By Deborah A. Widiss. Full Text.
The Civil Rights Act of 1964 and other employment discrimination laws make the “simple but momentous” declaration that it is illegal to deny employment on the basis of race, sex, religion, or other key aspects of identity. But when employees who have been treated unfairly turn to the courts for relief, courts rarely assess whether their claims meet the statutory standard. Instead, they funnel the evidence through a convoluted body of judgemade law known as McDonnell Douglas burden-shifting.
This is a serious problem. McDonnell Douglas is not only unanchored to the statutory language—it is deeply in tension with the statutory language. Bostock v. Clayton County and other decisions analyzing the causal language in employment discrimination statutes expose the extent of this disconnect. These cases reaffirm that, at most, an employee needs to prove that a protected trait made a difference in an employment decision. Under this standard, it is irrelevant whether other factors also played a role. McDonnell Douglas, by contrast, effectively requires a plaintiff to prove discriminatory bias was the only cause of an adverse action. Other aspects of the test also function as judicially-created hurdles that may be irrelevant to the ultimate question a jury would decide.
Although this doctrine can seem technical, its effects are very real. Workplace discrimination remains distressingly common, but plaintiffs are more likely to lose on summary judgment in employment cases than in other kinds of civil litigation. Often, this is because they cannot satisfy elements of the McDonnell Douglas test, even when there is ample evidence suggesting bias or harassment.
This is the first Article to lay bare the fundamental inconsistency between the statutes’ causal language, as definitively interpreted by the Court in recent cases, and the burden-shifting process. Emerging circuit case law takes steps to address these tensions. The Article argues such “middle down” reform is a promising development, but ultimately, the Supreme Court should provide clear guidance on how to modify the test to conform to the statute, or simply instruct courts to no longer employ McDonnell Douglas burden-shifting.
The Article also contributes to a nascent but growing literature on the progressive possibilities of textualism. While textualism has long been associated with right-leaning judges, there is nothing inherently conservative about prioritizing fidelity to statutory language. A fair reading of a progressive statute will often—and should often—advance progressive objectives. Refocusing attention on the text of anti-discrimination laws is necessary to realize their promise of equality.