Arkansas Law Review
58 Ark. L. Rev. 159
LENGTH: 13388 words
ARTICLE: The Allure and Danger of Practicing Law as Taxonomy
NAME: Marcia L. McCormick*
* Visiting Assistant Professor of Law, Chicago-Kent College of Law. For their suggestions, assistance with resources, and comments on prior drafts of this article, I would like to thank Katharine Baker, Mark Bauer, Howard Eglit, Doug Godfrey, Sanford Greenberg, Dan Hamilton, Sarah Harding, Hal Krent, Martin Malin, Nancy Marder, Joe Morrissey, John O'Connell, Michael Pardo, Joan Steinman, Mary Rose Strubbe, and Carolyn Shapiro. I would also like to thank all of the members of the Chicago-Kent faculty who participated in a workshop on this paper. Thanks, as well, to Paul Mollica, who allowed me to use his summary of employment discrimination cases, which was a valuable research tool. Finally, thank you to David Leavitt for outstanding research assistance. Any errors, technical or substantive, are mine alone.
... This country's approaches to the problem of discrimination, the laws we have enacted to resolve that problem, and the way that courts have interpreted those laws have been extensively criticized. ... In 1973, in McDonnell Douglas Corp. v. Green, the Supreme Court set forth what is now known as the McDonnell Douglas test which assigned burdens and order of proof in Title VII cases that alleged discriminatory treatment but lacked direct evidence. ... " The McDonnell Douglas test has been adopted, and the prima facie case has been subsequently modified to apply to disparate treatment cases brought under other employment discrimination statutes, non-employment discrimination contexts, and other portions of Title VII. ... This interpretation comports with the interpretation of nearly identical language in Title VII, which provides that employers may not "fail or refuse to hire or ... discharge any individual, or otherwise to discriminate against any individual ... because of such individual's race, color, religion, sex, or national origin ... ." That language has been interpreted to prohibit decisions made because of a prohibited characteristic, rather than membership in a subset of the class defined. ... The district court granted summary judgment on the ground that O'Connor could not make out a prima facie case of age discrimination because he was replaced by someone within his protected class. ...
This country's approaches to the problem of discrimination, the laws we have enacted to resolve that problem, and the way that courts have interpreted those laws have been extensively criticized. 1 While these critiques are related and form some of the basis for the critique presented here, this article accepts for the moment the positive law and structure already in place. 2 Rather than attempting to provide any definitive answer to these large issues, which is a project for another day, this article [*160] focuses more narrowly on the basic framework for evaluating employment discrimination cases at the summary judgment stage, and suggests a way to reform that framework with the goal of developing a more workable model.
One of the things that we all learn in law school is that lawyers love multi-part tests, multi-factor analyses, and shifting burdens. We love to break problems into smaller and smaller pieces, and then rearrange those pieces to best serve our purposes. Perhaps we use this process as a way to establish order in a chaotic world or as a way to make our subjective judgments seem more scientific. Perhaps it is a way to feel like we are using the expertise we spent so much time, sweat, and money developing. Perhaps it just makes us feel smart. Whatever the reason, the process is as natural to us as breathing.
But what happens when we get so caught up in the process that we lose sight of the ends that process was originally designed to serve? It would seem that when the means become self-serving or when they actually frustrate the ends they were originally intended to serve, we must discard them and refocus on a new way to serve those ends. To do otherwise would not only frustrate achieving the desired ends, but also might camouflage the fact that the ends are not actually being served. Sometimes, we have to start over from scratch.
We have reached this point in employment discrimination. The multi-part, burden-shifting test designed as a "sensible, orderly way to evaluate the evidence" 3 in an employment discrimination case has taken over how we define discrimination, and the Supreme Court itself has finally fallen into the trap. It is not necessarily that the test is too difficult for courts to apply. 4 On the contrary, the McDonnell Douglas Corp. [*161] v. Green 5 test [hereinafter "McDonnell Douglas test"] is satisfying, even fun, to apply and encourages lawyers and courts to get so caught up in principles promoted by the test that they lose sight of the law. Rather than being a way to assess the evidence in order to discover whether an inference can be made that the law has been violated, the test has replaced the law and redefined what discrimination means. By eclipsing the law, the test has made it nearly impossible to combat discrimination in all but the most egregious cases. Thus, it is time to recalibrate how we view the evidence that might prove discrimination in order to reorient ourselves to what discrimination actually is.
Part II of this article examines "discrimination" by looking at legal definitions and to the work of social psychologists. Part III then traces the development of the current analytical structure for employment discrimination cases. Part IV explains how the ubiquity of the current test has caused courts to substitute it for the laws they are supposed to apply, resulting in the misapplication of antidiscrimination laws and an overreliance on summary judgment. Finally, Part V proposes reformulating the test, making it clear that in summary judgment proceedings or proceedings to set aside a judgment, the explicit question for the court should be: would a minimally rational factfinder be required to find that the employer took an adverse employment action against the plaintiff solely for reasons unrelated to discrimination or discriminatory beliefs? In other words, only if a minimally rational factfinder were required to find that the employer acted solely for reasons unrelated to discrimination would the employer be granted summary judgment or the judgment of the jury be set aside. Otherwise, the case should either proceed to a trial, or the judgment of the jury should stand. At trial, the trier of fact should focus simply on whether the plaintiff has shown that the employer took the adverse job action at least in part for a reason related to discrimination. This test restores the original understanding of courts that employers [*162] generally act for some reason, and, when the most common reasons for an adverse employment action are rejected, all that remains is an inference that discrimination was the real reason. The ultimate burden on the plaintiff would be the same, and the employer would retain the same opportunity to argue that it had a legitimate reason for the action taken. However, the plaintiff would have a greater chance to influence how the court defines discrimination, and the court's perspective would better reflect the pervasive and subtle nature of discrimination in the workplace.
Discrimination, at its most basic and benign level, is the act of differentiating. 6 We discriminate all of the time in order to function. 7 In a positive sense, we discriminate between foods that are poisonous and those which are not in order to live. In a less beneficial, but still benign sense, we discriminate between the colors we like and those which we do not in decorating our living spaces. In a legal sense, we typically use discrimination to mean the unequal treatment of people, an act that often has far-reaching effects. While this is an informal definition of discrimination that most people would agree on, there is no societal consensus on what kind of discrimination the law should prohibit.
Players within the United States legal system have a rather narrow definition of equality and a narrow view of the role that government may legitimately play in ensuring equality. As a general rule, our legal system tends to protect equality of opportunity, or formal equality, rather than substantive equality. 8 As a result, our working definition of illegal discrimination focuses on what is in the mind of a particular decision-maker at [*163] the time a decision is made. 9 Illegal discrimination is generally thought of, in a doctrinal sense, as the self-aware, 10 intentional treatment of a particular person because of a single-dimensioned characteristic that is both irrelevant, at least to the decision being made, and also outside of the person's control. 11 This is the type of discrimination that the Equal Protection Clause has been interpreted to prohibit, and this model has been the basis for our antidiscrimination statutes.
While this intentional discrimination model is the paradigm, legislators, courts, and scholars have recognized that the government might appropriately work to provide a more substantive equality. 12 This approach tends to focus on the effects of particular actions, rather than on the intent of the actors. 13 As a result, some actions will be considered illegal discrimination if their effects perpetuate a system of inequality, even if the actors did not have the specific, fully self-aware intent to treat people badly based on a prohibited characteristic.
This latter model more accurately reflects how we as [*164] humans actually discriminate. 14 The things in our world contain infinite variations and if we were constantly confronted with having to process the impact of each variation we would not be able to function. 15 To cope, we define categories of things, and then quickly sort that which we encounter into those established categories. 16 We then use the definition of the categories to explain what the "thing" we have encountered is, and how it is likely to act or be acted upon. 17
Assigning a "thing" to a group or creating group identity has far-reaching effects. For instance, we perceive objects belonging to a defined group as more similar to each other, and more unlike things outside of the group, than we would if the objects were not attached to a group. 18 Studies have shown that when individuals are assigned group identities, even on an arbitrary basis, these effects are personalized: the individuals see members of their own group (the ingroup) as more like themselves, and others (the outgroup) as more different from themselves than they would in the absence of the group identity. 19 Additionally, they are much less able to see [*165] differences among members of the outgroup. 20 Groups are created by the salience of characteristics. Once a characteristic becomes salient (matters or makes a difference), like gender or race, that characteristic defines a group; however, individuals define what is salient in any given context. 21
These cognitive structures create the tendency to stereotype, and stereotypes are essentially cognitive shortcuts that link personal traits with group membership, in order "to simplify the task of perceiving, processing, and retaining information about people in memory." 22 Once set, stereotypes bias "in predictable ways the perception, interpretation, encoding, retention, and recall of information about other people" and influence judgment continuously. 23 Like salience, which defines groupness in the first place, we decide what behaviors to attribute to particular groups. 24
Stereotypes create expectations that bias the way individuals perceive others, remember things about others, and assign cause to the actions of others. 25 We determine whether a particular person is suited for a job by comparing the stereotypes we associate with that person to the stereotypes we associate with the job. 26 We also tend to remember the things a person [*166] actually did that fit our stereotypes of that person; we believe we remember a person doing things consistent with the stereotypes even if the person never did them, and we forget the things about them that do not conform to those stereotypes. 27 Additionally, we assign causes to the actions of people in accordance with our stereotyped expectations. For instance, we assume a person who acts consistently with a stereotype did so because of innate characteristics, while we assume that a person who acts inconsistently with a stereotype did so because of transitional or situational factors. 28
Thus, discrimination is accomplished, at least in part, through an ongoing process of interaction that often happens outside of a person's normal self-awareness, and which manifests in many small things over time, and may culminate in larger actions. Yet, stereotypes do not function entirely automatically and can be controlled by conscious effort. 29 Therefore, even though some discrimination may happen without full, contemporaneous self-awareness, discrimination is still an appropriate subject of regulation by the government. 30
Based on this social science research showing that the doctrinal model of discrimination is inaccurate, a wave of scholarly criticism has recently focused on that model and the [*167] courts' adherence to it to enforce our antidiscrimination laws. 31 As the next section demonstrates, there is nothing in our positive law that requires this model, and so courts are free to adopt a definition of discrimination that comports with how social science tells us discrimination operates. And yet, courts have done so very rarely. I submit that the reason is because the analytical structure itself hides the issue of discrimination, so that litigants never have a chance to challenge the judges' assumptions about what discrimination is.
