecognized impairments are properly set apart as the truly disabled. Foxglove seems to understand herself as masquerading a bit; she received a promotion because she is not ‘‘malformed’’ and still ‘‘functional,’’ though she has gotten the advantage of being on the ‘‘handicapper list.’’ Are disability rights for potentially everyone, or are they for a certain class of impaired people? If it is the latter, do they depend on a medicalized view of personhood, bringing with it notions of ill health and dysfunction, to demarcate that class? Even as their defensive posture means they must overwhelmingly reject being labeled ‘‘disabled,’’ fat advocates hone in on the undecided questions of disability rights in contemporary American law and society. Nondiscrimination as a Universal Ethical Imperative
The second most common formulation of deservingness after functional individualism was to say that of course nobody should be discriminated against. Speaking in terms of ‘‘nobody’’ or ‘‘anybody’’ rejects categorizing people into traits entirely, shifting antidiscrimination from its status as a gate-keeping mechanism for arranging concern for only certain operations of stigma to status as a universal ethical imperative. My exchange with Macskat was typical: [Do you think it ought to be illegal to discriminate against fat people?] Oh, absolutely. It ought to be illegal to discriminate against anybody. I would love it on the books if it were, ‘‘You cannot discriminate against a person, regardless of their sexual orientation, and they should be able to enjoy any privilege that
13 The ADA permits lawsuits based on being ‘‘regarded as’’ disabled even if one really is not, which has been a way for some fat citizens to win employment discrimination cases without having to claim that obesity is a disability. I explore these cases much more fully elsewhere (Kirkland 2008). Many ordinary people do not know about this prong of the ADA, however, which might help explain why the people I interviewed for the most part assumed that disability protection meant agreeing that there is something wrong with them.
Kirkland
423
any human being enjoys, including the privilege of marriage.’’ And I would like that to be like, you can’t discriminate against somebody, period. I would like simple laws like that. You can’t discriminate [laughs].
Vicky’s formulation is similarly universal, noncategorical, and presented as an ethical imperative: ‘‘I think all human beings have the right to be treated with dignity and respect. No matter what. And I don’t think it’s right that anybody should be treated that way either publicly or privately for any reason. And that’s it.’’ Both Vicky and Macskat expressed impatience with the idea that fatness must be shown to be entirely outside of one’s control. Vicky said she did not really care if getting fat was a choice or not, and Macskat rebelled by saying that losing weight was possible, but too unpleasant to be worth it. They transformed antidiscrimination law from functional individualism, with its delimited boundaries and usages, into an exhortation that they could use to describe a positive vision of a better world on a much grander scale than remaining within pre-given questions (‘‘Is being fat a choice that makes you unhealthy?’’) would permit. They found fitting into antidiscrimination norms exasperating and useless. Their frequent invocation of sexual orientation discrimination as an analogy is significant. A gay or lesbian person could go on a strict ‘‘diet’’ and refrain from same-sex contact, of course; it is just miserable, confining, and not worthy of a full and dignified life to have to do so. It is possible to measure a person’s job skills on an individual level without knowing her sexual orientation as well. But many gay and lesbian employees find that remaining closeted at work is an onerous burden, especially since heterosexual coworkers talk freely about their relationships. Deliberate fat activist strategies such as resistance to dieting or referring to oneself as ‘‘fat’’ in a straightforward and uncritical manner are refusals to ‘‘cover’’ fat identity (Yoshino 2006). Vicky’s and Macskat’s reflections show that refusals to cover can be deeply antagonistic to the core of our antidiscrimination tradition and may require developing new vocabularies that are currently very difficult to come up with. This universal ethical imperative is wholly inassimilable into the ways we currently reason about law. That is because antidiscrimination categories as traits one should ignore misses the whole point, as Macskat and Vicky would see it. The point is that one should be able to be fully oneself, in the body one has, eating, drinking, sitting in comfortably sized chairs, going out to dinners and movies with one’s daughter, and still be treated as a respected citizen. This is the world Vicky imagines herself in as a hippopotamus in the water: she’s still fat, but ‘‘graceful and amazing’’ because of the changed environment. It is not getting
424
Think of the Hippopotamus
accommodations to adapt to the other world; it is a shift in the terms of what is all around. Michelle’s view of disability similarly dodges the issue of parsing exactly who is and is not disabled in favor of just accommodating everyone. Here we see how the dominant logics in our antidiscrimination consciousnessFreasoning through the narrow list of analogous traits that should be ignored while true merit is measured insteadFdemarcate and sustain a very narrow range of imaginable injustices. Only harms that come to an otherwise normal, striving person can be fit in easily. That person is just like the other deserving people but for this one little irrelevant thing. While this criticism of antidiscrimination law has been articulated from many perspectives (Freeman 1995; MacKinnon 1991), we can see here some of the discrete discursive formations that sustain it in ordinary speech, and how even articulate and thoughtful advocates have great difficulty finding words to get around it.
Conclusion: Deterring and Channeling Arguments for Rights The national focus on obesity has provided a unique opportunity to study the invocation and deployment of legal consciousness in a very specific but crucial dimension: a moment of great attention (and extremely negative attention, at that) on a highly visible group of citizens who must then struggle to muster an account of themselves as deserving of rights, despite their near-total exclusion from legal protection now. Undocumented immigrants, terror detainees, and gay and lesbian couples seeking to marry are related groups in this sense who must also draw upon settled logics of dedication to work, civil liberties or religious freedom, or monogamous commitment, respectively. In these cases as well, reference to already-dominant logics may be politically expedient, but they submerge more difficult questions about how to deal with difference. Hardworking and eagerly assimilated immigrants may be well regarded, but what about those who really do transform the American national identity rather than embrace it? One may easily agree that terror detainees need access to lawyers, but the question of whether terrorism is better conceived of as criminal activity, as wartime aggression, as resistance, or as something else still remains. Heartwarming photos of couples lined up outside San Francisco City Hall to marry set aside the question of why marriage remains such a legally privileged status in the first place. This group of fat acceptance advocates has embraced a rights discourse that resonates strongly with the pre-existing antidiscrimination ethic. They sit uneasily within it, though, and their formulations
Kirkland
425
often set aside pressing dilemmas within our antidiscrimination tradition. Moreover, simply proliferating identity politics means replicating a strategy that has in the past produced bitterly limited demands for other groups while failing to achieve much politically (Brown 1995). Perhaps time spent so far outside of inclusion in antidiscrimination laws will mean that fat acceptance politics will develop more varied and creative avenues for protecting and celebrating fat people (like some being developed at the conferences). Only time will tell if fat advocates are presciently ahead of public opinion despite their exclusion from most formal legal protections, in which case, as Gould’s work on campus hate speech codes shows (2005), they might obtain concessions and protections at work, as consumers, and in public spaces without any formal legal account of their identity. For now, fat people’s liminal position outside civil rights protections has let us see the possibilities and limits for fat citizens to articulate why they should be let in. Studying legal subjects at the margins so often turns out to be a way to better understand the center. Antidiscrimination consciousness, as I have argued, is sustained at the center by a set of ideological commitments to a certain kind of deserving person. The description of this deserving person slides off the tongue quite easily, as does its accompanying logic of ‘‘just treat a person as a person, not a size.’’ One might think the biggest challenge is changing public opinion about whether fat people can reasonably lose weight, and then legal protections will soon follow. But I suggest in these concluding observations that fat advocates have shown us a greater challenge: how to push beyond the tensions that assail the center of antidiscrimination consciousness as we realize just how fragilely imaginary the truly deserving person is. Real questions of justice loom for fat persons, who are both like and unlike those imagined deserving ones. In fact, sometimes they are hard to tell from everyone else. Silbey suggests that perhaps the study of legal consciousness has run its course and drifted too far from its original critical focus on ideologies. She urges scholars to ‘‘recapture the critical sociological project of explaining the durability and ideological power of law’’ (2005:358). I have suggested here that the durability and ideological power of law is built out of the terms even oppositional activists must use to confront the status quoFcritically, out of the terms they do not use, the arguments that cannot be coherently formulated. Haltom and McCann argue that ‘‘legal lore contributes to hegemony to the extent that it sustains a pervasive takenfor-granted, commonsense knowledge on which the prevailing order rests’’ (2004:304). The common sense of antidiscrimination law is that it protects deserving people who can function just as well as anyone else from being unfairly judged based on an irrelevant
426
Think of the Hippopotamus
but historically stigmatized trait. ‘‘The common sense about law does not preclude or determine contests over legal meaning,’’ Haltom and McCann continue, ‘‘but it works to deter some conflicts while channeling others into safe, manageable trajectories and venues’’ (2004:304–5). Legal lore is an expression of liberal ideology that ‘‘defines both causes and evidence of failure in individualistic terms that efface attention to larger patterns of harm, systemic analysis of power relations that structure interaction, and collective responses to shared loss, gain, and aspiration’’ (2004:304–5). So how have contests over the legal meaning of fat identity been either deterred or channeled? How do fat acceptance advocates’ own arguments share in and react to those deterrences and dislocations? What have we lost in this channeling-off of fat rights claims? What can we see more clearly about civil rights consciousness generally? First, deterrence comes from mass culture as elites simply make the inclusion of height and weight clauses sound ridiculous. Lumping in weight-based discrimination claims with the ‘‘allergicto-cashews’’ folks (per Maguire 2007) reduces those claims to simple whines by overly sensitive people who do not really deserve rights. Of course, critics are quick to offer, we oppose discrimination against race and sex, but come onFfat people? The effect is to shore up and reify acceptance of ‘‘true discrimination’’ (though, as Herman [1996] points out, conservatives eager to parse those deserving of rights from those who are asking too much often fail to support measures devoted to the supposedly urgent problems of the truly deserving, as in the case of race discrimination, for example). Pointing out how undeserving fat people are holds up the ideal of the citizen who takes care of her own health through the cultivation of virtuous personal habits. A ‘‘collective response to shared loss,’’ as Haltom and McCann put it (2004:304–5; think of national health insurance), slips away in favor of the fantasy that we can each control our own bodies if we just try hard enough. Second, we have seen here how fat citizens channel energy into practices of self-management. These techniquesFmoral instruction, redirecting shame, scanning, positive self-presentation, and ignoring the mistreatmentFare sometimes empowering but often are not. They attempt to replicate but cannot replace what formal legal protections might provide. The few thousand NAAFA members in the country, led by a small cadre of activists, mobilize in more traditional ways, but for the most part these enactments are scattered, localized, and interpersonal. They are not yet a match for the relentless messages of mass culture. Perhaps advancements in fat acceptance will not come through anything like rights or social movements but through market forces (adapting to our changing bodies) and as more and more people come to see themselves as fat.
