Does Asking More Questions Tip the Outcome of Supreme Court Cases?--

Tony Mauro has an interesting article on whether the number and tone of Supreme Court questions can tip off the final result in the case:

In a new study entitled "The Illusion of Devil's Advocacy: How the Justices of the Supreme Court Foreshadow their Decisions During Oral Argument," Sarah Shullman came up with a surprisingly simple and accurate way of predicting outcomes based on the number and tenor of oral argument questions by justices.

Shullman's article, in The Journal of Appellate Practice and Process, reports on oral arguments in 10 cases she observed during the October 2002 term. As she watched, she tallied the number and tenor — helpful or hostile — of all the questions asked by all the justices. Then a student at Georgetown University Law Center, Shullman is now an associate at Steel Hector & Davis in West Palm Beach, Fla.

After seven of the 10 cases she studied were decided, Shullman looked for correlations — and found them. In all of the cases, the justices in aggregate asked more questions, and more hostile questions, of the party that ultimately lost the case. The model of the devil's advocate — peppering the side you favor with tough questions — did not appear prevalent enough to derail this conclusion. . . .

In any event, on the basis of her early success, Shullman proceeded to predict the remaining three cases she had charted that were still pending before the high court. And bingo! She was correct each time. A couple of justices strayed and asked more questions of the side they ultimately favored, but overall the justices turned out to be "quite predictable," she says. . . .

Shullman acknowledges that her sample was small, but the methodology has already been tested since she did her study. John Roberts Jr., one of the masters of the trade before taking the bench in 2003, used her theory for a talk he gave on oral advocacy before the Supreme Court Historical Society last year. Picking 14 oral arguments from the 1980 term and 14 from the 2003 term, Roberts found that in fact the most questions went to the losing party in 24 of the 28 cases — an 86 percent rate of accuracy.

"The secret to successful advocacy," Roberts deadpanned in conclusion, "is simply to get the Court to ask your opponent more questions."

An 86 percent success rate in making predictions compares favorably with that of other players in the growing field of Supreme Court prognosticators. Political scientists have gotten into the game to test the relative importance of precedents and politics in Supreme Court decision making. The Supreme Court Forecasting Project, based at Washington University in St. Louis, used statistical models and a panel of experts to predict the results in the cases argued in the 2002 term. The statistical method, based on data such as the circuit of origin and an analysis of precedents, came out right 75 percent of the time, while the human experts predicted outcomes correctly in 59 percent of the cases.

At a symposium on the project, Linda Greenhouse of The New York Times got into the spirit of things and looked back at her stories from the same term and found that in the 16 decided cases in which she ventured a prediction, she was right 75 percent of the time.

It's not surprising that this might be so, but the numbers in the small-scale studies being discussed were stronger than I would have thought. The story didn't indicate whether any significance tests were done on the data, and without the exact baseline on the % of reversals of the lower court for these cases, I couldn't do the statistics myself.

I went back and roughly counted the number of questions in Raich, the marijuana Commerce Clause case that Randy Barnett argued. Randy got MANY more questions than the government, a bad sign for Randy (and the Constitution).

In skimming the transcript in Raich, at pages 40-41 I also noticed a somewhat playful disagreement between Barnett and Justice Scalia over what "home consumption" of wheat meant in Wickard v. Filburn. Randy said it included wheat consumed by the livestock on the farm; Scalia disagreed, saying it meant wheat eaten by the farmer and his family. I went back and read Wickard. Although there are some ambiguous passages, two passages are clear:

It is urged that under the Commerce Clause of the Constitution, Article I, ยง 8, clause 3, Congress does not possess the power it has in this instance sought to exercise. The question would merit little consideration since our decision in United States v. Darby, 312 U.S. 100, sustaining the federal power to regulate production of goods for commerce, except for the fact that this Act extends federal regulation to production not intended in any part for commerce but wholly for consumption on the farm. The Act includes a definition of "market" and its derivatives, so that as related to wheat, in addition to its conventional meaning, it also means to dispose of "by feeding (in any form) to poultry or livestock which, or the products of which, are sold, bartered, or exchanged, or to be so disposed of." Hence, marketing quotas not only embrace all that may be sold without penalty but also what may be consumed on the premises.

And this:

had he chosen to cut his excess and cure it or feed it as hay, or to reap and feed it with the head and straw together, no penalty would have been demanded. Such manner of consumption is not uncommon.
So Randy is correct on Wickard. Let's hope he is also determined to be correct that the interstate commerce clause applies to "interstate commerce," not intrastate noncommerce.

Reading Wickard, which is a pretty weakly reasoned case by my lights, reminds me that the Supreme Court is in a bind in Raich. Either the Court has to follow the Constitution and strike down federal drug regulation of intrastate noncommercial uses of marijuana (a controversial decision to follow the rule of law), or it has to expand Wickard radically, rendering the Commerce Clause almost (though not quite) a dead letter. Stated another way, the Court either has to expand its federalism jurisprudence slightly (eg, Lopez & Morrison, but in the controversial drug area), or it has to limit Lopez & Morrison to their facts by radically expanding federal power under the Commerce Clause. It can't stand still. Perhaps that is why the Court has been so slow to render an opinion.

Related Posts (on one page):

  1. Ann Althouse Ponders Raich,--
  2. Does Asking More Questions Tip the Outcome of Supreme Court Cases?--
Ann Althouse Ponders Raich,--

Ann Althouse has a nice post analyzing Raich, the medical marijuana Commerce Clause case. She thinks that the government is likely to win.