Waiting for the Future: Some Realism about Empirical Legal Studies

“For the rational study of the law the black-letter man may be the man of the present, but the man of the future is the man of statistics and the master of economics.” (The Path of the Law, 1897)

No figure was more iconic for the small group of academic lawyers who thought of themselves as “realists” than Justice Oliver Wendell Holmes (1841-1935, Supreme Court 1902-1932) and no statement better exemplified his appeal to them than his bold prediction quoted above.  Yesterday, coming from my second Berkeley Law job talk in as many days in which candidates presented empirical scholarship and statistical data, and having just taught Holmes' essay in a sociology of law seminar, I found myself wondering if this prediction is finally coming true.

There are many ironies about this quote.  First, as Jack Schlegel shows in his great American Legal Realism and Empirical Social Science, just about the only people any where near the realists who could be described as statistically sophisticated were women like Dorothy Swain Thomas.  Indeed, as my colleague Kristin Luker argued in a talk she gave here a couple of years ago, quantitative methods were not as prestigious in the social sciences in the early 20th century, as they would be after World War II, and the often hard work involved in assembling and analyzing statistics was largely left up to women.

Second, hardly any of the published work associated with Legal Realism in the 1920s and 1930s consists of statistical analysis.  As Schlegel argues, most of the Realists were producing scholarship that played by the conceptualist rules of Classical Legal Thought.  They frequently asserted their belief that the legal rules they discussed should be reformed to achieve socially optimal outcomes, but they did so as another variation on the conceptual board, only a few (Douglas, Clark, and Moore on Schlegel’s account) actually worked with statistics.

Third, while Holmes privileges economics, he had in mind and economic science focused on institutions of the labor market that looks more like empirical sociology or labor economics today then most of what has passed under the banner of law and economics in the post-60s law school.

Holmes (or at least his readers) may have conflated two quite distinct ways that law might become statistical; as a tool of predicting judges, and as a method for analyzing optimal social rules.  He opined in the same article that the real business of law was predicting the decisions of judges (and presumably other legal decision makers).  Presumably such prediction could be improved by the development of actuarial methods and their utilization by lawyers and law professors (after all, regression equations are about making predictions).  Indeed, the project of modeling judicial decisions has long been a substantial focus of empirical scholarship on law (especially by Political Scientists) and seems to be enjoying a considerable vogue in law schools. 

Holmes did not make explicit reference to the second meaning of evidence based legal rule fashioning (indeed he had a kind of nihilistic confidence that popular decision making should prevail over expertise, note that having praised master of economics, he was famous for dissenting in Lochner because “this case is decided upon an economic theory which a large part of the country does not entertain.”) but his Realist fans could easily see their own preferred model of law as social engineering celebrated in his bold prediction.  Consider Roscoe Pound:

“The sociological movement in jurisprudence, the movement for pragmatism as a philosophy of law, the movement for the adjustment of principles and doctrines to the human conditions they are to govern rather than to assumed first principles, the movement for putting, the human factor in the central place and relegating logic to its true position as an instrument, has scarcely shown itself yet in America.  Perhaps the dissenting opinion of Mr. Justice Holmes in Lochner v. New York, is the best exposition of it.”

Unfortunately for the Realists, it was far from clear that the social sciences in 1928 had the tools to answer questions about social optimality  (even assuming the politics of law were such as to permit agreement on the substantive ends against which optimality would be judged).  In fact, there were important studies being done that had considerable relevance to law, including the criminological studies of Robert Park and Ernest Burgess at Chicago, but with the possible exception of sociologist Sheldon Glueck (appointed by Pound to the Harvard Law faculty in the ‘30s), none of the Realists were reading or citing them.  Perhaps the most important academic figure, was pragmatist philosopher John Dewey, who like them articulated a vision of governance and law with an experimental approach directed toward explicit purposeful social ends, and Dewey himself was no more empirically oriented than most of the Realists themselves were.

But the Realists were enthralled with more than a mere idea.  There was one extraordinarily powerful example of what they seemed to be getting at, an example that combined almost all the heterogeneous elements of their aspirations, the abrogation of the law of employers liability (including its defenses) and the adoption of workers compensation in its place during the first three decades of the 20th century. [ fortunately a plethora of outstanding socio-legal historians have plumbed this topic including Lawrence Friedman (article with Jack Ladinsky), Chris Tomlins (article) and John Fabian Witt (book)].  The workers compensation example combined massive law reform, constitutional battles (eventually most state Supreme Courts rejected Lochner type attacks), a triumph of social legislation to enhance the lives of the working classes (although not all workers cheered), and a whole new legal framework based on administrative processes and anchored, yes, in statistics.

For the Realists, and many of their contemporaries this story was one that seemed to confirm the expectation that a major transformation of law and governance was about to unfold and that statistics would play a critical role in it.  But while the administrative state built by the New Deal, shared some of these features (and imitated others, like the supposed “insurance” aspects of social security), it did not in the end presage a transformation in the broader work of law and lawyers along the lines of the Realist vision.

Has the hour of Holmes’ prediction come round at last?  Classical legal thought has remained remarkably enduring as a model for legal teaching. Even today, the law school classroom is dominated by teachers and a teaching model that demonstrates deft and crisp articulation of the axioms and principles that are believed to dominate particular doctrinal fields and to be in play in any particular case. If you interrupt the flow of that articulation to raise questions about the social context in which that case is operating, you will annoy your students (get tenure first, then annoy them).  Holmes wrote in 1897 about "the present divorce between the schools of political economy and law" (when did he think they were married?).  At Berkeley and some other schools this second marriage seems well underway.  This post has gone on too long, but I'm going to try and be more explicit next post about why I think the relationship is more durable this time.