Kirkland
427
The techniques my interviewees described will be shared by enough people that it will be much more common to ask, after scanning a room, ‘‘Why aren’t there chairs here that will fit us all?’’ The third way that contests over the meaning of fat in antidiscrimination consciousness are channeled and deterred is through advocates’ self-descriptions using the terms set by healthism and functional individualism. Again, because of the weakness and diffusion of the advocates’ message in the context of an overwhelmingly negative cultural message about fat, their first responses must be defensive. Defensive use of the master’s tools means that more radical conceptions of what our society could be do not have a chance to be well articulated. ‘‘But isn’t it unhealthy? It’s not like being black, right? Shouldn’t you just try harder next time to lose the weight? Shouldn’t businesses be able to protect themselves from the costs of fat employees?’’ In this first line of defense, advocates are up against an edifice of professional knowledge and must cite opposing studies. The next formulation is to make analogies to already protected groups. These analogies are highly constrained because advocates know perfectly well the comparisons are not exactly on point and because the one that ties in most closelyFdisability, because it actually confronts bodily differenceFis highly medicalized and defanged, drawing them right back to the health issue. The last move is the one I want to draw attention to here. It is to object to those grounds for debate in the first place, invoking dignity or the misery of dieting. Because of the need to talk about health and abilities of fat people first, the opportunity to let the gravity of the second critique sink in is lost. After talking for a while, Vicky, Macskat, and many others proposed that the real harm of fat discrimination is the punishment of variation and deviance from the norm. They suggested that instead of ushering fat people into the category of the potentially healthy (depending on the individual), we should simply stop caring so much about the fact that fewer and fewer of us fit into the ‘‘normal’’ Body Mass Index (BMI) range of 18–24.9. Perhaps it is a deeply unsettling idea that the goal of the war on fat is a world in which everyone’s body fits into this normalized range. Perhaps we could also be more relaxed about other ‘‘others’’ such as immigrants, wearers of the hijab, people who need help going to the bathroom, conservative talk show listeners, and pit bull owners. A fully realized sense of calm pluralism could slide into a shallow relativism, to be sure. (Vicky and Macskat cannot really mean we should respect absolutely everyone, after allFpeople who love to torture puppies?) They mean much more, but also a bit less than they were able to explain. Their approach works better as an account of what the world could be like, coaxing the interlocutor along, than as a logical progression in legalistic terms (‘‘Is it a choice? If not, then is there
428
Think of the Hippopotamus
evidence of group-based discrimination? Can people with this trait still do the job? Then maybe they should be included.’’) But we have scarcely given it a chance. As difficult as it will be, we might try talking less about fat people’s health and more about our own attitudes toward difference.
References Bagenstos, Samuel R. (2004) ‘‘The Future of Disability Law,’’ 114 The Yale Law J. 1–83. Balzer, Harry (2005) NPD Reports Americans Eat Differently 20 Years Later. Report, October 6, 2005. Port Washington, NY: The NPD Group. Brown, Wendy (1995) States of Injury: Power and Freedom in Late Modernity. Princeton, NJ: Princeton Univ. Press. Brownell, Kelly D., et al. (2005) Weight Bias: Nature, Consequences, and Remedies. New York: Guilford Press. Bumiller, Kristin (1988) The Civil Rights Society: The Social Construction of Victims. Baltimore: Johns Hopkins Univ. Press. Cachelin, Fary M., et al. (2002) ‘‘Does Ethnicity Influence Body-Size Preference? A Comparison of Body Image and Body Size,’’ 10 Obesity Research 158–66. Campos, Paul F. (2004) The Obesity Myth: Why America’s Obsession with Weight Is Hazardous to Your Health. New York: Gotham Books. Cowan, Dave (2004) ‘‘Legal Consciousness: Some Observations,’’ 67 Modern Law Rev. 928–58. Crawford, Robert (1980) ‘‘Healthism and the Medicalization of Everyday Life,’’ 10 International J. of Health Services 365–88. Engel, David M., & Frank W. Munger (2003) Rights of Inclusion: Law and Identity in the Life Stories of Americans with Disabilities. Chicago: Univ. of Chicago Press. Ernsberger, Paul, & Richard J. Koletsky (1995) ‘‘Weight Cycling,’’ 273 JAMA: The J. of the American Medical Association 998–9. Ewick, Patricia, & Susan S. Silbey (1998) The Common Place of Law: Stories from Everyday Life. Chicago: Univ. of Chicago Press. Feldman, Gary, & Judith Ashton (2007) ‘‘Jumping the Gun on Weight Discrimination,’’ The Boston Globe, 2 June, p. A11. Flanagan, Sahale A. (1996) ‘‘Obesity: The Last Bastion of Prejudice,’’ 6 Obesity Surgery 430–7. Fleury-Steiner, Benjamin (2004) Jurors’ Stories of Death: How America’s Death Penalty Invests in Inequality. Ann Arbor: Univ. of Michigan Press. Freeman, Allen David (1995) ‘‘Legitimizing Racial Discrimination through Antidiscrimination Law: A Critical Review of Supreme Court Doctrine,’’ in K. Crenshaw et al., eds., Critical Race Theory: The Key Writings That Formed the Movement. New York: New Press. Gaesser, Glenn A. (2002) Big Fat Lies: The Truth About Your Weight and Your Health. Carlsbad, CA: Gu ¨ rze Books. Gard, Michael, & Jan Wright (2005) The Obesity Epidemic: Science, Morality, and Ideology. New York: Routledge. Garland-Thomson, Rosemarie (2005) ‘‘Feminist Disability Studies,’’ 30 Signs: J. of Women in Culture and Society 1557–87. Gilliom, John (2001) Overseers of the Poor: Surveillance, Resistance, and the Limits of Privacy. Chicago: Univ. of Chicago Press. Gimlin, Debra L. (2001) Body Work: Beauty and Self-Image in American Culture. Berkeley: Univ. of California Press.
