Wednesday, August 6, 2014
Now, I know you were just sayng to yourself, "Why hasn't another article about the empirical study of Twombly and Iqbal come out recently? I'm having withdrawal symptoms." Rejoice -- here it is.
Professor Jonah Gelbach of the University of Pennsylvania Law School has published in the most recent issue of the Stanford Journal of Complex Litigation (and earlier posted on SSRN) his article Can the Dark Arts of the Dismal Science Shed Light on the Empirical Reality of Civil Procedure?
Empirical questions in civil procedure are too important to be answered as if motivated people weren’t involved in the legal system. Parties don’t conduct their primary behavior that way, lawyers don’t plead or brief that way, and judges don’t decide cases that way. We ought not to study litigation that way, either. This paper is a step toward a better alternative.
Empirical researchers must take seriously the fact that litigation involves human beings, who are motivated and have agency. To make this point concrete, I first step outside the realm of civil procedure and illustrate the importance of accounting for human agency in empirical research. I use the canonical problem of demand estimation in economics to show how what I call the “urn approach” to empirical work fails to uncover important empirical relationships by disregarding behavioral aspects of human action.
I then show how these concerns permeate a prominent empirical issue in contemporary civil procedure debates: the changes in pleading policy wrought by Bell Atlantic, Corp. v. Twombly and Ashcroft v. Iqbal. Revisiting my own earlier work, I embed the question of how changes in the pleading standard will affect case outcomes in a broad behavioral framework that takes parties’ agency seriously. In the process, I address recent critiques, both of the very idea of using behavioral frameworks to understand civil litigation policy changes, and of my use of real-world litigation data collected by the Federal Judicial Center. These criticisms implicate all aspects of the process of empirical research: the notion of using a behavioral framework at all, the type of data needed, and the question of how best to estimate effects that the behavioral framework indicates are important, given the data. As I show, these criticisms are straightforwardly (if verbosely) refuted on the merits.
The alternative to taking seriously the behavioral context created by the civil justice system — what has occurred so far in too much of the debate over Twombly and Iqbal — is, as one critic of early 20th-century empirical research by legal scholars once put it, “a mindless amassing of statistics without reference to any guiding theory whatsoever.” To do better, we will need to take behavior seriously in studying civil litigation.