III. LEGAL RESPONSES TO DISCRIMINATION
The first legal mechanisms created to address social inequality and the problems of discrimination came at the end of the Civil War and during Reconstruction. 32 While there were [*168] some steps forward during the Civil War, 33 the most notable developments came after, with the Thirteenth, 34 Fourteenth, 35 and Fifteenth 36 Amendments to the United States Constitution. 37 [*169] The effect of these amendments was quickly narrowed, 38 and [*170] few efforts were made until the mid-twentieth century to address discrimination through law. 39 Commonly considered the spark of the Black Civil Rights Movement, Brown v. Board of Education 40 was decided in 1954, when the Court declared that racial segregation in schools violated the Equal Protection Clause. 41 After Brown helped set the stage, 42 and when activists [*171] undertook massive efforts at civil disobedience to integrate public accommodations, 43 Congress began enacting antidiscrimination legislation again in 1957. 44 It passed a second civil rights act in 1960, 45 and finally enacted the most sweeping and widely used antidiscrimination legislation in 1964. 46
A model legal strategy for civil rights movements evolved out of this history. After an initial period of dawning, group awareness and local and national activism that resulted in an organized, coherent movement, a group seeking greater civil rights would focus its energies at the federal level by seeking a declaration from the Supreme Court that particular discriminatory practices violated the Constitution. 47 If that top- [*172] down approach was not successful, the group would seek a constitutional amendment and/or federal legislation outlawing the practices. 48 If those efforts were unsuccessful or struck down by the courts, the group would head to the states or other local bodies to try the same tactics and engage in more grass-roots efforts. 49 After some time and state or local successes, the group would again focus its efforts on the federal government, beginning with the Supreme Court and repeating the process. 50
This strategy was probably chosen to maximize the effect of group resources. There is only one federal government, while there are fifty state governments, and thousands of smaller local government bodies. Focusing on the federal government allows a greater concentration of resources. Additionally, a declaration by the highest court of any jurisdiction that the jurisdiction's constitution requires a certain outcome provides the most absolute protection. For example, if the United States Supreme Court finds that the Constitution protects a particular right or a particular class, then no government within the United States may restrict that right or burden that class, and governmental bodies may be empowered to prohibit private parties from restricting the right or burdening the class as well. Moreover, [*173] once a court finds a right grounded in or a class protected by its constitution, neither it nor any other court can easily roll that protection back. 51 Even if the right itself is not grounded in a constitution, a legislature may still have the power to recognize and protect the right under a more general type of power. And so, even if the group is more likely to get a good result at a local level, for maximum effect it makes sense to focus resources first on the United States Supreme Court, then Congress, then individual state supreme courts, then individual state legislatures, and then other local governmental bodies.
At least in part because of this strategy, which focuses on federal constitutional protections as an ideal, our model for defining illegal discrimination is the Equal Protection Clause. So, it is not surprising that when the Civil Rights Act of 1964 was passed, the Equal Protection model of illegal discrimination would be applied to Title VII, 52 the employment discrimination provisions.
Title VII provides, in part:
It shall be an unlawful employment practice for an employer -
(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's race, color, religion, sex, or national origin; or
(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's race, color, religion, sex, or national origin. 53
[*174] Section (a)(1) has been interpreted to prohibit only intentional discrimination, which has been labeled disparate treatment. 54 Section (a)(2), on the other hand, has been interpreted to prohibit discriminatory effects, and that theory is labeled disparate impact. 55 Even though disparate impact cases look to effects rather than intent, "the necessary premise of the disparate impact approach is that some employment practices, adopted without a deliberately discriminatory motive, may in operation be functionally equivalent to intentional discrimination." 56 In other words, some seemingly neutral policies have hidden biases so strong that the decision-maker may as well have used an explicit classification to keep members of the protected class out of the job. Moreover, despite initially giving broad effect to the disparate impact provisions of Title VII, 57 the courts have become progressively less willing to enforce those provisions. 58 Part of the reason for [*175] that could be that the Equal Protection Clause has been interpreted to prohibit only acts taken with a discriminatory purpose and not acts that have a discriminatory effect. 59 [*176] Because the Equal Protection Clause serves as the paradigm for defining illegal discrimination, any deviation from that paradigm may seem less legitimate and more problematic. Thus, most of the cases involving Title VII these days concern disparate treatment, and that is where this article will focus. 60
A. The McDonnell Douglas Test
After Title VII became law, employers who intentionally discriminated quickly learned not to admit that they were taking the particular employment action based on the person's protected class. Thus, direct evidence of the employer's motive dwindled. In 1973, in McDonnell Douglas Corp. v. Green, 61 the Supreme Court set forth what is now known as the McDonnell Douglas test which assigned burdens and order of proof in Title VII cases that alleged discriminatory treatment but lacked direct evidence. 62 Under this test, once a plaintiff demonstrates a prima facie case, the court presumes that discrimination has occurred. 63 The defendant must then rebut that presumption by [*177] producing evidence that the employment decision at issue was motivated by a non-discriminatory reason. 64 If the defendant produces such evidence, the plaintiff must show that the reason given by the defendant is a pretext for discrimination. 65 The plaintiff's membership in the protected class need not be the sole reason for the employer's actions; it need merely be a motivating factor. 66 The defendant cannot avoid liability at this stage, but it can severely limit the plaintiff's remedies if it proves that it would have made the same decision even if the protected characteristic had played no role in the decision. 67
If, however, the defendant fails to come forward with evidence of a non-discriminatory reason for its actions, a "court must enter judgment for the plaintiff because no issue of fact remains in the case." 68 While the Court explained what would happen when a defendant failed to carry its burden of production, it did not, at that time, explain what should happen when the employer's stated reason is shown not to be the true reason for its actions. 69 A split developed in the lower courts on [*178] what they should do in that situation. One group of courts held that the effect was the same as if the employer had submitted no evidence at all (pretext only). 70 Another group of courts held that disbelief of the employer's stated reason had no effect, and that the plaintiff had to provide additional specific evidence of discrimination (pretext plus). 71 The Supreme Court attempted to resolve the split in St. Mary's Honor Center v. Hicks 72 and clarify that a court was not required to find in favor of the plaintiff if it found that the defendant's stated reason was not the real reason, but that it could do so, because an inference of discrimination could be made from the prima facie case, and that inference would be bolstered by the fact that the employer offered only a cover-up. 73 However, the Court only moved more courts toward the pretext plus test by, in a majority of the discussion, emphasizing the plaintiff's burden of persuasion. 74 The Court did resolve the issue, however, in Reeves v. Sanderson Plumbing Products, Inc., 75 making clear that a court could infer discriminatory intent from the fact that the [*179] employer's reason was disbelieved, 76 and that even though the presumption of discrimination dropped out once the defendant articulated a non-discriminatory reason for its action, a permissive inference of discrimination remained. 77 Thus, a plaintiff may show intentional discrimination simply by proving a prima facie case, and that the defendant's "explanation is unworthy of credence." 78 The McDonnell Douglas test has been adopted, and the prima facie case has been subsequently modified to apply to disparate treatment cases brought under other employment discrimination statutes, 79 non-employment discrimination contexts, 80 and other portions of Title VII. 81
B. Despite its Flexibility, Courts Apply McDonnell Douglas Rigidly
The Supreme Court originally designed this test as a "sensible, orderly way to evaluate the evidence," 82 understanding that "there will seldom be "eyewitness' testimony to the employer's mental processes." 83 The logic behind the presumption created by the prima facie case was that the Court "presumed these acts, if otherwise unexplained, are more likely than not based on the consideration of impermissible factors." 84 The Court recognized that discrimination was [*180] pervasive but often subtle and that "the question facing triers of fact in discrimination cases is both sensitive and difficult." 85 Title VII embodied "important national policy" which sought to eradicate societal discrimination. 86 At the same time, mental processes and states of mind are nearly impossible to prove, particularly in situations where the actions taken could have been taken for any number of legitimate reasons, because the person whose state of mind is at issue has sole access to that information. 87 Yet, the adjudicative process requires a certain form and quality of proof. 88
Consequently, in an effort to serve the policy of eradicating discrimination, but in recognition that the intent of an actor would be difficult to prove, the McDonnell Douglas test does not ask whether there has been discrimination. It endeavors to take that question out of the picture, and focuses instead on the acts from which we can infer intent. The Court's decision not to have the lower courts define discrimination on the basis of a protected class was probably deliberate. Either the Court believed that there was a consensus, in 1973, on what constituted discrimination, or it recognized that there was no consensus and attempted to guide the lower courts by giving them the tools to avoid that sticky issue.
Regardless of the exact reason, the rule has had the opposite effect. Despite the Supreme Court's expansive language about the subtlety of discrimination, the flexibility of the test, and the broad inferences that can be drawn from particular kinds of evidence, 89 courts find for plaintiffs in employment discrimination cases significantly less frequently than in other types of civil cases. 90 Furthermore, verdicts in [*181] favor of plaintiffs are more than four times as likely to be reversed than verdicts in favor of defendants. 91 While these statistics could demonstrate many things, for example that there are more frivolous employment discrimination cases filed than in other civil contexts, most scholars who have done empirical studies on verdicts and reversal rates have concluded that courts find ways to rule for defendants in cases that fit the prohibition of Title VII and satisfy the McDonnell Douglas test. 92 In other words, the courts are dismissing, granting summary judgment, or (if courts of appeal) reversing judgments in meritorious cases. Moreover, the reason cannot be that there is no longer any discrimination: discrimination has not been eliminated; many scholars have documented the way that discrimination continues to operate in our society. 93
The pressure posed by the national policy against discrimination in the face of the difficulty of proving states of mind and the form and quality of proof required in our legal system is likely one reason that the lower courts have been so reluctant to find in favor of plaintiffs when the McDonnell Douglas test is applied. Other reasons could be that judges do not believe discrimination exists or that to the extent it may, that type of discrimination is not the kind of discrimination that the government should penalize. 94 In other words, there appears to be no consensus on what discrimination means.
This differentiation between what types of discrimination should and should not be penalized has deep roots, reaching back to the debates over the Reconstruction Amendments. At the time that the country was debating how (and whether) to [*182] tackle the problem of race-based slavery, much of the debate over the Reconstruction Amendments and civil rights legislation focused on what civil rights were. Those resisting any efforts at equality suggested that the measures proposed were targeted at social inequality, a vague concept that tied into the public/private dichotomy. 95 Proponents of the measures denied that they had anything to do with personal prejudices that would force people to accept any member of a particular class into their homes or their circle of friends. 96 Yet many feminist scholars have suggested that the public/private distinction is a social convention often used to justify discrimination. 97 For example, because home and family are parts of the private sphere, intimate violence was historically not penalized. 98 And while few would suggest that the government should be able to mandate who one's friends are, there is often little clear division between the public and the private spheres. 99 Sexual harassment in the workplace was once thought of as private because of the sexual component, even though it took place in the public sphere of the workplace. 100 Conversely, same-sex sodomy was criminalized as a public wrong even when it took place in the private realm of the home. 101 Thus, relegating the entire category of social inequality to a particular category of things not appropriate for government regulation merely legitimizes more subtle forms of discrimination.
IV. THE TEST'S ECLIPSE OF THE LAW
Just as neutral policies can mask intentional discrimination, the structure of the McDonnell Douglas test has obscured the failure of courts to apply a stricter definition of discrimination than exists or than is required by our antidiscrimination laws. It allows one to get bound up in the taxonomy of the evidence, placing each piece into a separate box and evaluating each box separately rather than looking at the whole picture to determine whether discrimination occurred. 102 Additionally, it has allowed us to avoid grappling with the definition of discrimination and, for many courts, has actually defined "discrimination."
The best example of the McDonnell Douglas test's effect is the split that resulted from St. Mary's Honor Center v. Hicks 103 and led to the decision in Reeves v. Sanderson Plumbing Products, Inc. 104 Recall, the Court in Hicks held that a trial court is not required to enter judgment for the plaintiff if it disbelieves the employer's reason, but it may, as long as it inferred from the prima facie case that the reason offered was not the real reason and that instead discrimination was the real reason. 105 Despite this permissive language, many courts interpreted Hicks to mean that trial courts were not allowed to enter judgment for the plaintiff unless the plaintiff offered specific evidence that would alone suggest discrimination. 106 The courts got caught up in categorizing the evidence based upon whether it was direct or circumstantial, or whether it was part of the prima facie case or proof of pretext. In the process, the courts lost sight of the ultimate issue: do the facts as a whole [*184] suggest that the employer discriminated?
