Supreme Court Independence, by the Numbers

By Guy Gugliotta
Washington Post Staff Writer
Monday, July 28, 2003; Page A07

For years, two schools of thought have debated how the Supreme Court makes decisions. Are the nine justices simply "politicians in robes," destined by ideology to vote a certain way, or are they independent actors, whose opinions reflect their times, their experience and, most of all, the law itself?

Mathematician Lawrence Sirovich, of Mount Sinai School of Medicine's Laboratory of Applied Mathematics in New York, stepped into this morass last month, introducing a purely mathematical model to gauge the justices' independence simply by cataloguing how often each one sides with the majority or the minority.

Using the techniques of information theory, Sirovich analyzed 468 opinions by the "Second Rehnquist Court" between 1994 and 2002 to assess how often the justices seemed to fit into predictable ideological boxes. Information theory is a mathematical tool designed to highlight the "unexpected" in complicated systems -- such as a nine-headed Supreme Court.

Sirovich used a scale of 1 to 9, where a score of 9 meant the justices always made decisions without regard to ideological alliances or factions, and a score of 1 demonstrated ideological lockstep. The court scored 4.68. Get rid of the unanimous findings -- 47 percent of the total, which Sirovich suggested were "no-brainers" -- and the rating soared to 6.2.

"That's pretty good, in terms of how close it is to the ideal," Sirovich said. "In my personal opinion there's a certain appeal to 9, but that would show tremendous independence and no baggage at all. I think even 4.68 is high."

He acknowledged, however, that the justices' perceived politics do play a significant, if not dominant role. After unanimity, the second-most frequent breakdown he found was the 5-to-4 "conservative" majority, with conservative Justices Clarence Thomas, Antonin Scalia and William H. Rehnquist and "centrists" Anthony M. Kennedy and Sandra Day O'Connor, prevailing over "liberals" John Paul Stevens, Ruth Bader Ginsburg, Stephen G. Breyer and David H. Souter.

Not being a constitutional scholar, political scientist or court historian, Sirovich offered no analysis of his results, simply publishing them in June in the Proceedings of the National Academy of Sciences and leaving others to offer interpretations. "It's great that math can do something interesting," he said. "I don't go much further than that."

Others do. Yale University legal scholar Yochai Benkler suggested that Sirovich's work offers a fresh perspective on the debate about how the court arrives at its opinions. "What is new and interesting about the Sirovich analysis is that it said to the debaters that 'You're both right, and it's a lot more complicated than you think,' " Benkler said.

The politicians-in-robes school uses Bush v. Gore, in 2000, as its recent exemplar, Benkler said, holding that the conservative majority voted ideologically to deliver the presidential election to George W. Bush.

The legal-historical school, Benkler said, "concedes that there are some political leanings, but that law, the facts of the case, the historical moment and the justices' backgrounds are what determine an outcome. The court itself makes that claim."

A good example for this view might be this year's American Insurance Association v. Garamendi, which struck down a California law allowing Holocaust survivors to demand information from insurance companies that might help them collect benefits from policies sold to family members before and during World War II.

This 5-to-4 decision brought Souter together with Breyer, Rehnquist, O'Connor and Kennedy, who argued that the law unacceptably undermined the president's ability to make foreign policy. Ginsburg wrote the dissent, joined by fellow "liberal" Stevens and "arch-conservatives" Scalia and Thomas.

"But there's not a clear answer about what to do with it," Benkler said of the Sirovich model. "It provides a new answer to those who say it's all about politics, but it can't give you an answer on a single case."

Sirovich said he decided to study the court after reading a newspaper article last year that noted that the current group of justices had been together since 1994, a record of Supreme Court stability unmatched since 1823.

This offered an opportunity to analyze a large number of cases, he said.

He eliminated opinions in which fewer than nine justices voted, those in which one or more justices voted differently on different parts of a case, and anonymous per curiam ("by the court") opinions. He was left with 468 opinions.

He sorted them by describing the frequency with which vote breakdowns occurred. Of 256 possibilities, the most frequent by far was the 9 to 0 unanimous decision (47 percent). The conservative majority stuck together on 9.6 percent, followed by cases decided by 8 to 1 votes with Stevens as the lone dissenter.

"For comparison purposes," Sirovich wrote, think of the justices' overall performance as falling along a scale of 1 to 9. Nine is the "Platonic Court" in which "each justice is free of ideology and sees equally compelling arguments on both sides of each issue."

One is the "Omniscient Court," in which each justice "always makes the right decision." The justices therefore act as one. "We have nine justices," he said in a telephone interview, "but suppose sometime in the future identical people are put on the court and vote the same way all the time. It would be as if we only had one justice.

"I point out that Thomas and Scalia vote the same way 95 percent of the time," Sirovich said. "We don't have truly nine justices. On the other side of the fence, Ginsburg and Souter vote the same way 90 percent of the time. Each one of these correlations erodes the number 9."

For some in the "politicians-in-robes" school, Sirovich's analysis is a modest result that uses a more complicated method to belabor the obvious.

"The court's all ideological," said University of Houston political scientist Keith Poole. "Line up the justices liberal to conservative, and every time [the opinion] was 7 to 2 and the liberals won, you'd know that Thomas and Scalia were the losers." If it was 6 to 3, add Rehnquist to the dissent, he said. This simple political spectrum would yield 93 percent of the court's opinions, he said.

But not all. "Like any theory, [the Sirovich analysis] confirms what is obvious but poses new questions," Benkler said. "There is empirical support for the claim that justices are independent and cast interesting votes that are not as predictable as one would have thought."