Kirkland
429
Goffman, Erving (1986) Stigma: Notes on the Management of Spoiled Identity. New York: Simon & Schuster. Goode, Erich (2002) ‘‘Sexual Involvement and Social Research in a Fat Civil Rights Organization,’’ 25 Qualitative Sociology 501–34. Gould, Jon B. (2005) Speak No Evil: The Triumph of Hate Speech Regulation. Chicago: Univ. of Chicago Press. Greenhalgh, Tricia, & Simon Wessely (2004) ‘‘‘Health for Me’: A Sociocultural Analysis of Healthism in the Middle Classes,’’ 69 British Medical Bulletin 197–213. Haltom, William, & Michael McCann (2004) Distorting the Law: Politics, Media, and the Litigation Crisis. Chicago: Univ. of Chicago Press. Herman, Didi (1996) ‘‘(Il)Legitimate Minorities: The American Christian Right’s AntiGay-Rights Discourse,’’ 23 J. of Law and Society 346–63. Hull, Kathleen (2006) Same-Sex Marriage: The Cultural Politics of Love and Law. Cambridge, United Kingdom, and New York: Cambridge Univ. Press. Kemper, Karen A., et al. (1994) ‘‘Black and White Females’ Perceptions of Ideal Body Size and Social Norms,’’ 2 Obesity Research 117–26. Kirkland, Anna (2003) ‘‘Representations of Fatness and Personhood: Pro-Fat Advocacy and the Limits and Uses of Law,’’ 82 Representations 24–51. FFF (2006a) ‘‘What’s at Stake in Fatness as Disability?,’’ 26 Disability Studies Q., http:// www.dsq-sds.org/2006_winter_toc.html. FFF (2006b) ‘‘What’s at Stake in Transgender Discrimination as Sex Discrimination?,’’ 32 Signs: J. of Women in Culture and Society 83–111. FFF (2008) Fat Rights: Dilemmas of Difference and Personhood. New York: New York Univ. Press. Kolata, Gina (2002) ‘‘Asking If Obesity Is a Disease or Just a Symptom,’’ The New York Times, 16 April, p. F5. FFF (2003) ‘‘What We Don’t Know About Obesity,’’ The New York Times, 22 June, p. WK12. FFF (2004a) ‘‘Data on Deaths from Obesity Is Inflated, U.S. Agency Says,’’ The New York Times, 24 Nov., p. A16. FFF (2004b) ‘‘The Fat Epidemic: He Says It’s an Illusion,’’ The New York Times, 8 June, p. F5. FFF (2007a) ‘‘Genes Take Charge, and Diets Fall by the Wayside,’’ The New York Times, 8 May, p. F1. FFF (2007b) Rethinking Thin: The New Science of Weight Loss–and the Myths and Realities of Dieting. New York: Farrar, Straus and Giroux. Kostiner, Idit (2003) ‘‘Evaluating Legality: Toward a Cultural Approach to the Study of Law and Social Change,’’ 37 Law & Society Rev. 323–68. Lears, Jackson (2006) ‘‘The American Way of Debt,’’ The New York Times Magazine, 11 June, p. 13. LeBesco, Kathleen (2004) Revolting Bodies? The Struggle to Redefine Fat Identity. Amherst, MA: Univ. of Massachusetts Press. Luker, Kristin (1984) Abortion and the Politics of Motherhood. Berkeley: Univ. of California Press. MacKinnon, Catharine A. (1991) ‘‘Reflections on Sex Equality under Law,’’ 100 Yale Law J. 1281–1328. Maguire, Ken (2007) ‘‘Mass. Bill Would Prohibit Bias Based on Height or Weight,’’ Richmond Times-Dispatch, 18 May, p. B13. Maurer, Donna, & Jeffery Sobal (1995) Eating Agendas: Food and Nutrition as Social Problems. New York: Aldine de Gruyter. McCann, Michael (1994) Rights at Work: Pay Equity Reform and the Politics of Legal Mobilization. Chicago: Univ. of Chicago Press.
430
Think of the Hippopotamus
FFF (2006) ‘‘On Legal Rights Consciousness: A Challenging Analytical Tradition,’’ in B. Fleury-Steiner & L. B. Nielsen, eds., The New Civil Rights Research: A Constitutive Approach. Aldershot, United Kingdom, and Burlington, VT: Ashgate. Millman, Marcia (1981) Such a Pretty Face: Being Fat in America. New York: Berkley Books. National Association to Advance Fat Acceptance (2008a) ‘‘NAAFA Information,’’ http:// www.naafa.org/documents/brochures/naafa-info.html#whatis (accessed 22 Feb. 2008). FFF (2008b) ‘‘NAAFA Official Documents,’’ http://www.naafa.org/documents/official/ constitution.html (accessed 22 Feb. 2008). Nielsen, Laura Beth (2000) ‘‘Situating Legal Consciousness: Experiences and Attitudes of Ordinary Citizens About Law and Street Harassment,’’ 34 Law & Society Rev. 1055–90. Oliver, J. Eric (2006) Fat Politics: The Real Story Behind America’s Obesity Epidemic. New York: Oxford Univ. Press. Oliver, Eric, & Taeku Lee (2005) ‘‘Public Opinion and the Politics of Obesity in America,’’ 30 J. of Health Politics, Policy and Law 923–54. Polletta, Francesca (2000) ‘‘The Structural Context of Novel Rights Claims: Southern Civil Rights Organizing, 1961–1966,’’ 34 Law & Society Rev. 367–406. Post, Robert, et al. (2001) Prejudicial Appearances: The Logic of American Antidiscrimination Law. Durham, NC: Duke Univ. Press. Rose, Nikolas S. (1990) Governing the Soul: The Shaping of the Private Self. London and New York: Routledge. Saguy, Abigail C., & Rene Almeling (2008) ‘‘Fat in the Fire?: Science, the News Media, and the Obesity Crisis,’’ 23 Sociological Forum 53–83. Saguy, Abigail C., & Kevin W. Riley (2005) ‘‘Weighing Both Sides: Morality, Mortality, and Framing Contests over Obesity,’’ 30 J. of Heath Politics, Policy and Law 869–921. Sarat, Austin (1990) ‘‘‘. . .The Law Is All Over’: Power, Resistance, and the Legal Consciousness of the Welfare Poor,’’ 2 Yale J. of Law and Humanities 343–97. Shakespeare, Tom (2006) ‘‘The Social Model of Disability,’’ in L. J. Davis, ed., The Disability Studies Reader. New York: Routledge. Siebers, Tobin (2006) ‘‘Disability in Theory,’’ in L. J. Davis, ed., The Disability Studies Reader. New York: Routledge. Silbey, Susan (2005) ‘‘After Legal Consciousness,’’ 1 Annual Rev. of Law and Social Science 323–68. Sobal, Jeffery (1995) ‘‘The Medicalization and Demedicalization of Obesity,’’ in D. Maurer & J. Sobal, eds., Eating Agendas: Food and Nutrition as Social Problems. New York: Aldine de Gruyter. Sobal, Jeffery, & Donna Maurer (1999) Interpreting Weight: The Social Management of Fatness and Thinness. New York: Aldine de Gruyter. Solovay, Sondra (2000) Tipping the Scales of Justice: Fighting Weight-Based Discrimination. Amherst, NY: Prometheus Books. Stearns, Peter N. (1997) Fat History: Bodies and Beauty in the Modern West. New York: New York Univ. Press. Sturges, Judith E., & Kathleen J. Hanrahan (2004) ‘‘Comparing Telephone and Face-to-Face Qualitative Interviewing: A Research Note,’’ 4 Qualitative Research 107–18. Terlep, Sharon (2005) ‘‘Paying a Heavy Price,’’ The Detroit News, 6 Nov., p. 1A. Yoshino, Kenji (2006) Covering: The Hidden Assault on Our Civil Rights. New York: Random House.
Kirkland
431
Cases Cited Cook v. State of Rhode Island Department of Mental Health, Retardation, and Hospitals, 10 F.3d 17 (1993). Equal Employment Opportunity Commission v. Watkins Motor Lines, 463 F.3d 436 (2006).
Statutes Cited Civil Rights Act of 1964, 42 U.S.C. § 2000e (2008). Compliance Guidelines to Prohibit Weight and Height Discrimination, San Francisco Administrative Code Chapters 12A, 12B and 12C (2007). District of Columbia Human Rights Act, D.C. Code § 2-1401.01 (2007). Elliot-Larsen Civil Rights Act, Michigan Compiled Laws § 37.2202 (2007). Santa Cruz Municipal Code, § 9.83.010 (2007).
Anna Kirkland is Assistant Professor of Women’s Studies and Political Science at the University of Michigan. She holds a J.D. and Ph.D. (Jurisprudence and Social Policy) from the University of California, Berkeley. This article extends the theoretical framework of her first book, Fat Rights: Dilemmas of Difference and Personhood (NYU Press, 2008), which explores the more formal justifications available for weightbased civil rights. She has also written on transgender rights and antidiscrimination in health care. Her current projects include a multi-method survey of diversity rhetoric in college application essays and a study of obesity policy and neoliberalism.