The best illustration of the courts' digression is the decision by the Fifth Circuit Court of Appeals in Reeves. Reeves involved a fifty-seven-year-old factory supervisor who was terminated after an investigation into employee timekeeping practices. 107 He was replaced by someone substantially younger. 108 Several months before his discharge, Reeves's supervisor said that he "was so old [that he] "must have come over on the Mayflower,'" and that he "was "too damn old to do [the] job.'" 109 After each side rested its case, the jury found that the discrimination against Reeves was willful, returned a verdict in favor of Reeves, and awarded him damages. 110
The Fifth Circuit reversed the jury's verdict, holding that Reeves had not presented enough evidence to show that discrimination was the reason for his discharge. 111 After noting that the employer had articulated a non-discriminatory reason for its actions, the court of appeals failed to consider the evidence that made up the prima facie case and looked only to the age-related comments as potential evidence of discrimination. 112 It found that those comments could not be linked to Reeves's discharge since they were made months before. Without those comments, the evidence showed that Reeves was treated like any other employee under investigation and that there was no widespread age-based animus in the employer's practices. 113 Thus, the court of appeals boxed each piece of evidence into its own category, and then used those categories to disregard the probative value of that evidence. When the Supreme Court reversed, it admonished the Fifth Circuit for rejecting the evidence contained in the prima facie case and found that when viewed as a whole, the evidence supported the verdict. 114 The Court went further, as well, and directed that a factfinder could infer a discriminatory motive from the fact that the reason offered by the employer for its [*185] action was not the real reason. 115
While the Supreme Court's decision in Reeves resolved the issue in that particular case, it has not solved the problems associated with the boxing of evidence. 116 Although McDonnell Douglas has been called an empty ritual because courts focused primarily on the pretext prong for a number of years, 117 Reeves revitalized the power of the prima facie case. 118 Courts are more frequently examining whether the plaintiff has provided evidence of a prima facie case. 119 And, the elements of the prima facie case are being applied very rigidly. For example, one way to show that the plaintiff's prohibited characteristic was the reason for an employer's action is to show that someone outside the plaintiff's class was treated better under substantially similar circumstances. Yet some courts have taken it too far and found this comparator evidence a necessary part of the prima facie case. 120 However, that is not the only way to show that the prohibited characteristic was the reason for the employer's action. For example, in Price Waterhouse v. Hopkins, 121 comparative evidence was unnecessary. It was irrelevant that some women could become partners at the accounting firm. 122 [*186] It did not matter that men who exhibited the same behaviors as plaintiff, Ann Hopkins, were promoted to partner. 123 The problem was that Hopkins was not promoted because she was a woman who acted too manly. 124
Categorizing the evidence is not the only way that courts and litigants get caught up in the principles that the McDonnell Douglas taxonomy promotes. Courts are importing the test as if the test defines what violates the statute at issue rather than being a way to infer intent with the idea that discrimination operates in the background. In fact, sometimes the test is imported into areas where it is simply inapplicable. A prime example of this improper importation is retaliation under Title VII. Title VII prohibits retaliation by employers for opposing any unlawful employment practice or participating in proceedings under Title VII. 125 The only question relevant to a retaliation charge is whether the employer took some act and whether the employer took the act because the employee opposed an unlawful employment practice or participated in proceedings under Title VII. 126 However, where there is no direct evidence of the employer's intent, courts use the McDonnell Douglas test to determine whether retaliation has occurred. One element of that test is whether the employee suffered an adverse employment action. 127 While a modified version of the test could be useful to isolate the evidence that might show the employer's intent, the courts have not modified the test to remove the adverse action requirement, a requirement that has nothing to do with the intent of the actor. The circuits are split on what constitutes an employment action adverse [*187] enough, in the context of retaliation, to provide a person a cause of action. 128
However, no court has questioned the utility of using McDonnell Douglas in the first instance, particularly the adverse job action requirement. 129 That requirement makes sense in the context of Title VII, which prohibits an employer from "discriminating against any individual with respect to ... compensation, terms, conditions, or privileges of employment ... ." 130 This language suggests that only discrimination that results in adverse employment actions will be actionable. The retaliation provision, on the other hand, prohibits "discrimination" without any qualifiers, 131 and thus mere differentiation should be sufficient.
Another, even more glaring example is the importation of the test into the Family and Medical Leave Act ("FMLA"). FMLA requires covered employers to allow covered employees to take up to twelve weeks of unpaid leave to care for a newly born or adopted child, a family member, or the employee's own serious health condition. 132 In addition, it gives employees a cause of action against employers who "interfere with, restrain, or deny the exercise of or the attempt to exercise, any right provided under [FMLA]." 133 Despite the fact that interference with rights is much broader than discrimination and looks to the effects of actions rather than the intent of actors, 134 some courts [*188] have imported the McDonnell Douglas test into the FMLA. 135 Thus, many actions that could be said to interfere with an employee's exercise of rights have not been found to violate the FMLA despite the act's plain language.
Finally, the taxonomy promoted by McDonnell Douglas, particularly the issue of comparators, has so fully infected our thinking about what discrimination is that the Supreme Court has fallen into the trap in its recent decision in General Dynamics Land Systems, Inc. v. Cline. 136 Cline involved the question of whether an employer violated the Age Discrimination in Employment Act ("ADEA") by treating workers over forty less favorably because of their age than workers over fifty. 137 The antidiscrimination provision of the ADEA prohibits employers from "failing or refusing to hire or to discharge any individual or otherwise discriminate against any individual ... because of such individual's age ... ." 138 However, not everyone is protected from age discrimination: the class of workers protected by the ADEA is limited to those forty and older; which leaves workers under forty unprotected. 139 The plain meaning of the two sections suggests that as long as a person is in the protected class, an employer may not rest decisions affecting that person on that person's age. 140
This interpretation comports with the interpretation of [*189] nearly identical language in Title VII, which provides that employers may not "fail or refuse to hire or ... discharge any individual, or otherwise to discriminate against any individual ... because of such individual's race, color, religion, sex, or national origin ... ." 141 That language has been interpreted to prohibit decisions made because of a prohibited characteristic, rather than membership in a subset of the class defined. In other words, white people may sue if they are discriminated against because they are white, men may sue if they are discriminated against because they are men, and so on, even though Title VII was originally passed to promote equality for African-Americans, women, and other disempowered groups. 142
The application of these principles seems relatively simple in the abstract. In Cline, the employer decided to take away future benefits at retirement for those employees between forty and fifty based solely on their current age. 143 This was direct evidence that the employer was motivated to take action solely because of age. Based on the plain language of the statute, stating that employers cannot discriminate because of an individual's age, this action should have been held to violate the ADEA. Because it was a direct evidence case, McDonnell Douglas considerations should never have entered the picture. However, there was a wrinkle presented by Cline. Employees over fifty would have greater benefits when they retired, and they retained these benefits in part because the employer cut benefits for the younger employees. 144 Thus, the same action that harmed one group of employees within the protected class helped another group also within the protected class.
The presence of these comparators confused the issue. Suddenly the issue was no longer the employer's intent. The employer's intent was undisputedly to classify on the basis of age. Instead, the question became whether the ADEA prohibited classifying on the basis of age when that classification benefited one segment of the protected class - not [*190] just any segment of that class, but the segment with the greatest amount of, for lack of a better word, protectedness. Even though the language of the ADEA was clear, the Court just could not conceive of Congress intending to protect all members of the class equally. And so, in a very tortured line of reasoning, the Court found that Congress meant "relatively older age" when it used the word "age," 145 and that the members of the protected class were protected only when they were harmed in favor of relatively younger members of the class. 146 This cannot be distinguished from Title VII by the fact that the comparators were members of the same class. We know from Price Waterhouse that it is still sex discrimination if a woman does not receive a benefit because she is not womanly enough. 147 Certainly, no court would find that refusing to offer benefits to a group of Black people because they were not Black enough was legal under Title VII. 148
Further demonstrating the confusion engendered by the desire to look at comparators, the Court looked to one of its prior decisions applying the McDonnell Douglas test in an ADEA case for guidance even though this was a direct evidence case, and the test would not be used. 149 In the prior case, O'Connor v. Consolidated Coin Caterers Corp., 150 the Court applied the McDonnell Douglas test to a case under the ADEA in which a fifty-six-year-old was discharged and replaced with a [*191] forty-year-old. 151 The district court granted summary judgment on the ground that O'Connor could not make out a prima facie case of age discrimination because he was replaced by someone within his protected class. 152 The Supreme Court held in O'Connor that the class membership of the replacement was irrelevant, and that an inference that age was the reason for the employer's decision could be made since the replacement was "substantially younger." 153
The Court in Cline took that point out of context, and inflated its importance. It ignored its prior reasoning in O'Connor, which recognized:
The discrimination prohibited by the ADEA is discrimination "because of [an] individual's age," 29 U.S.C. 623(a)(1), though the prohibition is "limited to individuals who are at least [forty] years of age," 631(a). This language does not ban discrimination against employees because they are aged [forty] or older; it bans discrimination against employees because of their age, but limits the protected class to those who are [forty] or older. The fact that one person in the protected class has lost out to another person in the protected class is thus irrelevant, so long as he has lost out because of his age. 154
Thus, the reasoning in O'Connor would have supported interpreting the ADEA according to the plain meaning of the statutory language. But the end result, the discussion of the comparator, and the presence of a comparator class in Cline, allowed the Court to interpret the ADEA as if the McDonnell Douglas test defines unlawful discrimination.
V. THE BETTER TEST
Given that the McDonnell Douglas test as applied is so flawed, and courts have proven so resistant to moving away from the most narrow interpretations of it, I suggest replacing it with a much simpler test. Judges 155 are just as prone to [*192] cognitive bias and unreflective discrimination as the rest of us. 156 Any test that fails to account for this fact is doomed to the same fate as the McDonnell Douglas test. Thus, I propose coming at the ultimate issue from the opposite angle, to reaffirm the notion that discrimination is not merely an exception to the rule.
In pre-trial proceedings or proceedings to set aside a judgment, the explicit, and only question for the court is whether a minimally rational factfinder would be required to find that the employer took an adverse employment action against the plaintiff solely because of a reason unrelated to discrimination or discriminatory beliefs. Only if a minimally rational factfinder would be required to find that the employer acted for reasons unrelated to discrimination would the employer be granted summary judgment or the judgment of the jury be set aside. Otherwise, the case should either proceed to a trial or the judgment of the jury should stand.
This formulation of the test requires no taxonomy of the evidence. Rather, it requires the court to look at the evidence as a whole with the ultimate issue at trial in mind, just like in any other type of civil litigation. Thus, it runs significantly less risk that the court will get so caught up in boxing the evidence into separate categories that it loses sight of the permissible inferences from that evidence.
