It is no secret that the Supreme Court's Title VII(1) jurisprudence cloaks substance in the "curious garb" of procedure.(2) When the Supreme Court talks about employment discrimination under Title VII, it generally does so by creating and refining special proof structures--different methods of proving discrimination.(3) This emphasis on procedure comes at the expense of discussions of what one naively might call "substance." Indeed, with the exception of some prominent sex discrimination cases,(4) the Supreme Court has taught us little in the past twenty-five years(5) about what discrimination is, how pervasive it is, and how we are to recognize it in the world. Refining the special proof structures thus constitutes the Supreme Court's major mode of discourse on the subject of employment discrimination.(6)
The liberal legal community(7) has sought--and claims to have found--a set of substantive judgments embedded in the Court's procedural decisions. What is said to exist is a substantive consensus that the eradication of discrimination is a high societal priority, and that discrimination is pervasive but difficult to prove. This consensus claims that the Court fashioned Title VII procedure to give plaintiffs a significant helping hand, to make sure their prospects are better than they would be under the rigors of the ordinary rules of litigation.(8)
When the Court's procedural decisions take a conservative, prodefendant turn, critics decry the departure from this substantive consensus. These critiques nostalgically seek a return to what they deem the correct, liberal past in which a deep societal commitment to the eradication of discrimination shaped a plaintiff-friendly procedural jurisprudence.(9) The calls for reform are essentially restorative: they seek to correct the Court's error of today by going back to the liberal consensus of yesterday. So went the intense academic and political response to Wards Cove Packing Co. v. Atonio,(10) in which that much-maligned political actor, a ,bare majority" of the supreme Court.(11) stood accused of reworking the procedures for proving Title VII disparate impact claims.(12)
Nostalgia also forms the core of the debate over St. Mary's Honor Center v. Hicks.(13) In Hicks, a "bare majority" made intentional discrimination more difficult to prove under the special proof structure for individual intentional discrimination cases--which I shall refer to as McDonnell Douglas-Burdine," in honor of McDonnell Douglas Corp. v. Green(14) and Texas Department of Community Affairs v. Burdine,(15) the two major cases in which it was set forth. Proof of discrimination under McDonnell Douglas-Burdine plays itself out in a three-part "minuet."(16) The first step is taken by the plaintiff, who must begin by proving a "prima facie case." In some legal contexts, the "prima facie case" consists of evidence sufficient to prove all required elements of the plaintiff's claim. Not so under McDonnell Douglas-Burdine.(17) The McDonnell Douglas-Burdine prima facie case merely attempts to "rule out the most common reasons for adverse job actions,"(18) entitling the plaintiff who proves a prima facie case to a presumption that intentional discrimination has taken place. In order to avoid a directed verdict, the employer(19) must then meet a burden of production--as opposed to a burden of persuasion--by introducing evidence of a legitimate, nondiscriminatory reason for its decision.(20) Once the employer has met its rebuttal burden, the presumption of discrimination created by the prima facie case "drops from the case,"(21) and the plaintiff must prove that the employer's stated justification was pretextual in order to prevail. In a much-cited dictum in Burdine, the Court stated that pretext could be proven by either direct or circumstantial evidence, and that, more specifically, pretext "may" be proven by showing that the employer's stated reason was "not worthy of credence."(22)
For at least ten years after Burdine, the circuits disagreed as to whether a plaintiff who proves a prima facie case and convinces the factfinder to reject the employer's stated justification is entitled, on that basis alone, to judgment as a matter of law.(23) The Court addressed this question for the first time in Hicks and, by a 5-4 vote, held that judgment for the plaintiff is not required in such a case.
