The Irrational Turn in Employment Discrimination Law: Slouching Toward a Unified Approach to Civil Rights Law - John Vallery White

• Publication year: 2002

The Irrational Turn in Employment Discrimination Law: Slouching Toward a Unified Approach to Civil Rights Lawby John Valery White^*

I. Introduction

This Article argues that the Supreme Court's recent disparate treatment decisions under Title VII of the Civil Rights Act of 1964¹ represent a trend toward unifying all civil rights law under an approach most closely akin to traditional equity. This trend explains the curious tension between substance and process in the Court's most recent decisions, St. Mary's Honor Center v. Hicks² and Reeves v. Sanderson Plumbing.³ It also explains the Court's uncommon confidence in its yet undefined notions of what constitutes discrimination on the basis of the several protected categories recognized in Title VII and related statutes. The trend toward equity reveals that the Court regards any precise definition of discrimination "because of the categories" as unnecessary, as it views the law of employment discrimination as applicable only to egregious cases. The perils of this approach are outlined, its necessity rebutted, and an alternative offered in succeeding sections.

Once regarded as a simple, even intuitive method of proving discrimination, the Title VII disparate treatment proof structure is now seen as wildly complicated. To commentators who hoped the Supreme Court would simplify the proof structure, the recent decision in Reeves was as disappointing as it was anticipated. Revisiting its 1993 decision in Hicks, the Court seemed poised to abandon one of two somewhat contradictory components of that decision. Instead, the Court in Reeves strongly reaffirmed Hicks on grounds that sustained the tensions inherent in the Hicks decision.⁴

Hicks and Reeves together represent a consistent, though seemingly contradictory approach to understanding proof of discrimination. Both opinions emphasize that in employment discrimination cases, a plaintiff is required to prove that the contested employment decision was motivated by a discriminatory reason.⁵ In this sense, both opinions emphasize a "substantive" approach to denning the parties' obligations. Notwithstanding intermediary burdens placed on the defendant, the plaintiff bears the burden of proof throughout and discriminatory intent is what the plaintiff must show to prevail. The opinions also embrace a "procedural" approach to structuring disparate treatment proofs. This approach is most closely associated with McDonnell Douglas v. Green,⁶ the Supreme Court's first disparate treatment decision. In McDonnell Douglas, the procedural approach used emphasized that a party's failure to produce evidence at various "stages of proof might result in a decision against that party.⁷ Although this approach's roots in common law suggest that it is consistent with a substantive definition of the elements of a discrimination cause of action, the McDonnell Douglas structure is widely regarded as artifice, largely disconnected from the question of discrimination vel non.⁸

By emphasizing the substantive requirements of the cause of action, Hicks highlighted the duality of the disparate treatment proof. To the detriment of plaintiffs whose obligations were seen as heightened, Hicks was consequently viewed as a change in the law.⁹ The inertia of Hicks precipitated a shift in lower courts' view of discrimination claims, supporting the view that Hicks had circumscribed the cause of action.¹⁰ In the heated debate that Hicks generated, most commentators agreed that the artificial nature of the McDonnell Douglas proof structure, especially its very light prima facie case, was the root of the problem.¹¹ Though the debate raged over whether Hicks was an appropriate response to the problem it exposed, fewer commentators defended McDonnell Douglas.¹² Indeed, even among commentators who saw value in McDonnell Douglas, most focused on the need to preserve its "presumption" of discrimination by attaching significance to the third stage of proof, the showing of pretext.¹³ Few argued that the indirect, three-step proof was valuable, as such.

In this context, Reeves is potentially confusing. Contrary to the expectations of many, Reeves reaffirmed the three-step approach of

McDonnell Douglas.¹⁴ Reeves also chastened the Fifth Circuit Court of Appeals, restoring a plaintiff's claims and seeming to break with the conservative spirit of Hicks.¹⁵ As such, Reeves was viewed as a pro-plaintiff's decision by the journalists covering the Supreme Court.¹⁶ However, Reeves repeats the very language from Hicks which emphasized that, artifice notwithstanding, the core of the disparate treatment proof is the plaintiff's obligation to prove discriminatory intent.¹⁷ The core confusion, common to both Hicks and Reeves, is the retention of both substantive and procedural requirements in the disparate treatment proof.