Using a test that is so much more holistic may concern litigants and judges that the lack of defined standards will fail to give lower courts sufficient guidance. However, sociological research indicates that more complicated environments and [*193] instances with greater economic interests at stake result in tightly bound rules providing less stable outcomes than do more loosely defined principles. 157 It is difficult, the more strict and specific the rules, to apply them to situations that vary from the model for which the rules were designed. Certainly, the workplace is a complicated environment, particularly when discrimination is at issue. The norms governing what constitutes discrimination are in flux; any workplace's system of interpersonal interaction will be highly complex, and every case has high economic stakes. Given the complicated nature of the issues then, employment discrimination cases are much better suited to broader principles than bound rules.
The test might also give judges difficulty by requiring them to define what discrimination is, something the current McDonnell Douglas test does not explicitly require as part of the legal analysis. However, underlying every single judgment as to what inferences can be drawn and where the pieces of evidence fit, is the judge's view of discrimination. And that view is never exposed or subject to challenge. So giving the parties the opportunity to present expert testimony on what discrimination is and how it operates will result at the very least in documentation in the record about what definition of discrimination a judge is using. This, in turn will foster public debate, may lead to congressional action, and could lead us closer as a society to consensus on the issue. For example, one of the most noteworthy things about Price Waterhouse v. Hopkins 158 is that Hopkins presented expert testimony on cognitive bias and the operation of stereotypes, and this evidence influenced the Supreme Court's discussion of why the beliefs held about Hopkins were discriminatory. 159 After Price Waterhouse, more people, judges included, understand better how stereotypes are manifestations of discrimination.
Another consequence of the test is that fewer cases will be disposed of on summary judgment, and thus, more cases will go to trial. While this increase in the federal court caseload might appear to be a bad outcome due to a desire for judicial [*194] efficiency, the answer is not to grant summary judgment more frequently. Meritorious cases with evidence on both sides should go to trial. 160 That is what the statute requires and the purpose served by the summary judgment procedure. If caseloads are too heavy, the government should create more judicial positions, more ancillaries to judges, or an administrative adjudication procedure. It should not allow trial by affidavit. 161
Using this test will also make the inquiry at trial much more straightforward. At trial, the factfinder should focus simply on whether the plaintiff has shown that the employer took the adverse job action for a reason related to discrimination. That is the only inquiry that matters, as the Supreme Court has repeatedly made clear. 162 Moreover, using this test resolves the question of how to incorporate the employer's affirmative defense, that it would have taken the same action without the prohibited reason, which the Supreme Court recently made clear applies in circumstantial cases as well as direct evidence cases. 163 Without the trips and traps of McDonnell Douglas, the real issues can come out, and courts can take a much more realistic look at discrimination in the workplace.
As these examples demonstrate, while the McDonnell Douglas test was originally designed to recognize the subtlety of discriminatory practices and to make the inquiry into the employer's intent easier, it has instead frustrated the operation of our antidiscrimination laws. The alternative test that I propose restores the original understanding of the Court, that discrimination operates throughout our society and that when the [*195] most common reasons for an adverse employment action are rejected, all that remains is an inference that discrimination was the real reason. The ultimate burden on the plaintiff is the same, and the employer retains the same opportunity to suggest that the real reason was a legitimate one, but the courts' perspective better reflects the pervasive and subtle nature of discrimination in the workplace.
For related research and practice materials, see the following legal topics:
Labor & Employment LawDiscriminationDisparate TreatmentProofBurdens of ProofLabor & Employment LawDiscriminationDisparate TreatmentProofBurden ShiftingLabor & Employment LawDiscriminationActionable Discrimination
n1. See, e.g., Catharine A. MacKinnon, Sex Equality 2-43 (2001) [hereinafter MacKinnon, Sex Equality] (providing a feminist critique of formal equality in the context of sex discrimination law); Catharine A. MacKinnon, Toward a Feminist Theory of the State 215-34 (1989) [hereinafter MacKinnon, Feminist Theory of the State]; Derrick Bell, An Allegorical Critique of the United States Civil Rights Model, in Discrimination: The Limits of Law 3, 11 (Bob Hepple & Erika M. Szyszczak eds., 1992); Alan Freeman, Antidiscrimination Law: The View from 1989, in The Politics of Law: A Progressive Critique 121 (David Kairys ed., rev. ed. 1990) [hereinafter Freeman, Antidiscrimination Law]; Alan David Freeman, Legitimizing Racial Discrimination Through Antidiscrimination Law: A Critical Review of Supreme Court Doctrine, 62 Minn. L. Rev. 1049, 1052-57 (1978) [hereinafter Freeman, Legitimizing Racial Discrimination]; D. Marvin Jones, No Time for Trumpets: Title VII, Equality, and the Fin de Siecle, 92 Mich. L. Rev. 2311, 2334-44 (1994); Robert Post, Prejudicial Appearances: The Logic of American Antidiscrimination Law, 88 Cal. L. Rev. 1, 12-19 (2000).
n2. It is possible, and perhaps probable, that the only way to correct our approach to discrimination is to develop an entirely new socio-legal approach.
n3. Furnco Constr. Corp. v. Waters, 438 U.S. 567, 577 (1978). The test was created by the Court in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 (1973), and is commonly referred to as the "McDonnell Douglas" test. See infra notes 61-81 and accompanying text.
n4. Several scholars have argued that the McDonnell Douglas test is too difficult to apply, and cite to cases in which courts bemoan the difficult time they are having. See, e.g., Denny Chin & Jodi Golinsky, Moving Beyond McDonnell Douglas: A Simplified Method for Assessing Evidence in Discrimination Cases, 64 Brook. L. Rev. 659, 660-73 (1998); Ernest F. Lidge, III, The Meaning of Discrimination: Why Courts Have Erred in Requiring Employment Discrimination Plaintiffs to Prove That the Employer's Action Was Materially Adverse or Ultimate, 47 U. Kan. L. Rev. 333, 342 n.45 (1999); Deborah C. Malamud, The Last Minuet: Disparate Treatment After Hicks, 93 Mich. L. Rev. 2229, 2232-39 (1995). I would submit that the problem is not that the test itself is difficult to apply; instead, courts simply do not believe that satisfying the test is enough to prove discrimination. Accord Robert J. Gregory, There is Life in That Old (I Mean, More "Senior") Dog Yet: The Age-Proxy Theory After Hazen Paper Co. v. Biggins, 11 Hofstra Lab. L.J. 391, 423-24 (1994).
n5. 411 U.S. 792; see supra note 3.
n6. Webster's New World College Dictionary 411 (4th ed. 2000) (defining discrimination as "distinguishing differences").
n7. See Linda Hamilton Krieger, The Content of Our Categories: A Cognitive Bias Approach to Discrimination and Equal Employment Opportunity, 47 Stan. L. Rev. 1161, 1163-64 (1995).
n8. See Derrick A. Bell, Jr., Race, Racism and American Law 41 (2d ed. 1980); MacKinnon, Feminist Theory of the State, supra note 1, at 215-34. Other scholars have labeled this distinction as deliberate exclusion rather than structural or institutional discrimination. See Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 Colum. L. Rev. 458, 467-69 (2001).
n9. Krieger, supra note 7, at 1168-77. The focus changes just a little when the validity of legislation is at issue. See Palmer v. Thompson, 403 U.S. 217, 224-25 (1971). Rather than looking at the intent of the legislators, we look primarily to the words of the legislation to evaluate discriminatory purpose. Id. Similarly, when the application of legislation is challenged, we look to the patterns of application and use those to infer discriminatory purpose. Yick Wo v. Hopkins, 118 U.S. 356, 374 (1886).
n10. Linda Hamilton Krieger calls this "transparency of mind," or a person's ability to accurately identify why that person made a particular decision. See Krieger, supra note 7, at 1167, 1185-86. This rule is applied differently when legislation, rather than an individual action, is being analyzed. See Palmer, 403 U.S. at 224-25. For legislation, the real intent of any particular legislator is irrelevant, and it is the language of the legislation that matters. Id.
n11. See Michael Selmi, Proving Intentional Discrimination: The Reality of Supreme Court Rhetoric, 86 Geo. L.J. 279, 288, 291-94 (1997).
n12. An example of a legislative effort to provide substantive equality might be the disparate impact provision of Title VII of the Civil Rights Act of 1964. 42 U.S.C. 2000e-2(a)(2) (2000). Cases interpreting this provision focus on the effects of employer actions. Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971). An additional example of a court focusing on effects rather than intent is Brown v. Board of Education, 347 U.S. 483 (1954), in which the Supreme Court looked at the effects of racial segregation in education and determined that such segregation caused harm and was unconstitutional. For examples of scholarly discussions of substantive equality, see Kimberle' Williams Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 Harv. L. Rev. 1331, 1341 (1988); Freeman, Antidiscrimination Law, supra note 1, at 124-26.
n13. See supra note 12.
n14. The former view of discrimination may actually have been an accurate model of how people discriminated when the Fourteenth Amendment was adopted. Ann C. McGinley, !Viva la Evolucion!: Recognizing Unconscious Motive in Title VII, 9 Cornell J.L. & Pub. Pol'y 415, 418, 426-27 (2000) (suggesting that overt discrimination was prevalent until 1964, after which people learned not to engage in such discrimination but were not made to examine their cognitive processes to root out stereotypes and more subtle forms of discrimination).
n15. See Eleanor Rosch, Human Categorization, in Studies in Cross-cultural Psychology 1, 1-2 (Neil Warren ed., 1977).
n16. See Eleanor Rosch, Principles of Categorization, in Cognition and Categorization 27, 28 (Eleanor Rosch & Barbara B. Lloyd eds., 1978).
n17. Krieger, supra note 7, at 1188-89.
n18. See id. at 1186 (citing studies by: Donald T. Campbell, Enhancement of Contrast as Composite Habit, 53 J. Abnormal & Soc. Psychol. 350, 355 (1956); Henri Tajfel, Cognitive Aspects of Prejudice, 25 J. Soc. Issues 79, 83-86 (1969); and Henri Tajfel & A.L. Wilkes, Classification and Quantitative Judgment, 54 Brit. J. Psychol. 101, 104 (1963)).
n19. See Marilynn B. Brewer, In-Group Favoritism: The Subtle Side of Intergroup Discrimination, in Codes of Conduct: Behavioral Research into Business Ethics 160, 161-62, 167-68 (David M. Messick & Ann E. Tenbrunsel eds., 1996); Anne Locksley et al., Social Categorization and Discriminatory Behavior: Extinguishing the Minimal Intergroup Discrimination Effect, 39 J. Personality & Soc. Psychol. 773, 776-83 (1980); David A. Wilder, Perceiving Persons as a Group: Categorization and Intergroup Relations, in Cognitive Processes in Stereotyping and Intergroup Behavior 213, 217 (David L. Hamilton ed., 1981).
n20. David L. Hamilton & Tina K. Trolier, Stereotypes and Stereotyping: An Overview of the Cognitive Approach, in Prejudice, Discrimination, and Racism 127, 131 (John F. Dovidio & Samuel L. Gaertner eds., 1986).
n21. That is not to say that in every instance individuals make a conscious choice about what characteristics matter. Conscious adoption could happen, but it is also likely that individuals absorb information from exposure to the culture in which they live. See Howard J. Ehrlich, The Social Psychology of Prejudice 35 (1973).
