Diversity and discrimination: a look at complex bias.

• Author: Kotkin, Minna J.

ABSTRACT

Multiple claims have become a fixture of employment discrimination litigation. It is common, if not ubiquitous, for court opinions to begin with a version of the following litany: "Plaintiff brings this action under Title VII and the ADEA for race, age, and gender discrimination." Equal Employment Opportunity Commission (EEOC) statistics show exponential growth in multiple claims in part because its intake procedures lead claimants to describe their multiple identities, at a time when they have little basis upon which to parse a specific category of bias. But increased diversity in workplace demographics suggests that frequently, disparate treatment may in fact be rooted in intersectional or "complex" bias: although stereotypes for "women" have somewhat dissipated, those for "older African American women" still hold sway. Complex bias provides a counter-narrative to the currently in-vogue characterization of workplace discrimination as "subtle" or "unconscious."

Despite the common sense notion that the more "different" a worker is, the more likely she will encounter bias, empirical evidence shows that multiple claims--which may account for more than 50 percent of federal court discrimination actions--have even less chance of success than single claims. A sample of summary judgment decisions reveals that employers prevail on multiple claims at a rate of 96 percent, as compared to 73 percent on employment discrimination claims in general. Multiple claims suffer from the failure of courts and intersectional legal scholars to confront the difficulties inherent in proving discrimination using narrowly circumscribed pretext analysis. Applying "sex-plus" concepts does not address the underlying paradox inherent in the proof of these cases: the more complex the claimant's identity, the wider must be cast the evidentiary net to find relevant comparative, statistical, and anecdotal evidence. Overcoming the courts' reluctance to follow this direction requires the development and introduction of social science research that delineates the nuanced stereotypes faced by complex claimants.

TABLE OF CONTENTS INTRODUCTION I. THE RISE AND FALL OF COMPLEX CLAIMS A. The EEOC and Complex Claims B. Empirical Evidence of Complex Claims at the Agency Level C. Demographics and Complex Claims D. An Empirical Look at Complex Claims in the Federal Courts E. Why Multiple Claims Fare So Poorly II. A DOCTRINAL FRAMEWORK FOR COMPLEX CLAIMS: THE "SEX-PLUS" ANALYSIS A. The Origin of "Sex-Plus" in Disparate Impact Cases B. The Expansion to Disparate Treatment Class C. "Sex-Plus" and Sexual Harassment D. Complex Claims Under Different Statutes III. INTERSECTIONAL SCHOLARSHIP IV. PROBLEMS OF PROOF: A LOOK AT TWO CASES A. Jeffers v. Thompson B. Wittenburg v. American Express C. Why Plaintiffs Lost and How They Might Have Won CONCLUSION APPENDIX INTRODUCTION

When an employee alleges discrimination on the basis of sex, age, and race, is she "crying wolf' or, as one judge put it, "throwing spaghetti at the wall to see what sticks"? (1) Or is she expressing the reality of today's workplace that diversity is tolerated, or may even be valued up to a point, but too much difference opens the possibility that an employee is singled out for disparate treatment?

Take, for example, the following cases. A female assistant stage director at the Metropolitan Opera claims that she was subject to a hostile work environment and discharged on the basis of her age, gender, and sexual orientation. (2) A file maintenance clerk alleges she was terminated because she is an older African American woman who is a Jehovah's Witness. (3) A hospital material distribution manager argues that he was fired due to his Italian ancestry, his gender, and his disability as a result of diabetes. (4) How do we react to these factual claims? Do we think, or perhaps more importantly, do judges think, "give me a break?" Or is there any recognition that subtle but real discrimination may be at work?