432
Think of the Hippopotamus
433
Book Reviews Scott Barclay, Editor
The Language of Law School: Learning to ‘‘Think Like a Lawyer.’’ By Elizabeth Mertz. Oxford, United Kingdom: Oxford Univ. Press, 2007. Pp. xvii1308. $35.00 paper. Reviewed by Marianne Constable, University of California, Berkeley
Mertz’s The Language of Law School uses ‘‘close analysis of classroom language to examine the limits that legal epistemology may place on law’s democratic aspirations’’ (p. 3). Mertz points to these limits in two ways: she shows how contract law and education promote ‘‘a common vision’’ of human conflicts that obscures particular aspects of social experience, and she explores the differences that gender and race make in the teaching and learning of law. Mertz’s painstaking research is a model empirical and sociolegal study of language. The first three chapters quite thoroughly survey related literatures and very clearly describe the research method. Mertz and her assistants observed, taped, coded, and transcribed eight first-year, semester-long contracts courses from a range of law schools and supplemented their material with interviews. These preliminary chapters explain factorsFsuch as turn-taking, repeated speech, pronomial usage, framing and footing, and role-playingFand termsFsuch as ‘‘pragmatics,’’ ‘‘Socratic method teaching,’’ and ‘‘metalinguistic filter’’Fused in the analyses. They introduce the issues through a compelling scenario beginning as follows: ‘‘Picture yourself entering a law school classroom on the first day of law school’’ (p. 7). The book’s greatest contribution comes in its middle chapters, which offer, in a smart and sophisticated reading of the classroom transcripts, subtle analyses of what legal discourse does and how. In these chapters, Mertz considers the implications of the commonalities in the dynamics of the eight classrooms. Although the eight classrooms differ in their use of lecture, free discussion, and shortfocused exchange, professors in all the classrooms, through their speech, model particular relations to texts, to parties involved in lawsuits, and to classroom interactions. A focus on dialogue even when presented in lectures, for instance, suggests the importance of the duelist mode in law. Students ultimately become strategists of speech. They learn to present themselves and the parties whom they animate in response to questions about cases in particular Law & Society Review, Volume 42, Number 2 (2008) r 2008 by The Law and Society Association. All rights reserved.
434
Book Reviews
ways. They learn to privilege texts not for their stories of human conflicts but for their answers to ‘‘a series of nested questions about the authority of various courts deciding the case at issue and also of the courts that authored precedents’’ (p. 62). They learn to treat legal texts as ‘‘detachable chunk[s] of discourse’’ (p. 45) that can be moved from one context to another. Vast differences in the cultural meanings of particular kinds of actions or items become elided into a common legal language (p. 64) that reworks temporality and history (p. 63). Persons in conflicts become types; often, students come to know ‘‘parties’’ through occupational status and worldly belongings, referring to them as ‘‘buyers’’ and ‘‘sellers’’ into whose mouths they put strategic language whose looseness contrasts to the precision demanded when quoting legal authorities. ‘‘Policy’’ becomes a catch-all phrase for matters unaddressed in written text. As students ostensibly prepare for legal practice, they engage in a landscape of argumentative positions, discourse frames, and participant roles, and come to inhabit an ‘‘I’’ that is not ‘‘their own self ’’ (p. 135). Toward the end of the book, Mertz places her study in the context of other studies concerned with classroom inclusion and exclusion by gender and race. Although Mertz finds that ‘‘white male students’’ who are ‘‘traditional insiders in the legal profession . . . tend to predominate,’’ she also finds ‘‘interesting fluctuations in the patterning’’ (p. 202). Classrooms may be inclusive along lines of gender but not race and vice versa; the rate and manner of student participation does not necessarily correlate to the race and gender of the professor. Here her results are not determinative and she calls for more research. From her perspective as a social scientist, Mertz shows how law’s erasure of certain aspects of social experience comes about through classroom dynamics that affirm the irrelevance of particular characteristics to legal equality. Unfortunately, according to Mertz, law has no mechanism for challenging its own orientation to what it takes for granted (p. 219). It thus falls to social science to draw attention to what she calls ‘‘the double edge’’ of law (pp. 63, 221): social science points not only to the invisibility of particular social factors in law but also to law’s neglect of social experiences that may be relevant to ‘‘democracy’’ after all. In characterizing the limitations of law in terms of social factors bearing on democracy, the book reveals its political and epistemological orientations to be those common to sociolegal study today. It privileges a social and empirical language for talking about issues of justice and power, and it attributes historically contingent ‘‘democratic aspirations’’ to law. However warranted as a sociolinguistic critique of contemporary legal education, the book leaves one wondering whether social science indeed has the
Book Reviews
435
mechanisms for or inclination to challenge its own orientation to what it takes for granted. n
n
n
Everyday Harm: Domestic Violence, Court Rites, and Cultures of Reconciliation. By Mindie Lazarus-Black. Urbana: Univ. of Illinois Press, 2007. Pp. xii1244. $22.00 paper. Reviewed by Keith Guzik, Bloomfield College
Everyday Harm: Domestic Violence, Court Rites, and Cultures of Reconciliation explores the impact of Trinidad and Tobago’s 1991 Domestic Violence Act (p. 23), the first legislation in the Englishspeaking Caribbean giving domestic violence victims the right to petition courts for orders of protection against their abusers. The subject matter, if seemingly esoteric, is vital. Relatively little sociolegal research has studied the globalization of domestic violence law, one of the most striking legal phenomena of recent decades. Conceptualizing law as power-laden events and processes, author Lazarus-Black aims to answer four key questions in this work: (1) Why and when do lawmakers create domestic violence law? (2) Why does such legislation usually produce few substantive outcomes for victims? (3) What does domestic violence law mean for women’s empowerment? (4) How does culture influence the law? Lazarus-Black investigates these questions through an ambitious research design combining quantitative and qualitative methods. She collected the records of all 1,463 protection order hearings that occurred in a magistrate’s court in ‘‘Pelau’’ over a two-year period (from January 1997 to December 1998). Over the course of three field visits to this town of some 15,000 residents, she conducted more than 100 interviews with legal professionals, litigants, and other community members regarding their views of the domestic violence law and experiences with protection order cases. On the basis of this rich data, Everyday Harm accomplishes most of what it sets out to do. Chapter 1 considers the postcolonial history of Trinidad and Tobago to explain how the country’s adoption of domestic violence legislation depended on different historical factors: a national embrace of modernist ideology; the expansion of public education, especially for women; relative economic prosperity that allowed residents to travel internationally and access global media; and feminist political activism. In Chapter 2, Lazarus-Black, together with Patricia McCall, provides a quantitative analysis of protection order applications in Pelau and finds much of what past research
Book Reviews
435
mechanisms for or inclination to challenge its own orientation to what it takes for granted. n
n
n
Everyday Harm: Domestic Violence, Court Rites, and Cultures of Reconciliation. By Mindie Lazarus-Black. Urbana: Univ. of Illinois Press, 2007. Pp. xii1244. $22.00 paper. Reviewed by Keith Guzik, Bloomfield College
Everyday Harm: Domestic Violence, Court Rites, and Cultures of Reconciliation explores the impact of Trinidad and Tobago’s 1991 Domestic Violence Act (p. 23), the first legislation in the Englishspeaking Caribbean giving domestic violence victims the right to petition courts for orders of protection against their abusers. The subject matter, if seemingly esoteric, is vital. Relatively little sociolegal research has studied the globalization of domestic violence law, one of the most striking legal phenomena of recent decades. Conceptualizing law as power-laden events and processes, author Lazarus-Black aims to answer four key questions in this work: (1) Why and when do lawmakers create domestic violence law? (2) Why does such legislation usually produce few substantive outcomes for victims? (3) What does domestic violence law mean for women’s empowerment? (4) How does culture influence the law? Lazarus-Black investigates these questions through an ambitious research design combining quantitative and qualitative methods. She collected the records of all 1,463 protection order hearings that occurred in a magistrate’s court in ‘‘Pelau’’ over a two-year period (from January 1997 to December 1998). Over the course of three field visits to this town of some 15,000 residents, she conducted more than 100 interviews with legal professionals, litigants, and other community members regarding their views of the domestic violence law and experiences with protection order cases. On the basis of this rich data, Everyday Harm accomplishes most of what it sets out to do. Chapter 1 considers the postcolonial history of Trinidad and Tobago to explain how the country’s adoption of domestic violence legislation depended on different historical factors: a national embrace of modernist ideology; the expansion of public education, especially for women; relative economic prosperity that allowed residents to travel internationally and access global media; and feminist political activism. In Chapter 2, Lazarus-Black, together with Patricia McCall, provides a quantitative analysis of protection order applications in Pelau and finds much of what past research
436
Book Reviews