n22. Krieger, supra note 7, at 1187-88; see also Barbara F. Reskin, The Proximate Causes of Employment Discrimination, 29 Contemp. Soc. 319, 321-22 (2000). While this description of what stereotypes are may sound very benign, stereotypes in a society with power imbalances such as ours operate to perpetuate and even aggravate such imbalances.
n23. Krieger, supra note 7, at 1188.
n24. Again, this decision could be consciously adopted or learned through exposure to culture. David L. Hamilton, A Cognitive-Attributional Analysis of Stereotyping, in 12 Advances in Experimental Social Psychology 53, 64 (Leonard Berkowitz ed., 1979).
n25. See Krieger, supra note 7, at 1200-09.
n26. Id. at 1200-04. Krieger uses the example of a small woman with paralegal training and a large man with physical education training both applying for the position of police officer to illustrate this point. Id. at 1200. If our stereotype of a police officer includes a physically imposing person, we would be more likely to perceive the large man as a better candidate. Id. at 1201. On the other hand, if our stereotype of a police officer includes a person able to defuse tense situations or apply the law correctly to particular conduct, we would be more likely to perceive the smaller woman with legal training as the better candidate. Id.
n27. See Krieger, supra note 7, at 1206-10; see also Nancy Cantor & Walter Mischel, Traits as Prototypes: Effects on Recognition Memory, 35 J. Personality & Soc. Psychol. 38, 41-45 (1977).
n28. See Krieger, supra note 7, at 1204-07. A good example of such attribution bias is given by Joan C. Williams. See Joan C. Williams, The Social Psychology of Stereotyping: Using Social Science to Litigate Gender Discrimination Cases and Defang the "Cluelessness" Defense, 7 Employee Rts. & Emp. Pol'y J. 401, 433-34 (2003). Because women with children are presumed to put their children as their first priority, when such a woman is late to work, her boss is likely to assume that the presumed innate characteristic, priority of childcare responsibilities, was the cause. Because men are assumed to put work first, a man who is late for work is assumed to have been caught in traffic, a transitional cause. Id.
n29. See Irene V. Blair, The Malleability of Automatic Stereotypes and Prejudice, 6 Personality & Soc. Psychol. Rev. 242, 244-47, 255-56 (2002); McGinley, supra note 14, at 430-32.
n30. See Marc R. Poirier, Is Cognitive Bias at Work a Dangerous Condition on Land?, 7 Employee Rts. & Emp. Pol'y J. 459, 478 (2003) (analogizing liability for discrimination caused by cognitive bias to the law related to invisible dangerous conditions on land); Michael Selmi, Discrimination as Accident: Old Whine, New Bottle, 74 Ind. L.J. 1233, 1233-34 (1999). But see Amy L. Wax, Discrimination as Accident, 74 Ind. L.J. 1129, 1131-34 (1999).
n31. See, e.g., Tristin K. Green, Discrimination in Workplace Dynamics: Toward a Structural Account of Disparate Treatment Theory, 38 Harv. C.R.-C.L. L. Rev. 91, 91-94 (2003); Krieger, supra note 7, at 1241-48; Charles R. Lawrence III, The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism, 39 Stan. L. Rev. 317 (1987); McGinley, supra note 14, at 430-32; Sturm, supra note 8, at 468; Williams, supra note 28, at 433-37.
n32. Before this time, states used the law to enforce inequality by prohibiting anti-slavery speech, prohibiting the education of Black people, restricting the power of manumission, prohibiting Black people from testifying in court, and prohibiting the presence of free Black people in slave-holding and free western states. The Reconstruction Amendments' Debates: The Legislative History and Contemporary Debates in Congress on the 13th, 14th, and 15th Amendments ii-iv (Alfred Avins ed., 1967) [hereinafter The Reconstruction Amendments' Debates]; Cong. Globe, 36th Cong., 2d Sess. 83-84 (1861), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 29; Cong. Globe, 36th Cong., 1st Sess. 1677-87, 1839 (1860), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 20-29; Cong. Globe, 35th Cong., 2d Sess. 948, 951-53, 970, 974-75, 980, 982-87, 1006 (1859), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 14-19; Cong. Globe, 35th Cong., 1st Sess. 402, 1967 (1858), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 13; Cong. Globe, 34th Cong., 3d Sess. 140 (1857), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 12; Cong. Globe, 34th Cong., 1st Sess. 124, 1598 (1856), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 12; Cong. Globe, 33d Cong., 1st Sess. 234, 240, 1012-13, 1155, 1556 (1854), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 11; Cong. Globe, 31st Cong., 1st Sess. 123-24, 288-89, 482, 1654-64, 1674-78, 2066 (1850), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 10; Cong. Globe, 30th Cong., 2d Sess. 418-19 (1849), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 1.
The federal government also promoted discrimination by, for example, not allowing Black people equal pay for military service or for carrying the mail. The Reconstruction Amendments' Debates, supra note 32, at iv-v. The United States Supreme Court reinforced these kinds of measures. See Dred Scott v. Sandford, 60 U.S. (19 How.) 393, 454 (1857) (holding that Black people were not citizens and therefore not entitled to the protection of the Privileges and Immunities Clause in Article IV, Section 2 of the United States Constitution); Moore v. Illinois, 55 U.S. (14 How.) 13, 16, 21 (1852) (upholding as constitutional a provision in the Illinois Constitution prohibiting the immigration of free Black people to the state).
n33. Massachussetts, for example, abolished segregation by race on streetcars and in schools. The Reconstruction Amendments' Debates, supra note 32, at 79.
n34. The Thirteenth Amendment provides: "Section 1. Neither slavery nor involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall exist within the United States, or any place subject to their jurisdiction. Section 2. Congress shall have power to enforce this article by appropriate legislation." U.S. Const. amend. XIII.
n35. The Fourteenth Amendment provides, in pertinent part:
Section 1. All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws ... .
Section 5. The Congress shall have power to enforce, by appropriate legislation, the provisions of this article.
U.S. Const. amend. XIV, 1, 5.
n36. The Fifteenth Amendment provides:
Section 1. The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.
Section 2. The Congress shall have power to enforce this article by appropriate legislation.
U.S. Const. amend. XV.
n37. Although these Reconstruction Amendments are viewed generally as instruments of racial equality, the abolitionist and women's rights movements overlapped significantly, and many hoped that women's rights would benefit. See, e.g., Cong. Globe, 40th Cong., 3d Sess. 708-10 (1869), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 346 (discussing an amendment proposed by Senator Pomeroy to grant suffrage to all citizens including women); Nell Irvin Painter, Sojourner Truth 220-33 (1996) (describing the overlap and the role Sojourner Truth played in both movements); Nadine Taub & Elizabeth M. Schneider, Perspectives on Women's Subordination and the Role of Law, in The Politics of Law, supra note 1, at 126. One reason for the overlap could be that distinctions based on sex were well accepted and were used to justify distinctions made on race and color. Milton R. Konvitz, A Century of Civil Rights 128 (1961). Another reason may be that women formed extensive religious and secular welfare associations because that was encouraged by the cult of domesticity dominant at the time. Taub & Schneider, supra note 37, at 162.
Although there was tension between members of both movements over the issue of which group deserved equality more, members of both movements argued that these amendments would grant equal rights for women, as well. Their hopes were dashed when the word "male" was used in Section 2 of the Fourteenth Amendment, providing that representation in Congress would be restricted for any state where the franchise was abridged for "any ... male inhabitants ... , being twenty-one years of age, and citizens of the United States ... ." U.S. Const. amend. XIV, 2. This marked the first time that gender was introduced into the Constitution. See H.R. Rep. No. 41-22 (1871) (reporting the majority view that the Fourteenth Amendment did not prohibit discrimination against women in exercising the right to vote, and the minority view that disagreed on the ground that voting was a privilege and immunity of citizenship), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 466-71; S. Rep. No. 42-21 (1872) (reporting the unanimous view of the Senate that the Fourteenth and Fifteenth Amendments did not give women the right to vote and that denial of women's-suffrage did not result in an anti-republican form of government), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 571-73.
The failure of the women's movement at that time helped to set in concrete the conceptual division between protected classes and the tension created when a person is a member of more than one class. For example, Black women are subject to discrimination because they are Black women, but under our legal categories, this is often found not to be discrimination because they are not discriminated against because of their sex by itself, and they are not discriminated against because of their race by itself. See Painter, supra note 37, at 224-25; see also Race-ing Justice, En-gender-ing Power xxx (Toni Morrison ed., 1992) (describing the interplay between race and sex in the controversy surrounding Anita Hill and the appointment of Clarence Thomas).
n38. In the Slaughter-House Cases, 83 U.S. (16 Wall.) 36 (1872), the Court held that the Fourteenth Amendment protected the rights that owed their existence to the federal government and not some broader definition of "privileges and immunities." Slaughter-House Cases, 83 U.S. at 37; see Pamela Brandwein, Reconstructing Reconstruction: The Supreme Court and the Production of Historical Truth 66-68 (1999). The Court also held that the Thirteenth Amendment was relevant only in cases of chattel slavery. Michael Vorenberg, Final Freedom: The Civil War, the Abolition of Slavery, and the Thirteenth Amendment 240 (2001).
Subsequent to The Slaughter-House Cases, Congress passed the Civil Rights Act of 1875, which prohibited racial discrimination in public accommodations. Act of Mar. 1, 1875, Ch. 114, 18 Stat. 335, 336. The Supreme Court found that neither the Thirteenth nor Fourteenth Amendments gave Congress the power to enact this law. The Civil Rights Cases, 109 U.S. 3, 11-12, 17-18, 22-23 (1883). The Court found that Congress lacked the power under the Thirteenth Amendment because the Thirteenth Amendment granted only the right to be free from the most literal forms of slavery. Id. at 22-23. Congress had a wide view of its power under the Thirteenth Amendment to erase the "badges and incidents" of slavery, and the Court's narrower view created lasting impediments to racial equality because slavery in the United States was based on a belief that Black people were inferior.
Slavery as an economic system ... was of small account compared with slavery as a system of racial adjustment and social control ... . Slavery was not the source of the philosophy [of the biological inequality and the racial inferiority of the Negro]. It merely enshrined it, prevented a practical demonstration of its falsity, and filled public offices and the councils of religious, educational, and political institutions with men reared in its atmosphere ... . The defense of slavery was of a social system and a system of racial adjustment, not of an economic institution.
Konvitz, supra note 37, at 10 (quoting Dwight L. Dumond, Antislavery Origins of the Civil War in the U.S. 52 (1939)) (alterations in original) (citations omitted).
The Fourteenth Amendment was also viewed narrowly by the Court in The Civil Rights Cases, which held that Congress had the power to restrain only state and not private actors. The Civil Rights Cases, 109 U.S. at 25. Scholars have criticized the Slaughter-House Cases and The Civil Rights Cases for having eviscerated the purpose of the amendments. See, e.g., Vorenberg, supra note 38, at 240-41; Christopher P. Banks, The Constitutional Politics of Interpreting Section 5 of the Fourteenth Amendment, 36 Akron L. Rev. 425, 438-39 (2003).
