« Combining Quantitative and Qualitative Methods in Socio-legal Research | Main | New Book - Minority Rights in Asia »

22 June 2006


TrackBack URL for this entry:

Listed below are links to weblogs that reference Leiter on New Legal Realism:


Elizabeth Mertz

Thanks to everyone who contributed to this ongoing discussion. To Eileen: I did not read you as "off the point," but rather as undertaking the careful reflection on differences in scholarly premises that we hope to encourage. (I'd call this a form of metalinguistic discussion in which we pay conscious attention to what we mean when we talk across disciplinary (or sub-disciplinary) divisions.) I similarly appreciated Bill Henderson's quite open request to Bill Ford for an explanation of terms (in a different thread of this forum).

[[A quick aside, **this paragraph is only for those interested in the metapragmatics issue**: notice how much of discussion of the "meaning" of Brian L.'s previous post depended on differing readings of his metapragmatic signals (I suspect that Bill H. was responding to cues like "so-called", which often carry pragmatic/social messages far beyond the more obvious semantically-indicated "thin thread between OLR and NLR." Hence the notion of a "semantic retreat" in which we point back to the more static semantic meaning of an utterance.) (There were some similar comments exchanged on Ann Althouse's website.) When we are writing, increased ambiguity is created by the fact that we can't use paralinguistic signals like voice intonation, gesture, and facial expression to clarify metapragmatic meaning.]]

If you go to the NLR website (which is in the process of getting updated over this summer), you will see three subheadings under the logo: social science, law, policy. As we've noted, NLR scholars do hope to develop more systematic approaches not only to interdisciplinary social science translations, but also to integrating social science,law, and policy. It seems fitting that Brian (T.) and Dave should end this line of discussion by pointing to the full translation process envisioned by NLR. I appreciate both of their elaborations of the ways in which NLR can facilitate a better relationship between law and social science.

David M. Trubek

I was pleased that Stewart Macaulay recalled the article I wrote with Jack Schlegal about Charles Clark and empirical research on courts. Clark, who was Dean of the Yale Law School during the heyday of Realism at Yale and later became Chief Judge of the 2nd Circuit, was commited to empirical research on the operation of the courts. He was no social scientist but he worked with social scientists. The goal of this work was reform: Clark wanted to make civil justice work better. To do that, he had to know what really happened in the courts. He represented what we might call the "reform" strand of Legal Realism. This strand of the movement was commited to a view of law as a tool for social good. They helped develop the ideas that led to to the overthrow of the Lochner era and usher in the New Deal. For them, empirical work was anecessary part of the struggle for justice and a preconditon for reform. New Legal Realism carries on that aspect of the Realist tradition.

Brian Tamanaha

Interesting exchange. Let me suggest two different versions of the gap between "law in the books" and "law in action" which helps track a difference in the focus of the OLR and NLR, along the lines suggested by Brian Leiter.

The difference between the stated legal rules and the patterns of actual judicial decisions is one version of this gap. As Brian notes, the Realists emphasized this gap, and they were able to demonstrate it by comparing how judges actually decided cases with the rules they claimed to be applying (which is empirical in its own sense). They proposed to restate the rules to conform to the actual practice of judges to render law more predictable. More ambitiously, they tried to shift the focus from rules to situations, since that seemed to have a determinative effect on how the rules were applied.

A second version of the gap is the difference between stated legal rules and the actual behavior of people in society. The Realists also made a point of this gap, which they also sometimes (confusingly) referred to as the difference between law in the books and law in action, but few Realists actually studied this using social scientific methods, for the reasons Brian states (few had the training to do so). This is hard empirical work, which cannot be done in a library poring through volumes of judicial decisions.

NLR looks at both of these versions of the gap, and applies social scientific methods to both, but perhaps has a greater emphasis on the second gap than the first (which can be done without social scientific training, but not without legal training).

Realism, old and new, has room for and a need for both.

