Academic

Can you use plagiarism software to do legal research?  I've been exploring the question lately, and so have some others.  Pamela Corley has looked at relationship between the briefs filed before the Supreme Court and the ensuing opinions.  And plagiarism software, it turns out, is a good tool for that sort of comparison.

Her project is pretty neat.  As you might imagine for a very unbusy Court overloaded with law clerks, time, and amici, the Court didn't borrow too much from the party briefs when it writes its decision, at least for the opinions issued in the 2002, 2003, and 2004 terms.  The average overlap between opinion and winning party brief then was roughly 10%.  I bet the percentages are higher in the appellate courts, and especially in the district courts. 

Sometimes, however, the Court found winning briefs to be, shall we say, highly persuasive.  In one case, Justice O'Connor used 41% of a respondents' brief in her opinion, and in another, she used 33% of an appellant's brief.  Rehnquist and O'Connor were the justices most likely to borrow from the briefs (they comprised 14% WJR/11.5% SDO of the content of the justices' majority opinions authored during those three years, depending on respondent/appellant), Souter the least (7% either way).

Corley found that the justices are more likely to borrow from high quality briefs (proxied by a DOJ or DC return address), from ideologically compatible briefs (conservatives were more likely to use briefs advocating the conservative position), and from briefs in low-profile cases, or at least low-profile enough not to appear on the front page of the New York Times the day after the decision was handed down (which might just mean "statutory cases").

Now if she would only have broken down those DC addresses by Supreme Court litigation boutique, then we'd have some new league tables to report to you all and said boutiques would have some new marketing materials to write.

The paper is available in a gated version here, and is forthcoming in Political Research Quarterly.