on this topic has discovered. Courts see a high number of applications from women lodging complaints against men (more than 80 percent of all applications). Most of these (more than 77 percent) are dismissed or withdrawn, resulting in few actual orders being issued (fewer than 4 percent). Interestingly, however, and evidencing the way in which domestic violence law is localized, 19 percent of cases are adjudicated through ‘‘undertakings,’’ a unique provision of the Domestic Violence Act that ‘‘allows a person accused of domestic violence for the first time to sign an affidavit promising not to engage in the activities for which he or she is being charged’’ (p. 32). The remaining chapters of the book explain why so few victims secure protection orders. As Lazarus-Black convincingly argues, the answer lies in ‘‘court rites’’ or institutionalized legal practices. These practices include the creation of an intimidating environment at court through the enforcement of its rules, which preserve class and gender hierarchies; the ‘‘time’’ required to process cases; and the ‘‘cultures of reconciliation’’ in Trinidad and Tobago that pressure parties to settle their disputes outside of court. While each of these forces is familiar from past law-and-society research, Everyday Harm makes a key contribution by further elaborating and systematizing them. For instance, the failure of the law to provide victims formal protection stems not only from the excessive time a successful petition requires, but also from the fact that the different actors in the processFapplicant, courthouse, police, respondent, lawyer, probation officer, and magistrateFall possess their own time, which they use differently. The only question left unanswered is what the Domestic Violence Act has meant for women’s empowerment. Absent is any systematic consideration of how women assess their experiences with the law. And while Lazarus-Black interprets the gap between protection order applications and actual orders granted as proof of the law’s failure and the permanence of gender subordination, past domestic violence research (which she cites) and cases presented in the book indicate that women themselves may interpret such outcomes differently. Some women might obtain relief from their abusive relationships simply by petitioning the court (the abuser gets the message and either changes the cited behavior or moves out), thus making their attendance at future hearings on the case unnecessary in their minds. In addition, the relatively high number of undertakings suggests a unique reworking of the gender order in Trinidad and Tobago, whereby abusive men are made to (willingly) submit themselves to the authority of the court to monitor their behavior. Women’s empowerment and structural change, then, could actually accompany the Domestic Violence Act’s failure to provide protection orders. But without the views of
Book Reviews
437
the women who experienced these legal outcomes, it is difficult to determine how the law has affected them. This point notwithstanding, Everyday Harm is a conceptually innovative and rigorously researched book that represents a genuine contribution to sociolegal research on the power of law and domestic violence research in general. n
n
n
Words for the Taking: The Hunt for a Plagiarist. By Neal Bowers. Carbondale: Southern Illinois Univ. Press, [1997] 2007. Pp. xvii1152. $14.95 paper. Reviewed by Susan Burgess, Ohio University
Neal Bowers, a well-published poet and tenured university professor, is living the good life in Ames, Iowa. Or at least he is, until he learns that several of his poems have been plagiarized, meaning stolen and published nearly word for word under a name other than his own. This startling news rocks his life, setting him off on a search for the perpetrator that fundamentally alters his orientation to his life’s work. For the most part, this is a book of enormous disappointment and disillusionment. How could someone steal his work so brazenly? He is also disappointed in his friends and colleagues. Why do they not understand the enormity of this violation? Why are they not more sympathetic to his plight? Many of them cannot understand why he continues to be so concerned with the case, particularly once the plagiarist has been identified. Bowers cannot understand why they are not. In addition, Bowers is disappointed with the law. Why is it so ineffectual? After discussing his case with several uninterested attorneys, he hires a local lawyer and a private detective who never quite manage to live up to Bowers’s expectations despite succeeding in tracking down and extracting a letter of apology from the plagiarist, David Jones, complete with a token check for $100 in compensation for trouble caused. Jones appears ‘‘judgmentproof,’’ having earned a grand total of $650 in the previous calendar year. For this, Bowers accrues more than $4,000 in attorney’s fees (some of which he contests as unfair charges). Rather than experiencing the expected exhilaration at having finally caught up with the perpetrator, what he finds is a troublingly manipulative and rather pathetic compulsive liar who cannot seem to hold down a job teaching elementary school. When he is unsatisfied with this result, his attorney asks him what he
Book Reviews
437
the women who experienced these legal outcomes, it is difficult to determine how the law has affected them. This point notwithstanding, Everyday Harm is a conceptually innovative and rigorously researched book that represents a genuine contribution to sociolegal research on the power of law and domestic violence research in general. n
n
n
Words for the Taking: The Hunt for a Plagiarist. By Neal Bowers. Carbondale: Southern Illinois Univ. Press, [1997] 2007. Pp. xvii1152. $14.95 paper. Reviewed by Susan Burgess, Ohio University
Neal Bowers, a well-published poet and tenured university professor, is living the good life in Ames, Iowa. Or at least he is, until he learns that several of his poems have been plagiarized, meaning stolen and published nearly word for word under a name other than his own. This startling news rocks his life, setting him off on a search for the perpetrator that fundamentally alters his orientation to his life’s work. For the most part, this is a book of enormous disappointment and disillusionment. How could someone steal his work so brazenly? He is also disappointed in his friends and colleagues. Why do they not understand the enormity of this violation? Why are they not more sympathetic to his plight? Many of them cannot understand why he continues to be so concerned with the case, particularly once the plagiarist has been identified. Bowers cannot understand why they are not. In addition, Bowers is disappointed with the law. Why is it so ineffectual? After discussing his case with several uninterested attorneys, he hires a local lawyer and a private detective who never quite manage to live up to Bowers’s expectations despite succeeding in tracking down and extracting a letter of apology from the plagiarist, David Jones, complete with a token check for $100 in compensation for trouble caused. Jones appears ‘‘judgmentproof,’’ having earned a grand total of $650 in the previous calendar year. For this, Bowers accrues more than $4,000 in attorney’s fees (some of which he contests as unfair charges). Rather than experiencing the expected exhilaration at having finally caught up with the perpetrator, what he finds is a troublingly manipulative and rather pathetic compulsive liar who cannot seem to hold down a job teaching elementary school. When he is unsatisfied with this result, his attorney asks him what he
438
Book Reviews
wants from the law and Bowers somewhat sheepishly notes that what he desires is justice, or ‘‘mak[ing] him give back what he took from me’’ (p. 75). The damage already having been done, Bowers realizes that this is impossible. The confessional nature of Bowers’s poems seems to exacerbate his feelings of violation, as the stolen poems focus on unresolved feelings surrounding the death of Bowers’s father. In a 1998 interview, Bowers commented that the poems ‘‘have the stench of David Jones about them. . . . In an unexpected and bizarre way, Jones has invaded not just my poetry but also my private past. Almost every thought of my father evokes thoughts of Jones. That’s the thing for which I can never forgive Jones’’ (The Cortland Review 1998: n.p.). This perhaps explains why he continues to pursue more information about Jones. How many poems has he plagiarized? How many other poets have been victimized? Could he compel a more direct statement of guilt from him? The pursuit of Jones continues until, near the end of the book, Bowers discloses the startling fact that Jones is a convicted child molester. This book reveals an interesting paradox about plagiarism: continuing to discuss it publicly seems to be both the best defense against it at the same time that it appears to keep Bowers’s wound open (not to mention the enormous amount of attention it seems to provide to the plagiarist.) As a result of this book, originally published by W. W. Norton in 1997, Bowers becomes a national expert on plagiarism, called on for comments from a variety of news sources including The New York Times. (One imagines that its reprinting by Southern Illinois University Press will keep the issue alive longer still.) At one point, noted playwright Tony Kushner calls Bowers while reading his book during an intercontinental flight, enthusiastically talking about optioning it for a Hollywood movie. Though he never hears back from Kushner, Bowers does sell the film rights to a production company (and, perhaps predictably, is disappointed in the resulting script). Thus in an ironic twist the book ends with Bowers appearing to profit from the work of Jones. Though this book raises many important issues about plagiarism, the specifics of the case related to both the victim and the perpetrator detract from rather than enhance that discussion. I had a good amount of sympathy for the author, who was an accomplished poet and professor at the time of plagiarism but who actually stopped writing poems thereafter and turned instead to writing this book and novels. Thus the plagiarism changed the course of his life’s work. However, I never really felt that I understood what the author was trying to accomplish in this book beyond eliciting that sympathy. Although that may well be a worthy
Book Reviews
439
story, it is a familiar one, and not the more creative narrative that I hoped to encounter in this book. This is perhaps my disappointment, then, and it stems from the fact that I was quite intrigued by the author’s promise of using mystery (complete with crimes, detectives, and sensational and noir-ish revelations) as a means of getting a different purchase on a serious issue such as plagiarism. I was hoping that using a different form would allow the author to tell a different story or to alter the content of what might be said about plagiarism. I was disappointed not to find more evidence of that in this book. Reference The Cortland Review (1998). ‘‘Interview,’’ Neal Bowers, http://www.cortlandreview.com/ issuefive/interview5.htm.