The Fifteenth Amendment has never been applied outside of the voting rights context. Congress, however, had a rather broad view of its power within this context. It enacted a law almost immediately after passage of the Fifteenth Amendment that, among other things, prohibited private parties from trying to interfere with anyone's exercise of the right to vote. The Enforcement Act of 1870, 16 Stat. 141, reprinted in 1 Statutory History of the United States: Civil Rights 445-53 (Bernard Schwartz ed., 1970). The proponent of that provision argued that Congress had the power to enact any legislation that would protect against the states' failure to prevent interference with the right to vote. Cong. Globe, 41st Cong., 2d Sess. 3611-13 (1870) (statement and amendment of Senator Pool), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 447-48. In other words, Congress had the power and the duty to enact positive protections to ensure that people could actually exercise the right to vote.
n39. The federal movement for racial equality was abandoned in 1877, and most of the Reconstruction legislation was repealed in 1894. Konvitz, supra note 37, at 66, 69. In addition, other than the Nineteenth Amendment, adopted in 1920, which provided that "the right of citizens of the United States to vote shall not be denied or abridged ... on account of sex[,]" further national efforts to address social inequality were generally unsuccessful. See, e.g., Plessy v. Ferguson, 163 U.S. 537 (1896); see also U.S. Const. amend. XIX. There were some states, however, that passed civil rights legislation. Konvitz, supra note 37, at 130.
n40. 347 U.S. 483 (1954).
n41. Id. at 493-95.
n42. Arguably, President Truman's Committee on Civil Rights, formed in late 1946, set the stage both for Brown and for subsequent legislation. Konvitz, supra note 37, at 70-72. In its report, the Committee stated and recommended:
The elimination of segregation, based on race, color, creed, or national origin, from American life.
The separate but equal doctrine has failed in three important respects. First, it is inconsistent with the fundamental equalitarianism of the American way of life in that it marks groups with the brand of inferior status. Secondly, where it has been followed, the results have been separate and unequal facilities for minority peoples. Finally, it has kept people apart despite incontrovertible evidence that an environment favorable to civil rights is fostered whenever groups are permitted to live and work together. There is no adequate defense of segregation.
Id. at 72 (quoting President's Committee on Civil Rights, To Secure These Rights (1947)). Additionally, President Eisenhower vowed to end segregation in the District of Columbia, and the Supreme Court considered segregation cases from the District before cases involving the same issues in the states. Id. at 123.
n43. Id. at 136-52; Charles & Barbara Whalen, The Longest Debate: A Legislative History of the 1964 Civil Rights Act (1985) (describing the interplay between the civil disobedience and the eventual passage of the Civil Rights Act of 1964).
n44. Civil Rights Act of 1957, Pub. L. No. 85-315, 71 Stat. 634. The bill passed after numerous compromises and after a one-person demonstration in the nature of a filibuster by Senator Strom Thurmond, which lasted over twenty-four hours. Konvitz, supra note 37, at 74-75.
Among other things, the Civil Rights Act of 1957 created a federal Commission on Civil Rights, which was to investigate allegations of discrimination in voting and other denials of equal protection, and which was to advise the federal government on equal protection issues. Civil Rights Act of 1957, 71 Stat. at 635. The Commission was to last only two years. Civil Rights Act of 1957, 71 Stat. at 635. The act focused primarily on enforcement of voting rights. Civil Rights Act of 1957, 71 Stat. at 637-38. It was very modest, and disappointed liberals as too little and Southerners as too much. Konvitz, supra note 37, at 78.
The Commission issued its report in 1959 and found widespread discrimination in voting, education, and housing. U.S. Comm'n on Civ. Rts., Report 545 (1959). The report also noted the complex interrelationships of these kinds of discrimination, and noted that Black Americans had become a sort of permanent "demoralized" underclass. Id. at 545-46, 548.
n45. This act was also a modest one, due in part to a filibuster and other delay tactics that lasted eight weeks. Konvitz, supra note 37, at 84-89.
n46. Civil Rights Act of 1964, Pub. L. No. 88-352, 78 Stat. 241 (codified as amended in scattered sections of the Code). Originally, Title VII of the Civil Rights Act of 1964, the most well known part of that legislation, was to prohibit discrimination on the basis of race, color, religion, or national origin. However, as a last-minute amendment by a Southern Democrat, proposed as a means to defeat the bill, sex was added to the list of prohibited classifications. 110 Cong. Rec. 2577-84 (1964).
n47. The women's movement used this strategy, beginning by trying to have the Supreme Court declare that women were entitled to civil rights under the Privileges and Immunities Clause of the Constitution. See, e.g., Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 133, 138 (1872). They were unsuccessful at first. The gay and lesbian civil rights movement also used the strategy, although much later, and it too was initially unsuccessful. Bowers v. Hardwick, 478 U.S. 186 (1986) (holding that there was no fundamental right to engage in same-sex sexual activity). The disability and age rights movements also followed this strategy. See, e.g., City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 433 (1985) (holding that rational basis review applied to classifications based on disability but striking down the decision at issue nonetheless); Massachusetts Bd. of Ret. v. Murgia, 427 U.S. 307 (1976) (concerning mandatory retirement at a particular age).
n48. For example, women won the right to vote in 1920 with the Nineteenth Amendment. U.S. Const. amend. XIX. The anti-age discrimination movement won passage of the Age Discrimination in Employment Act (ADEA) in 1971. 29 U.S.C. 621-634 (2000). The disability rights movement won passage of the Americans with Disabilities Act (ADA) in 1991. 42 U.S.C. 12101-12213 (2000). The gay and lesbian civil rights movement, on the other hand, went straight to the states and local communities after the defeat in Bowers. See Eric Marcus, Making History: The Struggle for Gay and Lesbian Civil Rights 406 (1992).
n49. After unsuccessful efforts at federal legislation, the Black civil rights movement focused its attention on the states and won passage of some civil rights laws. Konvitz, supra note 37, at 130. Similarly, after Bowers, the gay and lesbian civil rights movement focused on gaining rights in state and local communities. See Marcus, supra note 48, at 406.
n50. See, e.g., Lawrence v. Texas, 539 U.S. 558 (2003) (discussing sexual orientation); Reed v. Reed, 404 U.S. 71 (1971) (discussing gender); Brown, 347 U.S. 483 (discussing race).
n51. The Supreme Court has
"always been reluctant to expand the concept of substantive due process because guideposts for responsible decisionmaking in this unchartered area are scarce and open-ended." By extending constitutional protection to an asserted right or liberty interest, [the Court] to a great extent, places the matter outside the arena of public debate and legislative action.
Washington v. Glucksberg, 521 U.S. 702, 720 (1997) (quoting Collins v. Harker Heights, 503 U.S. 115, 125 (1992)).
n52. 42 U.S.C. 2000e to 2000e-16 (2000).
n53. 42 U.S.C. 2000e-2(a).
n54. See Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 985-86 (1988); Goodman v. Lukens Steel Co., 482 U.S. 656, 664 (1987); International Bhd. of Teamsters v. United States, 431 U.S. 324, 335 & nn. 14-15 (1977); Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971). However, the language of this provision would encompass a causal connection without intent. See Price Waterhouse v. Hopkins, 490 U.S. 228, 240-41 (1989) (discussing the provision in causal terms).
n55. See Watson, 487 U.S. at 986-87; Griggs, 401 U.S. at 431.
n56. Watson, 487 U.S. at 987; see also Griggs, 401 U.S. at 431.
n57. Griggs, 401 U.S. at 431-34 (recognizing that Congress intended to prohibit employment decisions with discriminatory consequences, not merely those with discriminatory motivation).
n58. Demonstrating its discomfort with a broad application of disparate impact, the Court stated in Watson:
The plaintiff's burden in establishing a prima facie case goes beyond the need to show that there are statistical disparities in the employer's work force. The plaintiff must begin by identifying the specific employment practice that is challenged ... . Especially in cases where an employer combines subjective criteria with the use of more rigid standardized rules or tests, the plaintiff is in our view responsible for isolating and identifying the specific employment practices that are allegedly responsible for any observed statistical disparities.
Watson, 487 U.S. at 994; see also Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 656-57 (1989).
In Wards Cove, the Court held that if a plaintiff could demonstrate that a particular employment practice had produced a disparity, the employer must then produce evidence that would show that it had a business justification for the practice. Wards Cove, 490 U.S. at 659. The burden of persuasion, however, remained with the plaintiff. Id. Thus, a plaintiff could prevail by demonstrating that the practice did not have a business justification. Id. at 660. Alternatively, a plaintiff could show that the same goals could be reached without an undesirable effect. Id. at 660-61. This rule brought disparate impact analysis almost entirely within disparate treatment analysis, with the shifting burden of production and the opportunity to prove pretext alone, which is what the Court intended. See id. at 660.
The Court's holding in Wards Cove was superseded by the Civil Rights Act of 1991, 42 U.S.C. 2000e-2, which provides:
(k)(1)(A) An unlawful employment practice based on disparate impact is established under this title only if -
(i) a complaining party demonstrates that a respondent uses a particular employment practice that causes a disparate impact on the basis of race, color, religion, sex, or national origin and the respondent fails to demonstrate that the challenged practice is job related for the position in question and consistent with business necessity; or
(ii) the complaining party makes the demonstration described in subparagraph (C) with respect to an alternative employment practice and the respondent refuses to adopt such alternative employment practice.
(B)(i) With respect to demonstrating that a particular employment practice causes a disparate impact as described in subparagraph (A)(i), the complaining party shall demonstrate that each particular challenged employment practice causes a disparate impact, except that if the complaining party can demonstrate to the court that the elements of a respondent's decisionmaking process are not capable of separation for analysis, the decisionmaking process may be analyzed as one employment practice.
(ii) If the respondent demonstrates that a specific employment practice does not cause the disparate impact, the respondent shall not be required to demonstrate that such practice is required by business necessity.
(C) The demonstration referred to by subparagraph (A)(ii) shall be in accordance with the law as it existed on June 4, 1989, with respect to the concept of "alternative employment practice."
(2) A demonstration that an employment practice is required by business necessity may not be used as a defense against a claim of intentional discrimination under this title.
(3) Notwithstanding any other provision of this title, a rule barring the employment of an individual who currently and knowingly uses or possesses a controlled substance, as defined in schedules I and II of section 102(6) of the Controlled Substances Act (21 U.S.C. 802(6)), other than the use or possession of a drug taken under the supervision of a licensed health care professional, or any other use or possession authorized by the Controlled Substances Act or any other provision of federal law, shall be considered an unlawful employment practice under this title only if such rule is adopted or applied with an intent to discriminate because of race, color, religion, sex, or national origin.
Civil Rights Act of 1991, 42 U.S.C. 2000c-2.