Eileen Braman

Sean. I think you are right that we have been talking past each other. I am entirely on board with the extent of indeterminacy in interpretive strategies deriving from the fact (1) that no two cases are exactly alike and (2) that even if they were there is a great deal of inter-subjectivity in decision making -- deriving from linguistic imprecision and other sources.

My larger point in raising the "gorilla" in the Book review is that -- if as Friedman notes political science is a highly insular discipline - then at the very least the same could be said about legal scholars theorizing about constitutional interpretation. Given the huge normative implications of how ideology actually influences decisions and/or shades interpretation(which legal realists have ALWAYS been interested in) this is an area where scholars can perhaps break important new theoretical ground by exploring each others basic assumptions and acknowledging each others' work.

I have thoroughly enjoyed reading the other comments on this thread that do not go directly to the discussion Sean and I have been having. I hope our discussion is not too far off point.

Sean Wilson

Hello Eileen. Good to hear from you again. I think we talked past each other.

I do not dispute a thing of your second post. Indeed, that our scholarship should work in a post-Dworkin paradigm (attitudes in dialogue with a pre-existing cognitive decision structure) is something I have always agreed with. Much of what Friedman said in his article was in my dissertation prospectus. So we are all on the same page as far as that goes.

However, I was only replying to the empirical claim from the book review: that the empirical models constructed by political scientists proved that "a 2000 pound Gorilla existed" and that the models showed that "often the outcome in constitutional cases has more to do with the mix of preferences of the justices on the Court then their commitment to any interpretive principle." The models, as you know, show no such thing. And based upon how these models were constructed, I really doubt whether Constitutional legal theory should "consider the findings of empirical research on the influence of ideological preferences on Supreme Court decision making." The reason why is because a proper analysis of the data in those models demonstrates that they are CONSISTENT with an interpretive paradigm.

This is not to say that new, better models will not be constructed that show something else -- or that the true reality is quite bad for an interpretive paradigm in spite of the models -- it is only to say that ideology models in political science do not prove "attitudes rule." They people who promoted those models did not understand that they were explaining variation of an index, not variation in votes; the equating of the two being an ecological fallacy.


William Henderson

When I wrote this post, it was late ... the hazards of trading blogging for sleep. In full context, Brian's remarks were not harsh. I retract the statement.

But they certainly were relevant and informative. That's why I excerpted them at length.

Elizabeth Mertz

Yes, the famous Macaulay jazz tag line! I'm encouraged by this lively line of discussion, in which some of the premises and terms of debate are themselves under consideration, hopefully in ways that clarify rather than stifle further collaborative consideration of how best to combine disciplinary perspectives. From Eileen's post, it sounds as if some thinkers in political science are also working toward an "NLR-ish" attempt to understand and respect the quite different normative demands and goals of judicial discourse (rather than approaching consideration of this "indigenous" normative discourse as inherently opposed to study of judicial preferences). If these new labels are useful for nothing else, they seem to be generating some self-conscious discussion, bringing us together in new ways and hopefully charting some new paths through the jungle of law and social science discussions.

I should add that a number of NLR folks have been very interested in pragmatism as a part of their emphasis on translating between legal theory and empirical research. (My colleague Victoria Nourse has been especially vocal on this point, as have some of the folks interested in "new governance.") Professor Leiter has an interesting article that talks about this issue: "Rethinking Legal Realism: Toward a Naturalized Jurisprudence," 76 Texas L. Rev. 267 (1997).

Laura Beth Nielsen

Finally Stewart weaves in a jazz reference -- what took so long? April in Paris is what I was listening to when I read it. No, not Harry -- Count Basie.

Brian Leiter

My thanks to Professor Macaulay for those interesting examples and references.