n
n
n
Answering the Call of the Court: How Justices and Litigants Set the Supreme Court Agenda. By Vanessa A. Baird. Charlottesville and London: Univ. of Virginia Press, 2007. Pp. xii1225. $45.00 cloth. Reviewed by Udi Sommer, University at Albany, SUNY
In Answering the Call of the Court: How Justices and Litigants Set the Supreme Court Agenda, Baird attempts to explain the relationship between justices and litigants early in the agenda-setting process. Baird implements a remarkable research design that yields a wellwritten volume. The primary theoretical contribution of this project is in the notion that in order to create social change, the Court depends on the support structure of extrajudicial players. Legal entrepreneurs pay heed to signals concerning justices’ priorities and then sponsor cases in the appropriate issue areas. The cases sponsored present the type of legal questions and case facts that make them good vehicles for policy change. In other words, litigants have a pivotal role in translating shifts in the priorities of justices into actual changes in the agenda. The upshot of this symbiotic relationship is increased policymaking power to both justices and legal entrepreneurs. Agenda-setting, according to Baird, begins earlier than anybody else has ever argued. To a significant degree, it starts with politically salient decisions handed down by the Court approximately four years before the decision to grant certiorari in the
Book Reviews
439
story, it is a familiar one, and not the more creative narrative that I hoped to encounter in this book. This is perhaps my disappointment, then, and it stems from the fact that I was quite intrigued by the author’s promise of using mystery (complete with crimes, detectives, and sensational and noir-ish revelations) as a means of getting a different purchase on a serious issue such as plagiarism. I was hoping that using a different form would allow the author to tell a different story or to alter the content of what might be said about plagiarism. I was disappointed not to find more evidence of that in this book. Reference The Cortland Review (1998). ‘‘Interview,’’ Neal Bowers, http://www.cortlandreview.com/ issuefive/interview5.htm.
n
n
n
Answering the Call of the Court: How Justices and Litigants Set the Supreme Court Agenda. By Vanessa A. Baird. Charlottesville and London: Univ. of Virginia Press, 2007. Pp. xii1225. $45.00 cloth. Reviewed by Udi Sommer, University at Albany, SUNY
In Answering the Call of the Court: How Justices and Litigants Set the Supreme Court Agenda, Baird attempts to explain the relationship between justices and litigants early in the agenda-setting process. Baird implements a remarkable research design that yields a wellwritten volume. The primary theoretical contribution of this project is in the notion that in order to create social change, the Court depends on the support structure of extrajudicial players. Legal entrepreneurs pay heed to signals concerning justices’ priorities and then sponsor cases in the appropriate issue areas. The cases sponsored present the type of legal questions and case facts that make them good vehicles for policy change. In other words, litigants have a pivotal role in translating shifts in the priorities of justices into actual changes in the agenda. The upshot of this symbiotic relationship is increased policymaking power to both justices and legal entrepreneurs. Agenda-setting, according to Baird, begins earlier than anybody else has ever argued. To a significant degree, it starts with politically salient decisions handed down by the Court approximately four years before the decision to grant certiorari in the
440
Book Reviews
current case. The pattern of the Court’s changing interests is clearFwhen Baird employs different measures for justices’ priorities, the result is that four to five years after the Court signals its interest in a certain issue area, the number of cases in this area on the Court’s agenda swells significantly. Quantitative and qualitative analyses are sophisticatedly combined in this volume. Despite ‘‘black-boxing’’ some of the processes in her statistical analyses, Baird carefully outlines what happens in the box in the qualitative sections of the book. For instance, using various illustrative examples from Supreme Court cases, in Chapter 2 the author discusses the signals the Court sends to litigants. These include separate opinions registered by justices (e.g., the separate opinions in Furman v. Georgia [1972]) and unexpected actions (e.g., Justice Anthony Kennedy’s decision to join the majority in Lawrence v. Texas [2003]). Baird also offers an in-depth analysis of how the Mexican American Legal Defense Fund (MALDEF) took advantage of legal cues in recent Supreme Court decisions to devise a legal strategy challenging the Personal Responsibility Act passed by Congress in 1996. In this case (Latino Coalition for a Healthy California v. Belshe [1997]), based on former Justice Sandra Day O’Connor’s approach to federalism in New York v. United States (1992), the MALDEF chose federalism over equal protection as grounds for litigation. In the quantitative portion of the book, Baird considers statistically the Court’s attention to policy areas over time, which is operationalized as the number of cases across eleven policy areas from 1953 to 2000. The explanatory variable, the Court’s priorities, is measured as an index comprising several indicators: the presence of the decision on the front page of The New York Times, the formal alteration of a precedent, and reversal. The resultant model demonstrates that consideration of an important case by the Court adds four additional cases to the Court’s agenda after four years and then again after five years (i.e., a total of eight additional cases). To elegantly refute the hypothesis that this lag is really a product of the justices’ strategic concern for legitimacy, Baird demonstrates that although justices are apparently willing to take cases earlier, such cases are just not available. It actually takes several years for good vehicles to percolate. And, as we would expect given this model, the priorities of the Court influence the nondiscretionary docket of the courts of appeals a year before they influence the agenda of the Supreme Court itself. The argument in this book has important implications that go beyond the role of extrajudicial players and pertain to the role of the Court within the American constitutional design. Legal reality, in Baird’s depiction, resembles a pure market model more than previously thought. Policy areas, which are of interest for the Court, are often niches where supply fails to meet demand. Legal
Book Reviews
441
entrepreneurs who look for profit in policy terms are quick to react. Based on signals from the Court, they identify the potential for policy gains and sponsor appropriate litigation. This puts the Court in a more proactive position than traditionally thought. Moreover, within a matter of four years the Court has available to it the right vehicles. Even compared to other branches of government, a period of four years is not a lot. Terms in office of elected officials are on average that long. This is the time officials have to influence policy, and once their term is over there is the potential for a substantial policy change. Clearly, the Court is fundamentally different from other branches of government. However, when it comes to its ability to play a proactive role in policymaking, Baird’s argument indicates that the Court is a rather potent player. Cases Cited Furman v. Georgia, 408 U. S. 238 (1972). Latino Coalition for a Healthy California v. Belshe, 785153-7, Calif. Sup. Ct., December 19, 1997. Lawrence v. Texas, 539 U. S. 558 (2003). New York v. United States, 505 U. S. 144 (1992).
n
n
n
Law as a Means to an End: Threat to the Rule of Law. By Brian Z. Tamanaha. New York: Cambridge Univ. Press, 2006. Pp. xii1254. $80.00 cloth; $31.99 paper. Reviewed by Roger Cotterrell, Queen Mary College, University of London
Tamanaha’s previous books have shown his ability to present vivid arguments on large themes of great contemporary interest. He engages provocatively with key debates; typically develops his arguments in clear, direct prose; and usually reaches strong conclusions that challenge the reader. His newest book shows all these characteristics and is also written with much passion, because its theme is nothing less than the health of, orFas he sees itFthe sickness of the U.S. legal system as a whole. He argues that a pernicious instrumentalism has taken over virtually all institutions of American lawFespecially the legislative and administrative processes, the Supreme Court, and much of lawyers’ practice, legal education, jurisprudence, and sociolegal scholarship. If law was always seen instrumentally to some extent, what Tamanaha thinks is new (roughly since the beginning of the twentieth century) is
Book Reviews
441
entrepreneurs who look for profit in policy terms are quick to react. Based on signals from the Court, they identify the potential for policy gains and sponsor appropriate litigation. This puts the Court in a more proactive position than traditionally thought. Moreover, within a matter of four years the Court has available to it the right vehicles. Even compared to other branches of government, a period of four years is not a lot. Terms in office of elected officials are on average that long. This is the time officials have to influence policy, and once their term is over there is the potential for a substantial policy change. Clearly, the Court is fundamentally different from other branches of government. However, when it comes to its ability to play a proactive role in policymaking, Baird’s argument indicates that the Court is a rather potent player. Cases Cited Furman v. Georgia, 408 U. S. 238 (1972). Latino Coalition for a Healthy California v. Belshe, 785153-7, Calif. Sup. Ct., December 19, 1997. Lawrence v. Texas, 539 U. S. 558 (2003). New York v. United States, 505 U. S. 144 (1992).