Despite restoration of the burden of proof, some courts have resisted using the theory, and plaintiffs usually try to style their cases as disparate treatment cases. See Elaine W. Shoben, Disparate Impact Theory in Employment Discrimination: What's Griggs Still Good for? What Not?, 42 Brandeis L.J. 597, 598-600, 620-21 (2004) (recognizing that lower courts are somewhat hostile to disparate impact claims and that plaintiffs rarely bring them, but not finding a causal connection).
n59. While the Court was not explicit that the Equal Protection Clause did not prohibit laws or policies with a disparate impact until 1976, well after the Civil Rights Act of 1964 was passed, it had never previously found that the Equal Protection Clause prohibited laws or actions with only discriminatory effects. Washington, 426 U.S. at 245-48. The Court did hold that there could be circumstances in which this discriminatory purpose can be inferred from disparate application or an extremely one-sided discriminatory effect. See Akins, 325 U.S. at 404; Yick Wo, 118 U.S. at 374. However, that discriminatory effect must be very stark to comprise evidence of discriminatory purpose. See Village of Arlington Heights, 429 U.S. at 266; see also Yick Wo, 118 U.S. at 358-59, 373-74.
n60. Some scholars have argued that disparate impact should be used more extensively, since it is a significantly more flexible doctrine. See, e.g., Shoben, supra note 60. However, other scholars suggest that courts are extremely wary of disparate impact because of its flexibility and, as a result, restrict its application severely. See, e.g., Malamud, supra note 4, at 2263-66. Given the Court's decision in Wards Cove, which essentially pulled the disparate impact standard into a replica of the disparate treatment standard, I am inclined to side with the latter view.
n61. 411 U.S. 792 (1973).
n62. Id. at 802. McDonnell Douglas concerned a motion to dismiss, but the rule was set up as a method "to govern the consideration of [the plaintiff's] claim." Id. at 798. The Court later made clear that this test did not apply at the point in trial where the factfinder must make the ultimate decision. United States Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 714-15 (1983). At trial, the straightforward issue is whether the employer discriminated. Id. at 715. Moreover, despite the procedural posture of McDonnell Douglas, the Court recently made clear that the test is not used to evaluate the sufficiency of a complaint. Swierkiewicz v. Sorema, 534 U.S. 506, 511 (2002).
n63. See Texas Dep't of Cmty. Affairs v. Burdine, 450 U.S. 248, 252-53 (1981) (building on McDonnell Douglas). The elements of the prima facie case laid out by the Court in McDonnell Douglas, a failure to hire, race discrimination case, were: 1) the plaintiff was a member of a racial minority; 2) the plaintiff was qualified for the job; 3) the plaintiff was rejected by the defendant despite his qualifications; and 4) after the rejection, the position remained open, and the defendant continued to seek applicants with the plaintiff's qualifications. McDonnell Douglas, 411 U.S. at 802. The Court added that these elements were flexible and would vary in different factual situations. Id. at 802 n.13. The generic prima facie case could be styled as: 1) the plaintiff is a member of a class protected by statute; 2) the plaintiff applied and was qualified for the job or performing adequately; 3) the employer took some kind of adverse employment action against the plaintiff; and 4) there is some information that suggests the reason for the decision was not related to how the business should operate or on economics alone.
n64. Burdine, 450 U.S. at 252-53. The employer has only a burden to produce admissible evidence. Id. at 255. Although the evidentiary burden shifts, "the ultimate burden of [persuasion] remains with the plaintiff at all times, to convince the factfinder of intentional discrimination on the part of the defendant." Id. at 253, 256.
n65. Id. at 253; McDonnell Douglas, 411 U.S. at 804.
n66. See Desert Palace, Inc. v. Costa, 539 U.S. 90, 100-02 (2003). Before Desert Palace, it was unclear whether, in a circumstantial case, discrimination had to be the sole reason for the discharge, or merely a motivating reason. See id. at 95. After Desert Palace, it is clear that whatever type of evidence proves liability, the discriminatory reason need only be a motivating reason and not the sole reason. Id. at 101-02.
n67. Id. at 95, 100-02 (citing 42 U.S.C. 2000e-2(m)).
n68. Burdine, 450 U.S. at 254.
n69. The courts usually define pretext as a lie. See, e.g., Russell v. Acme-Evans Co., 51 F.3d 64, 68 (7th Cir. 1995); Shager v. Upjohn Co., 913 F.2d 398, 401 (7th Cir. 1990); see also Black's Law Dictionary 602, 1187 (6th ed. 1990). However the Supreme Court has resisted using this language, and pretext could easily include a facially neutral reason based on a prohibited assumption or stereotype. For example, personal animosity would be a non-discriminatory reason for an employment action but not if that personal animosity were really sublimated racism or sexism. Compare Staggs v. Elk Run Coal Co., 479 S.E.2d 561, 581-84 (W. Va. 1996) (recognizing that discrimination is not always conscious), with State Dep't of Pub. Safety v. Sexton, 748 So. 2d 200, 213-14 (Ala. Civ. App. 1998) (holding that there is no pretext unless there is a lie); see generally Price Waterhouse, 490 U.S. 228 (recognizing that discrimination is illegal whether the animus is conscious or sublimated into something like stereotyping); Watson, 487 U.S. at 990-91 (recognizing that subjective decisionmaking can mask reliance on stereotypes and, thus, discrimination).
n70. See Dister v. Continental Group, Inc., 859 F.2d 1108, 1113 (2d Cir. 1988); MacDissi v. Valmont Indus., 856 F.2d 1054, 1059 (8th Cir. 1988); Chipollini v. Spencer Gifts, Inc., 814 F.2d 893, 899 (3d Cir. 1987) (en banc); Tye v. Polaris Joint Voc. Sch. Dist. Bd. of Educ., 811 F.2d 315, 319-20 (6th Cir. 1987); Thornbrough v. Columbus & Greenville R.R., 760 F.2d 633, 647 (5th Cir. 1985); see also Catherine J. Lanctot, The Defendant Lies and the Plaintiff Loses: The Fallacy of the "Pretext-Plus" Rule in Employment Discrimination Cases, 43 Hastings L.J. 57, 71-75 (1991) (discussing in detail the circuits that had clearly adopted the rule, and those that had decisions reflecting both the pretext only and pretext plus rules).
n71. See White v. Vathally, 732 F.2d 1037, 1042-43 (1st Cir. 1984); Lanctot, supra note 70, at 82-86 (discussing the circuits that had some decisions appearing to advocate the pretext plus rule).
n72. 509 U.S. 502 (1993).
n73. Id. at 511.
n74. See generally id. Four circuits held that courts could make no inference from the fact that the defendant's reason was not the true reason. See Reeves v. Sanderson Plumbing Prods., Inc., 197 F.3d 688, 693-94 (5th Cir. 1999), rev'd, 530 U.S. 133 (2000); Thomas v. Eastman Kodak Co., 183 F.3d 38, 56-57, 64 (1st Cir. 1999); Gillins v. Berkeley Elec. Coop., Inc., 148 F.3d 413, 416-17 (4th Cir. 1998); Fisher v. Vassar Coll., 114 F.3d 1332, 1344-46 (2d Cir. 1997) (en banc).
n75. 530 U.S. 133.
n76. An employer might offer a false explanation for reasons other than to cover up discrimination. For example, the employer might be embarrassed about the real reason. Still, it is reasonable to allow an inference of bad intent from the fact of the lack of candor in this context just as it would be in any other evidentiary context. See Reeves, 530 U.S. at 147.
n77. Id. at 142-43, 147.
n78. Id. at 143 (citing Burdine, 450 U.S. at 256).
n79. See Raytheon Co. v. Hernandez, 540 U.S. 44, 51-53 (2003) (noting the test's application to the ADA); O'Connor v. Consolidated Coin Caterers Corp., 517 U.S. 308, 311 (1996) (assuming without holding that the McDonnell Douglas test applies to ADEA claims); Hicks, 509 U.S. at 506 n.1 (assuming that the test applies to cases brought under 42 U.S.C. 1983).
n80. See, e.g., Batson v. Kentucky, 476 U.S. 79 (1986) (applying the test to challenges that jurors were struck on the grounds of race).
n81. See infra note 131.
n82. Furnco Constr. Corp. v. Waters, 438 U.S. 567, 577 (1978).
n83. Aikens, 460 U.S. at 716.
n84. See Furnco, 438 U.S. at 577. In making this statement, the Court asserted the presumption that employers acted rationally. Id. Many scholars have taken this to mean that the Court was assuming that employers exercise sound judgment and have demonstrated how irrationally some employers can act. See, e.g., Malamud, supra note 4, at 2255-57. However, this cannot be what the Court meant by rational. Since it required the employer to merely articulate a reason that was not based on a prohibited characteristic, it must have meant rational in a much narrower sense, to mean only that employers must be presumed to know why they terminated an employee enough to articulate that reason. See Selmi, supra note 11, at 326.
n85. Aikens, 460 U.S. at 716.
n89. Deborah Malamud contends that the Court created the standard to protect employers from a more reaching standard like that used in the disparate impact analysis. Malamud, supra note 4, at 2237.
n90. See Michael Selmi, Why Are Employment Discrimination Cases So Hard to Win?, 61 La. L. Rev. 555, 558 (2001).
n91. See Kevin M. Clermont et al., How Employment-Discrimination Plaintiffs Fare in the Federal Courts of Appeals, 7 Employee Rts. & Emp. Pol'y J. 547, 556-58, 566 (2003).
n92. See, e.g., Selmi, supra note 11, at 283-84, 309.
n93. See generally Federal Glass Ceiling Commission, Good for Business: Making Full Use of the Nation's Human Capital 75 (1995); Melvin L. Oliver & Thomas M. Shapiro, Black Wealth/White Wealth: A New Perspective on Racial Inequality 45-50 (1995); Donald Tomaskovic-devey, Gender & Racial Inequality at Work: The Sources and Consequences of Job Segregation (1993).
n94. See Deborah A. Calloway, St. Mary's Honor Center v. Hicks: Questioning the Basic Assumption, 26 Conn. L. Rev. 997, 998 (1994) (suggesting that the Court does not believe that discrimination really still exists); Selmi, supra note 11, at 284. Amy Wax, for example, has argued that if discrimination operates beyond normal self-awareness, it is not properly penalized. See generally Wax, supra note 30.
n95. See generally Cong. Globe, 43d Cong., 1st Sess. 344, 376-77, 379, 381-84, 565-66, 616, 618, 726, 741, 900, 902 (1873-74), reprinted in The Reconstruction Amendments' Debates, supra note 32, at 658-60, 670-71.
n96. See supra note 95.
n97. See, e.g., Mary Becker et al., Cases and Materials on Feminist Jurisprudence: Taking Women Seriously 642 (2d ed. 2001); Vicki Lens, Supreme Court Narratives on Equality and Gender Discrimination in Employment: 1971-2002, 10 Cardozo Women's L.J. 501, 520 (2004); Susan Moller Okin, Justice and Gender: An Unfinished Debate, 72 Fordham L. Rev. 1537, 1551-52 (2004).
n98. See generally Ruth Gordon, Heroes of Their Own Lives: The Politics and History of Family Violence 1-6 (1988); Susan Schechter, Women and Male Violence: The Visions and Struggles of the Battered Women's Movement 3-5 (1982).
n99. See Okin, supra note 97, at 1551-52.
n100. See Catharine A. MacKinnon, Sexual Harassment of Working Women 1-3 (1979) (discussing the issues presented by sexual harassment in the workplace).
n101. See Lawrence, 539 U.S. at 563-64.
n102. See Malamud, supra note 4, at 2237-38, 2319-20 (arguing that abandoning the McDonnell Douglas test would lead to a more holistic view of cases and creative understanding of discrimination in the workplace). But see Troupe v. May Dep't Stores Co., 20 F.3d 734, 736 (7th Cir. 1994) (purporting not to use the burden shifting method, but limiting the types of circumstantial evidence, which might possibly be used to prove discrimination, to those types comparable to the fourth prong of the prima facie case or pretext).
n103. 509 U.S. 502 (1993).