Stewart Macaulay

I tried to trace a connection between the old and the new legal realism in Stewart Macaulay, The New Versus the Old Legal Realism: "Things Ain't What They Used to Be," 2005 Wisconsin Law Review 365-403. Professor Leiter may well be talking about the majority or most of the realists or, perhaps, the most influential ones. Nonetheless, there is an empirical approach involved in realism that Professor Horwitz suggests that we can trace back to the Wisconsin law professors who worked with the Progressive reformers in the early years of the 20th Century. Carrington and King (47 J. Legal Educ. 297 (1997) point out that Professor Oliver Rundel studied delay in the Wisconsin criminal justice system as early as 1912.

You can brush aside Underhill Moore's parking studies (but there are lessons to be learned even from them, including the limits of total theories such as behaviorism). However, we should not forget the empirical work of other realists, such as Charles Clark at Yale in the late 1920s and 1930s. He and his colleagues did "discover" the largely administrative nature of state court civil litigation and the role of plea-bargaining on the criminal side. Schlegel and Trubek note: "What is noteworthy about all this work [by Clark and his colleagues] was that it pointed to the marginality of law, and suggested that researchers had to look beyond the law and legal rules if they were fully to understand the phenomena they were concerned with." I also consider work such as Kalven and Zeisel's jury project at Chicago and the projects at Wisconsin of Hurst, Beuscher and Remington as another stage of realism. All of this was something other than teasing out the "real" reasons for appellate decisions -- of course, Professor Leiter is right when he says that many or most of the realists did a lot of this. I see the jury project and the work at Wisconsin as one of the bridges to the Law and Society Association. (Of course, social scientists also played an important role in the process that created LSA; if I had to assign major credit for creating LSA to one person, I'd point to Harry V. Ball, then a young sociologist who was working on a Frank Remington grant for empirical work in the criminal justice process).

I think that the new legal realism continues at least part of the old. However, as my title notes, "Things Ain't What They Used to Be." (And if you don't get the reference to the Duke Ellington tune that featured Johnny Hodges, you have my deepest sympathy for what you have missed musically).

Laura Beth Nielsen

I can't comment on the debates within political science, but on Bill's orginal post, I didn't read the Leiter piece as harsh either.

I think he is saying New and Original Legal Realists share an interest in what courts do (I think and hope New Legal Realism and law and society and sociolegal studies and all the other things we call ourselves) are interested in the empirical study of law beyond the courtroom and the (law school) classroom.

The New Legal Realists (as far as I understand it) hope to use rigorous social science to show the "law in action" versus conjecture, impression, opinion, and misinformation about "law in action". This isn't meant to be critical of the original legal realists -- that was a huge advance in how we understand law. Now, the idea is to move it forward yet again.

Eileen Braman

Sean, I would like to respectfully respond by saying I think we have a sincere difference of opinion on this. Like you I came to a PhD program with JD in hand – EXTERMELY skeptical of social science approaches to studying decision making. I think my mentors and colleagues would testify that I would be to last to “buy into” any line of unsupported research on this subject. But my views have evolved. They have become much more complex. It is not law or ideology that wholly determines decisions but an interaction of the two. The great part about having empirical knowledge (for me) has been the ability to test these assertions in concrete ways – taking both preferences and what Friedman refers to as the "methodology" of legal decision making seriously. And I have learned that when you make a careful effort to do that ideological influences do not disappear. I’m not wholly convinced that they dominate – but I think that to imply the entire body of empirical research on attitudinal research is wholly without merit is somewhat overly simplistic.

I suggest people read the literature critically and decide for themselves rather than make any blanket assertions. I think the broader purpose of the discussion is to figure out what each approach has to add to our overall knowledge and how they can complement each other rather than calcify disciplinary divisions that have existed for years.

I hope this makes sense. I do respect your opinion – but I would suggest leaving it to others to make their own judgment about the overall merits of such research. Perhaps, as happened in my case, their views will evolve and change. At the very least I hope to see much more interesting work deriving from attempts to deal with these issues.


Brian Leiter

Gordon Smith correctly summarizes the basic point of my posting.

