Determining the Authors of Unsigned Opinions

At the California Appellate Report, Shaun Martin posts about the recent case, Blixseth v. Yellowstone Mountain Club. Here is the first paragraph of the opinion:

Timothy Blixseth sounds the clarion call of many a disappointed litigant: “It’s not fair!” He insists that the judge who presided over the administration of the Yellowstone Mountain Club ski resort’s bankruptcy was biased against him and should have recused himself. The bankruptcy judge denied the recusal motion and the district court affirmed. Blixseth has now filed a blunderbuss appeal.

Martin writes:

The panel consists of Judges Kozinski, Paez and Berzon. It's a per curiam decision.

Which means, of course, that we can only wonder baselessly about who the actual author of this sharply worded opinion could possibly be.

He then includes a picture of the judge he thinks is the author. I think he's right.

But if we want to be sure, we always check out this recent article by William Li, Pablo Azar, David Larochelle, Phil Hill, James Cox, Robert C. Berwick, and Andrew W. Lo. The title of the article is Using Algorithmic Attribution Techniques to Determine Authorship in Unsigned Judicial Opinions, and the citation is 16 Stan. Tech. L. Rev. 503 (2013). From the abstract:

Our work uses natural language processing to predict authorship of judicial opinions that are unsigned or whose attribution is disputed. Using a dataset of Supreme Court opinions with known authorship, we identify key words and phrases that can, to a high degree of accuracy, predict authorship. Thus, our method makes accessible an important class of cases heretofore inaccessible. For illustrative purposes, we explain our process as applied to the Obamacare decision, in which the authorship of a joint dissent was subject to significant popular speculation. We conclude with a chart predicting the author of every unsigned per curiam opinion during the Roberts Court.

One step that Li et al. took to determine the identity of per curiam opinion authors was to identify particular words that specific justices used uniquely and often. Some judges may be harder to identify than others, but I think that it would be pretty easy to work a judge like Alex Kozinski into the algorithm. For example, I think that the phrase "blunderbuss appeal" would be a solid indicator that Kozinski is writing the appeal.