This Article will first locate the simultaneous commitment to both the substantive and procedural requirements of disparate treatment proof in the Court's quiet unification of civil rights law. Though the tension between substantive and procedural proofs is real, this tension is not as important as the Court's reasons for ignoring it. Specifically, this Article will show that the Court's approach in Hicks and Reeves articulates an incomprehensible definition of discrimination, wrapped in a veil of coherence provided by the Court's ambivalent commitment to both substantive and procedural approaches. This ruse is maintained because the Court wishes to reconcile the more powerful antidiscrimination principles of Title VII with the remainder of its civil rights jurisprudence; Hicks and Reeves represent the transformation of antidiscrimination law into a form of old-fashioned equity. Under this view, discrimination cannot be proved by establishing elements, but rather is reserved to the considered judgment of the judge.

This is not to say that the Court should resolve the tension between substance and procedure. This Article instead seeks to demonstrate why the Court's recent emphasis on substance, an emphasis that has encouraged many to attack McDonnell Douglas, must be viewed with caution. Both Hicks and Reeves are wrong in over-emphasizing plaintiffs' need to prove discrimination vel non for the simple reason that the requirements of such a proof cannot be defined.¹⁸ In emphasizing this requirement, the Court's recent decisions ask plaintiffs to guess what a judge's definition of discrimination might be, provide evidence sufficient to satisfy the judge's view of such actions, and anticipate any excuses the judge may harbor for discriminatory behavior. Neither plaintiffs nor defendants can know what they must show or defend; rather they are left to trade in the very stereotypes and prejudices that the act presumably bans from the workplace in the hope that they can trigger the judge's sympathy to their cause.¹⁹ In showing why discrimi- nation cannot be sufficiently denned, this Article highlights the importance of the McDonnell Douglas three-step proof structure.

Recognizing that the concerns voiced in Hicks about false positives are significant and accepting that the promise of defining "elements" of a disparate treatment cause is illusory, this Article proposes a modified McDonnell Douglas proof. Contrary to the language of Hicks, a procedural approach to proving discrimination is not foreclosed by Rule 301 of the Federal Rules of Evidence, nor by common-law understandings of burdens of proof and production. Rather the substantial policy concerns that make antidiscrimination law controversial counsel for the very kind of complicated balance the Court attempts in Hicks and Reeves.

II. The Propriety of McDonnell Douglas and the Remedy of Hicks

Hicks and Reeves prove to be very confusing decisions because their balancing of the substantive and procedural aspects of disparate treatment veil a policy decision, the significance of which at all times exceeds the debate over substantive and procedural definitions of discrimination. The policy choice is to move employment discrimination law from the "law" to the "equity" side of the bar. Perhaps rooted in the belief that discrimination is relatively rare, this move deliberately remolds employment discrimination law to be more consistent with the Court's recent decisions on civil rights generally. It is problematic because it overlooks the inherent problems in defining discrimination.

Hicks and Reeves are the Supreme Court's most substantial modifications of disparate treatment since the approach was adopted in 1973. McDonnell Douglas first distinguished disparate treatment cases from disparate impact suits, holding that a plaintiff who showed a "prima facie case" of discrimination was entitled to a presumption of discrimination unless the defendant "rebutted" that presumption with admissible evidence of a nondiscriminatory reason for the adverse employment decision.²⁰ The McDonnell Douglas opinion held, however, that the rebuttal did not end the matter; a plaintiff was to be given an opportuni- ty to show that the rebuttal was pretext, allowing the case to go to the fact-finder on the ultimate question of discriminatory motive.²¹

Over the next twenty years two questions proved crucial: courts were forced to define the nature of both the plaintiff's and defendant's showings at the initial stages of the process. In Postal Service Board of Governors v. Aikens,²² the Supreme Court avoided defining precisely what the scope of the plaintiff's prima facie case was, but implied that the scope of the showing was both light and flexible.²³ Addressing the defendant's burden in Burdine v. Texas Department of Community Affairs²⁴ the Supreme Court held that the defendant's burden was only one of production, making that burden rather light as well.²⁵ Predictably, the key question quickly came to concern the effect of the pretext showing. With the 1991 Civil Rights Act's recognition of compensatory and punitive damages, the question...