« Legal Profession Panels at LSA | Main | New Book - Graber on Dred Scott »

July 13, 2006


William Henderson

Many thanks for this exchange. I read Spaeth and Segal as a 3L in my law and social science class (the same semester, in fact, as my Con Law Equal Protection / Sub Due Process class). The clouds literally parted; everything in Con Law made a lot more sense.

It is follow-up work like my Indiana colleague Eileen Braman that, I think, has the opportunity to eventually find its way into law school class rooms.

Finally, as a former federal appellate law clerk, I observed both jurisprudential and and attitudinal tendencies, though much more of the former than the later. Most cases have a statute or case on point and don't tap into intense preferences. Just reading the briefs and writing the bench memo, most competent clerks can predict on the outcome of 80 to 90 percent of the cases. In my experience (a sample of one), only about 10 percent of the *published* cases have sufficient legal uncertainty to invite judical lawmaking (though some jurists, I admit, are more activist than others).

My former professor said it best: former Circuit Court clerks tend to believe in jurisprudence while former SCOTUS clerks inevitably inch closer to a legal realist position.

Eileen Braman

Sean, knowing how to blog is one thing -- knowing how to post links is another altogether. The Journal of Politics piece Jeff mentioned (May 2006) is the one on threshold decision making. The analogy paper (coauthored with Tomas E. Nelson) is currently 'under review' and not available on line. I'd be happy to send you a copy of the latest iteration.

Sean Wilson


.. could you post a link to your research? I'd like to put it on my website on the "new research" page. (If you do not want to post it here, perhaps you could email it to me).

Eileen Braman

Thanks Jeff for mentioning the JOP piece.

I'd just like to mention for any legal academics who may have been told the model is strictly "at odds" with legal conceptions of decision making it is not necessarily so -- although that IS a big misconception.

My research has actually evolved via an effort to reconcile legal and attitudinal accounts of decision making. I'm basically interested in how legal decision makers reach conclusions consistent with their preference in the context of accepted decisional norms. Specifically, I've done this in a series of experiments looking at how attitudes influence analogical perception (how lay and legally trained decision makers perceive precedents related to pending litigation) and experiments looking at "threshold" decisions in complex (multidimensional) cases.

In each of my studies there is evidence of attitudinal influence AND legal constraint -- but quite honestly more attitudinal influence than I was expecting as a former law student and practitioner all too familiar with the legal socialization process.

I guess my point is there is great potential to do work at the intersection of these perspectives. In my next project I hope to test the conventional wisdom about the differential constraining influence of rules vs. standards on legal discretion where decision makers are asked to make similar judgments under different decision rules.

I wish I had heard more about this stuff as a law student -- then again maybe its better I didn't (I have tons of research questions now). Still I think legal academics who do not know about this model or dismiss it summarily are doing themselves (and perhaps ther students) a significant disservice. I can't believe there are many of those types reading this blog but I hope the number of scholars threatened by such research are diminishing.

Sean Wilson

... at least that's the version they tell to the graduate students. Although it is good to see Jeff (and Harold?) admit that they were working within a deficient philosophic paradigm when they constructed their models of judging -- they are apparently no longer doubting a Dworkin-effect that causes "law" itself to structure judicial attitudes -- it is nonetheless difficult to see certain empirical falsehoods continue to exist.

I really don't know how political science will ever become relevant if we treat models as we do rhetoric. What I mean by that is the bivariate model that Jeff refers to (the one with the high R-squared) loses about two-thirds of its explanatory value if you don't put the voting data into percentages before conducting regression analysis. As I have quite clearly shown, aggregation is a modeling technique that causes exaggerated and unreliable estimates of goodness of fit. Moreover, there is no sound empirical modeling protocol that suggests one should IGNORE individual level data in favor of a regression of percentages. (Ecological regression is performed only when you do not have individual-level data or when multi-level analysis accompanies it).

To see the verification of this, visit my website:


The truth is that Jeff and Harold never had a model of attitudes that explained anything other than about 24% of the civil liberties votes. Still a good model, of course, but not one that could support the large language that came with the packaging. Indeed, one wonders whether Jeff's model isn't more compatible with a Dworkin-like view of judging, and whether it was simply stuffed into a hard-core realists (or "critical") framework because of the motivated reasoning of the scientists.

Also, keep in mind that Jeff's model only explains about 9% of the votes on today's Court because the extreme justices that used to anchor Jeff's model -- Rehnquist and the liberal members of the Warren Court -- are now gone.

I don't have time to get into the search and seizure model that doesn't understand the difference between "facts" and "doctrine." But take a look at Barry Friedman's article; he understands the point.

The comments to this entry are closed.


April 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      

Site Meter

Creative Commons License

  • Creative Commons License
Blog powered by Typepad