The Supreme Court's decision in Hicks is viewed by many as reminiscent of Wards Cove and the other conservative civil rights decisions of the Supreme Court's 1988-1989 Term.(24) Indeed, the Court's action in Hicks appears to be a particularly extreme case of conservative judicial activism because the Hicks majority conceded that its decision directly contradicted the Burdine dictum.(25) The Court's explanation--that one need not, "where holdings of the Court are not at issue . . . dissect the sentences of the United States Reports as though they were the United States Code"(26)--reads to many as a declaration of war, given the fact that the Court had consistently developed the Title VII special proof structures through quasi-legislative "tests" set forth in dicta to guide future decisionmaking. Further fueling the charge of judicial activism was the fact that the position rejected by the Court in Hicks was embraced not only by the Clinton Justice Department in its amicus brief, but also by the Bush Justice Department in an earlier Supreme Court case that was not heard on its merits.(27) The criticism of Hicks was immediate, as was the cry for restorative legislation.(28)
The strategic importance of nostalgic critiques and restorative rhetoric makes it difficult for liberals, of which I am one, to abandon them. Thus, before undertaking a close analysis of the issue in Hicks, my sympathies were with the dissent. I wanted to be on the side of those who proclaim with moral confidence that the landmark Title VII cases of the early 1970s represented a clear, proplaintiff consensus, and that only a Court that no longer believes that race discrimination against nonwhites is--or ever was--a problem in our society(29) would destroy it. But when it comes to McDonnell Douglas-Burdine, I have reluctantly concluded that the nostalgic critique must fail.
The purpose of this article is to explain why the Court's much-maligned decision in Hicks was correct, and to further argue that in the aftermath of Hicks, the McDonnell Douglas-Burdine proof structure ought to be abandoned.
I begin in Part I by analyzing the Court's prior disparate treatment decisions and conclude that the Supreme Court never succeeded in setting the prima facie case threshold high enough to permit the proven prima facie case to support a sufficiently strong inference of discrimination to mandate judgment for the plaintiff when combined only with disbelief of the employer's stated justification. It is, of course, possible to argue that the Court interpreted the policies of Title VII to require a framework in which a mandatory inference of discrimination must be drawn even when the evidence itself is not strong enough to warrant it. But I fail to find in McDonnell Douglas-Burdine a sufficiently strong proplaintiff substantive consensus to support this result--a result that must, in all honesty, be viewed as a kind of affirmative action, in that it protects members of protected groups from discharge without just cause in any case in which a minimal showing can be made that discrimination could have been the cause, without proof that it was the cause. Instead, I conclude that the major thrust of the Court's disparate treatment jurisprudence is the attempt to insulate disparate treatment cases from the radical innovations of the disparate impact standard. There is a marked conservative overtone to the McDonnell Douglas-Burdine line of cases--and against its background, the nostalgic critique of Hicks is unacceptable.
If Hicks is correct, however, there remains what might well be thought of as a problem of judicial economy: McDonnell Douglas-Burdine is reduced to nothing but an empty ritual. If McDonnell Douglas-Burdine does nothing the normal rules of civil procedure cannot do, if it neither aids nor constrains judicial decisionmaking, one must ask whether it makes sense to continue to use the McDonnell Douglas-Burdine proof structure at all.
In Part II, this article questions the continued utility of McDonnell Douglas-Burdine by reviewing how district courts use McDonnell Douglas-Burdine at the pretrial stage, where it matters the most. A review of district court summary judgment cases demonstrates that to accord legal significance to the plaintiff's satisfaction of the "requirements" of the prima facie case "stage" and the pretext "stage" of McDonnell Douglas-Burdine is to engage in an act of misplaced concreteness. The world of practice under McDonnell Douglas-Burdine remains a disorderly one, in which the assignment of categories of facts to "stages" of the case is unstable. Furthermore, to the extent that McDonnell Douglas-Burdine does constrain factfinding, it tends to discourage the kind of holistic factfinding that is most likely to reveal the truth about discrimination in the workplace.
On the basis of this review, I suggest that it would be better to abandon McDonnell Douglas-Burdine than to repair it. Abandoning McDonnell Douglas-Burdine would leave courts with a less structured approach to disparate treatment cases, in which the only question would be whether the plaintiff has proved intentional discrimination by a preponderance of the evidence, both direct and circumstantial. It is not only intellectual honesty that would be gained from abandoning McDonnell Douglas-Burdine. There is also a possibility that abandoning McDonnell Douglas-Burdine will draw the attention of the courts--and of that sector of legal academia that still has some faith in the educability of the courts-to the substantive problem that McDonnell...