n104. 530 U.S. 133 (2000) [hereinafter Reeves II].
n105. Hicks, 509 U.S. at 510-11.
n106. See Reeves v. Sanderson Plumbing Prods., Inc., 197 F.3d 688, 693-94 (5th Cir. 1999) [hereinafter Reeves I], rev'd, 530 U.S. 133; Thomas v. Eastman Kodak Co., 183 F.3d 38, 56-57, 64-65 (1st Cir. 1999); Gillins v. Berkeley Elec. Coop., Inc., 148 F.3d 413, 416-17 (4th Cir. 1998); Fisher v. Vassar Coll., 114 F.3d 1332, 1344-46 (2d Cir. 1997) (en banc).
n107. Reeves II, 530 U.S. at 137-38.
n108. Reeves was replaced by a succession of three people in their thirties. Id. at 142.
n109. Id. at 151.
n110. Id. at 139.
n111. Reeves I, 197 F.3d at 694.
n112. Id. at 693.
n113. Id. at 693-94.
n114. Reeves II, 530 U.S. at 142, 146.
n115. Id. at 146.
n116. See Michael J. Zimmer, Slicing & Dicing of Individual Disparate Treatment Law, 61 La. L. Rev. 577, 592-99 (2001). Zimmer notes that the Court in Reeves failed to address the process that the Fifth Circuit used in categorizing the evidence, and this failure means that lower courts continue to get caught up in the taxonomy. Id. at 591-99.
n117. See Chin & Golinsky, supra note 4 (arguing that McDonnell Douglas should be discarded because it is an empty ritual that courts no longer employ).
n118. See Zimmer, supra note 116, at 600.
n119. See, e.g., Hudson v. Chicago Transit Auth., 375 F.3d 552 (7th Cir. 2004); Steinhauer v. DeGolier, 359 F.3d 481 (7th Cir. 2004). In fact, in a case in which the defendant conceded that a prima facie case had been made, the Seventh Circuit nevertheless found that the plaintiff failed to provide sufficient evidence of one. Davis v. Con-Way Transp. Cent. Express, Inc., 368 F.3d 776 (7th Cir. 2004).
n120. In fact, some courts even require evidence that a comparator was treated more favorably as an element of the prima facie case. See, e.g., Nieto v. L&H Packing Co., 108 F.3d 621, 624 n.7 (5th Cir. 1997) (recognizing a split within the Fifth Circuit concerning whether it was an element of the prima facie case but not requiring it); Suggs v. ServiceMaster Educ. Food Mgmt., 72 F.3d 1228, 1232 (6th Cir. 1996); Edwards v. Wallace Cmty. Coll., 49 F.3d 1517, 1521 (11th Cir. 1995); Simens v. Reno, 960 F. Supp. 6, 10 (D.D.C. 1997); see also Marla Swartz, Note, The Replacement Dilemma: An Argument for Eliminating a Non-Class Replacement Requirement in the Prima Facie Stage of Title VII Individual Disparate Treatment Discrimination Claims, 101 Mich. L. Rev. 1338 (2003) (discussing how various courts have used and required evidence of comparators).
n121. 490 U.S. 228 (1989).
n122. See id. at 233, 251-52.
n123. See id. at 236.
n124. See id. at 251-52.
n125. 42 U.S.C. 2000e-3(a) (2000). Specifically, it prohibits employers from "discriminating against any ... employees or applicants for employment ... because [they have] opposed any practice made an unlawful employment practice by this subchapter, or because [they have] made a charge, testified, assisted, or participated in any manner in an investigation, proceeding, or hearing under this subchapter." 42 U.S.C. 2000e-3(a).
n126. See Mark S. Brodin, The Demise of Circumstantial Proof in Employment Discrimination Litigation: St. Mary's Honor Center v. Hicks, Pretext, and the "Personality" Excuse, 18 Berkeley J. Emp. & Lab. L. 183, 188-89 (1997); Stephen W. Smith, Title VII's National Anthem: Is There a Prima Facie Case for the Prima Facie Case?, 12 Lab. Law. 371, 381 (1997).
n127. See McDonnell Douglas, 411 U.S. at 802 (requiring the plaintiff to show that he was not hired in a failure to hire case).
n128. A few courts have required that plaintiffs demonstrate that they suffered an "ultimate" job action, such as termination. See, e.g., Hernandez v. Crawford Bldg. Material Co., 321 F.3d 528, 531-32 (5th Cir. 2003); Ledergerber v. Stangler, 122 F.3d 1142, 1144 (8th Cir. 1997). Others require that the action be "materially adverse," although it need not be an ultimate action. Bell v. E.P.A., 232 F.3d 546, 555 (7th Cir. 2000); Nguyen v. City of Cleveland, 229 F.3d 559, 566 (6th Cir. 2000); Heno v. Sprint/United Mgmt. Co., 208 F.3d 847, 857 (10th Cir. 2000); Munday v. Waste Mgmt. of N. Am., Inc., 126 F.3d 239 (4th Cir. 1997); Robinson v. City of Pittsburgh, 120 F.3d 1286, 1300 (3d Cir. 1997); Torres v. Pisano, 116 F.3d 625, 640 (2d Cir. 1997). The Ninth Circuit and the EEOC have adopted the view that any treatment reasonably likely to deter protected activity constitutes retaliation. See Ray v. Henderson, 217 F.3d 1234, 1243 (9th Cir. 2000); EEOC, 2 EEOC Compliance Manual 8 (2002).
n129. The approach of the Ninth Circuit and the EEOC comes closest to removing the requirement altogether.
n130. 42 U.S.C. 2000e-2(a) (2000).
n131. See 42 U.S.C. 2000e-3(a).
n132. 29 U.S.C. 2612(a) (2000).
n133. 29 U.S.C. 2615(a)(1) (2000).
n134. See Smith v. Diffee Ford-Lincoln-Mercury, Inc., 298 F.3d 955, 960-61 (10th Cir. 2002); Bachelder v. America W. Airlines, Inc., 259 F.3d 1112, 1130-31 (9th Cir. 2001); see generally Martin H. Malin, Interference with the Right to Leave Under the Family and Medical Leave Act, 7 Employee Rts. & Emp. Pol'y J. 329 (2003) (arguing that the FMLA should be interpreted in the same way as the National Labor Relations Act, which prohibits interference with the right to organize).
n135. See Hunt v. Rapides Healthcare Sys., LLC, 277 F.3d 757, 768 (5th Cir. 2002); Nichols v. Ashland Hosp. Corp., 251 F.3d 496, 502 (4th Cir. 2001); Brungart v. Bellsouth Telecomm., Inc., 231 F.3d 791, 798 (11th Cir. 2000); Rice v. Sunrise Express, Inc., 209 F.3d 1008, 1019 (7th Cir. 2000); Rice, 209 F.3d at 1019-20 (Evans, J., dissenting); Gleklen v. Democratic Cong. Campaign Comm., 199 F.3d 1365 (D.C. Cir. 2000); Hodgens v. General Dynamics Corp., 144 F.3d 151, 160-72 (1st Cir. 1998); Morgan v. Hilti, Inc., 108 F.3d 1319, 1323 (10th Cir. 1997).
n136. 540 U.S. 581 (2004).
n137. See id. at 581.
n138. 29 U.S.C. 623(a)(1) (2000).
n139. 29 U.S.C. 631(a) (2000).
n140. See Cline, 540 U.S. at 602-05 (Thomas, J., dissenting); Cline v. General Dynamics Land Sys., Inc., 296 F.3d 466, 469-71 (6th Cir. 2002); Cline, 296 F.3d at 472-75 (Cole, J., concurring). The primary meaning of the word "age" is chronological age - the length of time that a person has been alive. American Heritage Dictionary 33 (3d ed. 1992).
n141. 42 U.S.C. 2000e-2(a).
n142. Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75, 79 (1998); McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273 (1976).
n143. Cline, 540 U.S. at 584.
n145. This analysis was tortured because the Court failed to follow the rules of statutory construction, which it has been adhering to very closely for at least the past few years. See Rafael Gely, Supreme Court's 2002 Term Employment Law Cases: Is This the Scalia Court?, 7 Employee Rts. & Emp. Pol'y J. 253, 254 (2003) (describing the Court's recent adherence to the primary rule of statutory construction). It began by looking to the legislative history, gave no weight to an EEOC regulation, and allowed the word "age" to mean different things in different parts of the ADEA. Cline, 540 U.S. at 585-99.
n146. Cline, 540 U.S. at 590-91.
n147. Price Waterhouse, 490 U.S. at 251-52.
n148. Moreover, the reasoning cannot be explained along affirmative action lines. There is no good analogy to affirmative action. Race-based decisions are valid under Title VII as long as they are made pursuant to a valid affirmative action plan. See generally United Steelworkers of Am. v. Weber, 433 U.S. 193 (1979). However, affirmative action involves distinctions between classes, both of which might be protected, but not distinctions within classes. This view that classes are truly distinct, and that they do not involve a continuum may be inaccurate, but that is the view embodied in the law.
n149. Cline, 540 U.S. at 592-93.
n150. 517 U.S. 308 (1996).
n151. Id. at 309-10.
n153. Id. at 311-12.
n154. Id. at 312 (emphasis in original).
n155. Members of juries, as well, would be just as prone to cognitive bias as anyone, but I do not address them here for several reasons. First, juries are made up of several people who must make decisions by consensus. The more balanced the jury in terms of class, race, gender, etc., the more likely that the biases of one group would be negated by the understandings of other jurors. Second, judges are not employees in the normal sense even if they are state judges who are elected. No one with authority can direct them in doing their jobs, with the limited exception of a higher court overturning a decision on appeal or directing a non-discretionary act through a writ of mandamus. Yet, nearly all judges are employers, or at least must manage their law clerks, secretaries, and other assistants. Thus, judges are likely to identify with employers as fellow members of an in-group. The vast majority of jurors are more likely to be employees and not employers or managers, or at least employees in addition to being managers. Thus, jurors are less likely to identify with the employer. Third, judges are left wholly to their own discretion in deciding how the law applies to the facts. Jurors are given instructions, which can be designed to guide that application in a way that would avoid the effects of cognitive bias.
n156. Chris Guthrie et al., Inside the Judicial Mind, 86 Cornell L. Rev. 777, 829-30 (2001).
n157. See John Braithwaite, Rules and Principles: A Theory of Legal Certainty, 27 Australian J. Legal Phil. 47 (2002).
n158. 490 U.S. 228 (1989).
n159. Id. at 235-36, 251-52.
n160. See Arthur R. Miller, The Pretrial Rush to Judgment: Are the "Litigation Explosion," "Liability Crisis," and Efficiency Cliche's Eroding Our Day in Court and Jury Trial Commitments?, 78 N.Y.U. L. Rev. 982 (2003) (arguing that courts are relying on summary judgment to resolve factual issues); Paul W. Mollica, Federal Summary Judgment at High Tide, 84 Marq. L. Rev. 141 (2000) (making the same argument).
n161. See Mollica, supra note 160, at 152 (quoting Benton-Volvo-Metairie, Inc. v. Volvo Southwest, Inc., 479 F.2d 135, 138 (5th Cir. 1973)).
n162. See Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 142-43 (2000); St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 510 (1993).
n163. Desert Palace, Inc. v. Costa, 539 U.S. 90, 101-02 (2003).