The Supreme Court by the Numbers

Statistical analysis has been successfully applied to all manner of social issues from demographics to economics. The Oakland Athletics and the Boston Red Sox have even demonstrated that breaking baseball down to the numbers can help compensate for a limited player payroll. It is also possible to misapply statistical analysis to mislead. Sometimes this is done in error, sometimes with a deliberate intention to deceive. Public opinion polls are perhaps the most common way to lend a patina of statistical certainty to partisan persuasion. Joel Best has even written a book, Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists. Hence, when someone successfully and honesty applies information theory to a political question it deserves special attention.

In June of this year in an article in the Proceedings of the National Academy of Science , Dr. Lawrence Sirovich of the Department of Biomathematical Sciences at the Mount Sinai School of Medicine attempted to mathematically characterize decisions by the second William Rehnquist Court. This Court is comprised of the members since the last appointment to the Court of Stephen Breyer in 1994 by President Bill Clinton. Sirovich included in his analysis only those decisions for which there were nine clear separate decisions. Some cases are just issued “by the court” with no delineation of the vote. In other cases, one judge may have recused himself or herself. In still other cases, judges may split their decision with respect to different parts of a case. In all, he retained about 70% of the cases.

Sirovich then considered two possible models of the court: an “omniscient” court and a “platonic” court. In the omniscient limit each judge knows the appropriate outcome and all nine judges arrive at the same decision. All court decisions are then 9-0 judgments. In the opposite, platonic limit, all the cases are extremely close and the likelihood of each judge voting a particular way is 50%. There is no way to predict which way a judge will vote on a platonic court. Of course the real statistics lie somewhere in between these extremes.

Nonetheless, it is remarkable that 47% of the Rehnquist Court’s decisions are unanimous 9-0 judgments, reflecting a tendency to an omniscient court. If the issues in this many cases are so clear that the Court has no difficulty arriving at a unanimous conclusion, it is a shame that these cases have to be adjudicated by the Supreme Court.

Not surprisingly justices have ideologies and perspectives that force them, like grapes, to group into bunches. It is no secret that Justices Antonin Scalia and Clarence Thomas concur 93% of the time. Of course, this includes the 47% of the time that all the Justices agree. This common Scalia-Thomas agreement has led to some off-color jokes about how Thomas does not think for himself. However, we hardly ever hear about the 90% of the time that Justices Ruth Bader Ginsberg and David Souter vote together. Is anyone asking whether Ginsberg, who is a junior member of the Court with respect to Souter, is not up to her position and just relying on the guidance of Souter in her decisions? Of course not.

Given the overwhelming number of times that justices agree with one another, it would be unfair to characterize any justice as “extreme” or “out of the mainstream.” Justice John Stevens is by far the justice most likely to march the path of a lone dissenter. This is hardly evidence that Stevens is, depending on your perspective, a bold independent thinker or a dangerous maverick. He was the lone dissenter 23 times out of 377 cases. Rehnquist, Scalia, and Thomas were sole dissenters only three times a piece.

If all decisions were 9-0 judgments, having nine justices is redundant. A single justice could perform the same job with equal skill. If the judges acted independently, we would still require all nine judges to arrive at the same set of results. The correlation of decisions by judges, Sirovich suggests, would require only 4.5 to 5 independent judges. This result is a mathematical one, not a practical recommendation.

The composition of the Court and potential Court appointments have been the object of much consternation on the part of the political Left and the Right. Democrats in the Senate are willing to sustain a filibuster in order to pour sand in the Bush judge appointment apparatus. It is the lower court judges that Bush is submitting now that may provide potential Supreme Court justices in the future. What Sirovich makes clear is that there are statistically few cases that contribute to this rancor. Nonetheless, it is those high-intensity cultural issues like abortion, affirmative action, the proper roles of religion and the state — cases that do not have much of a statistical impact on raw decision numbers — that drive our perception of the Court and individual justices.

Leave a Reply

You must be logged in to post a comment.