n
n
n
Law as a Means to an End: Threat to the Rule of Law. By Brian Z. Tamanaha. New York: Cambridge Univ. Press, 2006. Pp. xii1254. $80.00 cloth; $31.99 paper. Reviewed by Roger Cotterrell, Queen Mary College, University of London
Tamanaha’s previous books have shown his ability to present vivid arguments on large themes of great contemporary interest. He engages provocatively with key debates; typically develops his arguments in clear, direct prose; and usually reaches strong conclusions that challenge the reader. His newest book shows all these characteristics and is also written with much passion, because its theme is nothing less than the health of, orFas he sees itFthe sickness of the U.S. legal system as a whole. He argues that a pernicious instrumentalism has taken over virtually all institutions of American lawFespecially the legislative and administrative processes, the Supreme Court, and much of lawyers’ practice, legal education, jurisprudence, and sociolegal scholarship. If law was always seen instrumentally to some extent, what Tamanaha thinks is new (roughly since the beginning of the twentieth century) is
442
Book Reviews
that instrumental views have been established in a ‘‘specific historical contrast’’ (p. 35) with earlier noninstrumental views, now entirely displaced. Law tends now to be seen only as a mere technical means to achieve any chosen ends. Noninstrumentalism viewed law as having ‘‘an inviolable, built-in principled integrity’’ (p. 219); its special identity, making it more than just a tool, was ‘‘as a matter of principle, reason, immemorial customs of the community, a body of specialized knowledge and a science’’ (p. 58). On this noninstrumental view, law adjusts to social change but is not a tool of social engineering; judges should apply law ‘‘with no preconceived controlling end in view,’’ and legislators must ‘‘seek to declare the immanent norms of the community or natural principles’’ (p. 7). In Tamanaha’s argument, instrumental views of law cannot now be displaced and would not be dangerous but for the deterioration of belief in what he variously calls the common good, the public good, general welfare, or public purpose. Without this, nothing holds legal instrumentalism in check. With ‘‘rampant instrumental manipulation of the law’’ (p. 250), lawyers have little concern beyond furthering their clients’ interests (and so their own) by any means short of illegality; the selection of judges depends on whether their personal preferences will lead them to adjudicate consistently with the interests of those who decide their appointment; legislators tailor their votes on legislation to the demands of those who can influence their re-election; law students are taught that skill in arguing legally on either side of a case is more important than working out how it should be resolved. Tamanaha’s discussion of professional legal practice (especially corporate and tax), and its often ‘‘brutish conditions’’ (p. 136)F relentless competition, the billable hours system, the high costs of entering the profession, and financial pressures on partners to bring in clientsFis the angriest in the book, closely followed by his condemnations of legislative lobbying, judicial appointment methods, and the engineering of litigation through cause lawyering. By the end of the book, a truly ‘‘barren vision’’ has been conjured up, of ‘‘a war of all against all within and through law’’ (p. 225). Ultimately Tamanaha is not sure whether things are quite this bad. Judges generally have not yet been reduced to deciding instrumentally, though he sees much evidence that the Supreme Court has, and its influence threatens to popularize the idea of judicial instrumentalism. Everything depends on whether some remnant of the common good idea survives, and Tamanaha is not sure about that, one way or the other. This (non-American) reviewer cannot assess the book’s accounts of such a vast range of American legal experience. An important issue, however, is whether the conceptual framework Tamanaha uses can hold together this diversity of description and critique. Ostensibly the dominant concept is legal instrumentalism, but since he accepts that
Book Reviews
443
this, as such, does not indicate a new phenomenon, he makes two moves to sharpen the focus. The first is to see instrumentalism as newly significant as a displacement of noninstrumental views. But he recognizes that special interests were often served by the nineteenthcentury decisions of courts proclaiming the language of noninstrumentalism, and that now, as then, arguments that the pursuit of private claims is the essence of law can easily be made or assumed. The issue seems to come down to how far invocations of noninstrumental ideas actually determine outcomes, but that remains unclear. A second move is to claim that the sense of common good has declined, perhaps to the vanishing point, so it must have ceased to inform law. This deserves much discussion but seems ultimately beyond the scope of the book. Sociologists have told us much about social capital, ‘‘habits of the heart,’’ and the nature of contemporary values and beliefs. Without a careful study of such matters, as reflected in legal ideas and practices and in citizen demands on law, the book remains a broad, many-sided general polemic about important areas of legal disorder and dissatisfactionFfascinating, empirically rich and strikingly presented, but unified perhaps only by the author’s conviction that selfishness has at last overtaken law. n
n
n
Injury: The Politics of Product Design and Safety Law in the United States. By Sarah S. Lochlann Jain. Princeton, NJ: Princeton Univ. Press, 2006. Pp. xii1214. $55.00 cloth; $19.95 paper. Reviewed by Stephen Daniels, American Bar Foundation
Injury offers a challenging and provocative discussion of issues that are the subject of intense debate: tort law and product-caused injuries. Jain, a cultural anthropologist, challenges the reader to look at these familiar issues in a different way, ‘‘to step outside of the questions of frivolous cases and junk science’’ (p. 4). Instead, she encourages us to think more deeply about: the centrality and necessity of injury in the American economy; how injury and inequality are intertwined; and the promise, limits, and failures of law in dealing with injury in a way adequately recognizing the goal of human well-being. This goal, she says, ‘‘must have some rhetorical, if not material, purchase in any social economy that wants to pass as democratic’’ (p. 33). Jain’s introduction (‘‘Injury in U.S. Risk Culture’’), first chapter (‘‘American Injury Culture’’), and conclusion lay out the challenge in presenting her theoretical argument. At its heart is the idea of
Book Reviews
443
this, as such, does not indicate a new phenomenon, he makes two moves to sharpen the focus. The first is to see instrumentalism as newly significant as a displacement of noninstrumental views. But he recognizes that special interests were often served by the nineteenthcentury decisions of courts proclaiming the language of noninstrumentalism, and that now, as then, arguments that the pursuit of private claims is the essence of law can easily be made or assumed. The issue seems to come down to how far invocations of noninstrumental ideas actually determine outcomes, but that remains unclear. A second move is to claim that the sense of common good has declined, perhaps to the vanishing point, so it must have ceased to inform law. This deserves much discussion but seems ultimately beyond the scope of the book. Sociologists have told us much about social capital, ‘‘habits of the heart,’’ and the nature of contemporary values and beliefs. Without a careful study of such matters, as reflected in legal ideas and practices and in citizen demands on law, the book remains a broad, many-sided general polemic about important areas of legal disorder and dissatisfactionFfascinating, empirically rich and strikingly presented, but unified perhaps only by the author’s conviction that selfishness has at last overtaken law. n
n
n
Injury: The Politics of Product Design and Safety Law in the United States. By Sarah S. Lochlann Jain. Princeton, NJ: Princeton Univ. Press, 2006. Pp. xii1214. $55.00 cloth; $19.95 paper. Reviewed by Stephen Daniels, American Bar Foundation
Injury offers a challenging and provocative discussion of issues that are the subject of intense debate: tort law and product-caused injuries. Jain, a cultural anthropologist, challenges the reader to look at these familiar issues in a different way, ‘‘to step outside of the questions of frivolous cases and junk science’’ (p. 4). Instead, she encourages us to think more deeply about: the centrality and necessity of injury in the American economy; how injury and inequality are intertwined; and the promise, limits, and failures of law in dealing with injury in a way adequately recognizing the goal of human well-being. This goal, she says, ‘‘must have some rhetorical, if not material, purchase in any social economy that wants to pass as democratic’’ (p. 33). Jain’s introduction (‘‘Injury in U.S. Risk Culture’’), first chapter (‘‘American Injury Culture’’), and conclusion lay out the challenge in presenting her theoretical argument. At its heart is the idea of
444
Book Reviews
‘‘injury’’ in America: what counts as injury, what is recognized as a cause, and how responsibility is distributed. Hers is a broader view that starts by looking at injury as the wounding of the body in the vernacular rather than the legal sense. More, product-caused injury is an inevitable, central feature of American consumer-capitalism. It is so inevitable that its cost, discounted by its likelihood, is built into the design of products. To drive home the point of centrality, Jain even argues that whole areas of business and profit depend on there being a substantial amount of injury. In her view, these are crucial facts left out of most debates over injury lawFfacts that point to more fundamental issues that must be addressed. Focusing on injury offers a way to critique not only tort law, but also the system of American consumer-capitalism of which it is a part. Important for Jain, injury law offers one powerful discourse on what counts as an injury, what causes injury, and how the responsibility for injury is distributed. However, law fails in her estimation, because it does not fully comprehend the actuality of injury as it is suffered and its unequal distribution. Rather than seeing injury as a central, inevitable feature of the American economic system, the lawFwith its case-by-case approachFsees injury as an aberrant event. Consequently, ‘‘the law is ill-equipped to handle its [injury’s] crucial socioeconomic challenges . . . [injury laws] narrow our modes of apprehension of what counts as injury, they divert attention away from other ways of understanding injury, and they miss the cultural implications of objects and the ways that objects are situated in networks of power’’ (p. 151). Jain wants to redirect our attention to more fundamental issues such as human well-being and equality. Among other matters, Jain wants to focus our attention on product design in the American economy. Design incorporates key assumptions about users and their bodies. Often invisible to injury law, those assumptions may reflect the inequalities and biases endemic to the American system. Sometimes, as with the susceptibility of women to injury from airbags in cars, this may not be a conscious part of a design. Other times, as with mentholated cigarettes and the targeting of African Americans by tobacco companies, it may be all too intentional. Regardless, these designed-in assumptions have a great deal to do with who may suffer an injury and who may escape injury. As a result, the occurrence of injury may also reflect those inequalities and biasesFand remain invisible to injury law. While some scholars and reformers see injury law as a way of attacking inequality and fostering social justice (key goals for Jain), in Jain’s view any victoriesFto the extent they occurFare Pyrrhic victories. The middle three chapters of Injury are there to demonstrate that injury law is inadequate to this larger task. Each presents a story, ‘‘a genealogy of how particular injuries and objects have come to be understood at particular moments’’ (p. 8). The stories tell of the