Claims such as these are a fixture of current employment discrimination litigation. (5) Indeed, it is common, if not ubiquitous, for opinions to begin with some variation of the following litany: "Plaintiff brings this claim under Title VII and the ADEA for race, age, and gender discrimination." (6) Equal Employment Opportunity Commission (EEOC) intake procedures guarantee that many such claims will be filed without a factual foundation. (7) Courts have devised no consistent or fully articulated theory to address multiple claims. The courts that consider such claims seriously rely on a "sex-plus" analysis that does no more than acknowledge the possibility of subclass discrimination. (8) Although scholars have made much of the multiplicity, indeterminacy, and fluidity of identity, they have offered little in the way of guidance for the resolution of the everyday employment discrimination action that is a concrete manifestation of postmodern legal theory. (9) Empirical evidence demonstrates that multiple claims are all but impossible to win, more problematic even than single claims. (10) In this Article, I undertake to look more closely at claims brought by "complex subjects," to use Kathyrn Abrams's phrase. (11) My goal is not to consider whether such claims should be recognized, as has been the thrust of prior scholarship. (12) By and large, the courts have accepted, either explicitly or implicitly, their legitimacy. Rather, I examine how multiple claims should be analyzed to uncover what I suggest is the complex bias that underlies them.

This project might be considered a part of the more generalized body of recent scholarship articulating the view that there is something very wrong with employment discrimination law today. All of this work stems from the recognition that the federal courts increasingly reject the vast majority of such claims at a time when there is still substantial evidence of bias in the workplace. (13)

Several interrelated strands of this critique can be identified. The first, which in part underlies all of this scholarship, explores the concept of "subtle bias": (14) the proposition that decision making in the workplace is infected by unconscious attitudes, which create skewed results for protected groups, but discrimination law is too crude a vehicle to tease out these biases. (15) This concept was first articulated beginning in the late 1980s. Charles Lawrence, Linda Krieger, and David Oppenheimer all wrote ground-breaking articles that labeled the phenomenon, respectively, as "unconscious racism," (16) "cognitive bias," (17) and "negligent discrimination." (18) Another group of scholars has looked at employment discrimination litigation from an empirical perspective, demonstrating that plaintiffs have very little chance of success both at the summary judgment stage and at trial. (19) The skewed outcomes have been attributed not only to the difficulties of proving subtle bias, but also to negative judicial attitudes and doctrinal limitations. Several authors (including myself) have undertaken these projects explicitly to respond to conservative critics, who see employment discrimination legislation as primarily creating a new kind of lottery for protected classes, adding to the 'litigation explosion," and disadvantaging American business by necessitating the expenditure of resources on frivolous employment claims. (20)

A third take on subtle bias comes from the "behavior realism" school. This experimentally-based movement relies on the social science-based principle of implicit cognition: the theory that the perceptions and attitudes that motivate action are not under the conscious or intentional control of the actors. (21) Legal scholars exploring the relationship of this branch of social science to employment discrimination find support for subtle bias in the Implicit Association Test (IAT), which demonstrates the prevalence of unconscious stereotypes. (22) The IAT measures implicit attitudes by comparing, for example, the response times for associating positive words with African American faces, in comparison to European (white) faces. (23) Researchers have found that the IAT reveals far more bias than subjects explicitly express. (24) Normative suggestions arising from behavior realism include reconsideration of affirmative action in the workplace and a more critical look at doctrine that rests on some untested and intuitive notion of psychology, such as the "same actor" rule, which posits that someone who hires a member of a protected group will not thereafter evince bias toward that person. (25)

Another branch of the discrimination law critique comes from the "structuralists." In essence, they concede the impossibility of sorting out subtle bias in the individual disparate treatment case, and instead call upon the courts to concentrate on the internal mechanisms that employers have put in place to guard against bias. (26) Tristin Green describes this effort:

Recognizing that Title VII of the Civil Rights Act of 1964, the mainstay of legal prohibition on discrimination in employment, falls short of addressing the problem, legal scholars have begun to formulate a new paradigm of regulation that would impose an obligation on employers--through legal rights or otherwise--to take structural measures to minimize discriminatory bias in workplace decisionmaking. This "structural approach" aims to minimize discriminatory decisionmaking at the individual level and to reduce unequal treatment in the workplace by pushing change at the organizational level in work environments and decisionmaking systems. (27) This approach looks particularly to the "new workplace," in which long-term employment is not presumed, and in which strict hierarchies have been replaced with team-building. In these workplace settings, subtle bias can work to undermine opportunities for protected group members, through day-to-day decisions that may result in a definable adverse...