Sean Wilson

Actually, the truth is that the research on ideological preferences by political scientists suffers from serious methodological flaws. There is great reason to ignore it. It never proved what its proponents said it did. Much of what the research showed is that those political scientists who were constructing ideology models simply didn't understand how ecological inference manipulated their findings. There is no behavioral model in political science that shows that political ideology dominates voting or that constitutional interpretation does not meaningfully structure outcomes. This is a bunch of misinformation that political science departments have fed to graduate students for the last 10 years. The assertion is ultimately the product of: (1) incompetence in statistical reasoning (ecological fallacy); (2) a misunderstanding of Dworkin and other descriptions of how an epistemological constituence can meaningfully structure outcomes; (3) too great of a disciplinary urge to flirt with philosophic skepticism (without cause)and (4) a language game with the term "ideology" and "attitudes."

The best empirical evidence indicates that only about 20-30% of justice voting behavior in civil liberties cases over the last 60 or so years is due to "ideology" (depending upon the measure you choose). Today that percentage is much lower. That is not to say that the true percentage is not higher, however; it is only to say that a proper analysis of ideology models in political science never proved what scholars said.

The proof of this assertion is on its way out. You can find evidence of it here (more stuff is being posted about this daily):


Gordon Smith

Bill, I didn't read Brian's comments as "harsh" at all. He is just saying that "old legal realism" and "new legal realism" are tied by a fairly thin string: both "contrast[] what courts say they’re doing with what they actually do." Later in his post, he notes that NLR may have some value, but not because it is closely linked to OLR.

Eileen Braman

I recently had the opportunity to think about the relaionship between empirical research and interpretive theory in the context of constituional theory writing a review for the Law and Courts book review on Kenneth D. Ward and Cecilia R. Castillo's edited volume "The Judiciary and American Democracy: Alexander Bickel, the Countermajoritarian Difficulty and Contemporary Constiutional Interpretition" (a book I commend to those interested, by the way). Anyway some of that seems relevent to this general discussion about how empical research should inform legal scholarship on decision making. I copy the relevent part of the review bleow for anyone interested:

"I feel some obligation to mention the 2000 pound gorilla legal theorists have been so adept at ignoring in thinking about constitutional decision making. For years interpretive legal theory has evolved by scholars touting schools of interpretation that justify the outcomes in seminal decisions like BROWN and refute those in infamous decisions such as PLESSY V. FERGUSON (1896) and LOCKNER V. NEW YORK (1905). The merit of a particular interpretive approach is judged by its ability to justify widely acknowledged desirable outcomes. Stated simply, constitutional theory evolves “as if” there were some consistent interpretive approach that would yield such desirable outcomes.

"Of course, we, as political scientists, know better; often the outcome in constitutional cases has more to do with the mix of preferences of the justices on the Court then their commitment to any interpretive principle. Constitutional legal theory has completely failed to consider a half century of empirical research on the influence of ideological preferences on Supreme Court decision making. That being said, I would argue this is not entirely a bad thing. Interpretive legal theory covers the normative as well as empirical ground; it is properly concerned with how decisions should be made as well as how they have been made at critical junctures in our history. But perhaps the time has come for political scientists to take a more active role in the debate and consider the democratic implications of how judges actually make decisions in a broader theoretical context. Perhaps we should take a cue from Bickel and think about reassessing the role of the court in light of what we, as a discipline, understand about how judges actually exercise their substantial authority."


sorry for that typo; the "taken" shouldn't be there...


Leiter takes an odd little swipe taken at pol sci (see below) without elaboration ... wonder why?

"None of this is to deny the potential value of sound empirical work on law and the legal system. Such work might even illuminate the gap between what the courts say they're doing and what they're actually doing. (The formal political science literature on this subject is, alas, rather disappointing.)"

The comments to this entry are closed.


July 2014

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    

Site Meter

Creative Commons License

  • Creative Commons License
Blog powered by Typepad