Book Reviews
445
short-handled hoe and the plight of Mexican American farm workers in California; the QWERTY keyboard, repetitive strain injury, and the position of female office workers; and mentholated cigarettes and the targeting of African Americans by tobacco companies. Each of these three chapters details not only the litigation surrounding these objects, but also the subordinate positions of the main victims of the injuries involved. Each also emphasizes the limitations and outright failures of injury law in the face of the victims’ positions. At best, Jain says in her conclusion, these three stories argue that ‘‘[t]ort law offers a tinkering mechanism’’ (p. 149). Her concluding remarks about the litigation that successfully led to the short-handled hoe being declared unsafe for farm workers in California are illustrative. Despite the immediate victory, it ‘‘necessarily circumscribed workers’ ills within a narrow set of legal issues and away from the conditions of agricultural labor more generally’’ (p. 85). Jain offers no immediate solutionsFthis is not her purpose. With the provocative use of real-world examples, Injury is a firstrate work of critique. It should be on the reading list of anyone interested in the civil justice system and the political debates surrounding itFregardless of their position on the issues. n
n
n
Bodies in Revolt: Gender, Disability, and a Workplace Ethic of Care. By Ruth O’Brien. New York: Routledge, 2005. Pp. 198. $135.00 cloth; $36.95 paper. Reviewed by Daniel Santore, University at Albany, SUNY
What are the conditions under which workers’ rights legislation can produce revolutionary change in the workplace? O’Brien proposes an answer to this question by marshaling an array of Western theories of bodies and human action, and considering the ‘‘radical potential’’ of the Americans with Disabilities Act (ADA). According to O’Brien, the ADA’s open-ended definition of worker need and its (as yet unrealized) affinity with humanist values make it a potential agent of change in the logic of the capitalist workplace. The ADA can benefit all workers, not simply discrete factions of workers or those workers fitting a narrow definition of disability, by making employers and courts accommodate individuals’ varied modes of activity at work. O’Brien asserts that the ADA offers workplace accommodations on the basis of individual, and thus endlessly varying, needs. This
Book Reviews
445
short-handled hoe and the plight of Mexican American farm workers in California; the QWERTY keyboard, repetitive strain injury, and the position of female office workers; and mentholated cigarettes and the targeting of African Americans by tobacco companies. Each of these three chapters details not only the litigation surrounding these objects, but also the subordinate positions of the main victims of the injuries involved. Each also emphasizes the limitations and outright failures of injury law in the face of the victims’ positions. At best, Jain says in her conclusion, these three stories argue that ‘‘[t]ort law offers a tinkering mechanism’’ (p. 149). Her concluding remarks about the litigation that successfully led to the short-handled hoe being declared unsafe for farm workers in California are illustrative. Despite the immediate victory, it ‘‘necessarily circumscribed workers’ ills within a narrow set of legal issues and away from the conditions of agricultural labor more generally’’ (p. 85). Jain offers no immediate solutionsFthis is not her purpose. With the provocative use of real-world examples, Injury is a firstrate work of critique. It should be on the reading list of anyone interested in the civil justice system and the political debates surrounding itFregardless of their position on the issues. n
n
n
Bodies in Revolt: Gender, Disability, and a Workplace Ethic of Care. By Ruth O’Brien. New York: Routledge, 2005. Pp. 198. $135.00 cloth; $36.95 paper. Reviewed by Daniel Santore, University at Albany, SUNY
What are the conditions under which workers’ rights legislation can produce revolutionary change in the workplace? O’Brien proposes an answer to this question by marshaling an array of Western theories of bodies and human action, and considering the ‘‘radical potential’’ of the Americans with Disabilities Act (ADA). According to O’Brien, the ADA’s open-ended definition of worker need and its (as yet unrealized) affinity with humanist values make it a potential agent of change in the logic of the capitalist workplace. The ADA can benefit all workers, not simply discrete factions of workers or those workers fitting a narrow definition of disability, by making employers and courts accommodate individuals’ varied modes of activity at work. O’Brien asserts that the ADA offers workplace accommodations on the basis of individual, and thus endlessly varying, needs. This
446
Book Reviews
model of worker protection is substantively different from protections based on categorical identities, such as age, sex, or race. The ADA, O’Brien argues, provides a chance to view disability not as a fixed identity, but as a universal human condition that generates a range of unique needs. Disability so defined refers to variety in conditions of human activity (e.g., pregnancy, depression) and is not limited to those differences that tend to constitute the popular image of a ‘‘handicap’’ (e.g., developmental disability or physiological injury). O’Brien speculates that this inclusive interpretation of disability, if adopted energetically by political actors within certain ideal institutional arrangements, can generate a revolutionary collective consciousness among workers and perhaps even employers. The heart of O’Brien’s book presents (1) a conceptual framework for an ‘‘ethic of care’’ in the contemporary workplace, and (2) the social context in which this ethic might flourish. An ethic of care is an ideal orientation toward workers’ needs that flows from emerging, inclusive definitions of disability. Drawing on feminist theories of difference and power, O’Brien outlines in Chapters Two and Three a ‘‘workers’ cause’’ logic that eschews securing an equality of circumstances and rights among individuals, instead pursuing the benevolent accommodation of difference in the workplace. This represents a move away from using ‘‘normalcy’’ and ‘‘average capabilities’’ as measuring sticks for employer obligations toward employees; it also rejects the notion of disability as a static identity. O’Brien arrives at this alternative workplace ethic of care after thoughtfully discussing competing theories of ‘‘bodies and being,’’ including the work of Descartes, Deleuze, Guattari, Foucault, and Spinoza. Those readers engaged with social theory will find much to like in O’Brien’s broad theoretical readings. Ultimately O’Brien links the ethic of care to those perspectives that understand humanity as defined by what people can do (i.e., by action). In O’Brien’s own conceptual language, ‘‘animality’’ captures the essence of human action: mind and body as a fused process out of which myriad, unique needs are generated and, ideally, satisfied through law. (Dis)abilities exist for every individual along a continuum of different conditions of being. O’Brien acknowledges that there is very little in the actual history of work in the United States that involves inclusive disability interpretations, much less a broad-based ethic of care and accommodation for workers. Chapters Four and Five review the evolution in the United States of an industrialized, hierarchical workplace that sought standardization and worker subjugation in the name of efficiency. O’Brien astutely points out that ‘‘mass industrial capitalism’’ pivoted on the assumption of ‘‘identical
Book Reviews
447
human bodies . . . considered (to be) interchangeable parts on the factory line’’ (p. 73). Increasingly rationalized management had no time for considering workers as individuals with singular needs. O’Brien argues against standardized procedures of control in the workplace, advocating instead an ‘‘individualized workplace’’ in which each worker ‘‘fits into the workplace like a piece of a continually changing jigsaw puzzle’’ (p. 91). Recognizing and accommodating individual conditions of being need not generate fractious relations among employees. Rather, all workers (with the aid of union guidance) can share the fruits of individually tailored disability protection, so long as the ADA is used in its fullest breadth, politically and legally. Social scientists working in several disciplines will appreciate the theoretical depth of O’Brien’s arguments and her historical account of the relations between labor and management under capitalism. Readers will struggle, however, with O’Brien’s limited detail of the ADAFthe key legislative seed from which inclusive social understandings of disability might grow. While O’Brien refers on many occasions to the open-ended language of the ADA, the law is not once quoted directly. Indeed, much of the sociopolitical context of the law (e.g., its sponsors, its public discussion) is absent. O’Brien’s argument for the ADA’s transformative potential would have been strengthened by linking her innovative conceptual understanding of human action with a more specific discussion of the actual terms of the ADA. On balance, though, O’Brien has written a thought-provoking book about work, the cultural power of law, and the possibility of new collective consciousnesses.
