In the New York Times this morning, George Washington U. law professor and New Republic writer Jeffrey Rosen argues that Elena Kagan should trade in her admiration for Oliver Wendell Holmes, Jr., for membership in the Louis D. Brandeis fan club. I don’t have a lot of use for either Holmes or Brandeis, so I won’t go into the main points of Rosen’s argument. But there is one thing he says that I cannot let pass–when he notes that:

Democrats repeatedly insisted during the Kagan hearings that the Roberts court has shown a pro-corporate bias — a charge supported by a recent study by the Constitutional Accountability Center, which found that in the term that just ended, the United States Chamber of Commerce won 13 out of the 16 cases in which it filed briefs, a success rate of 81 percent.

Neither Rosen nor the Constitutional Accountability Center (go ahead, click on the link, the argument doesn’t get any better) has even begun to make the case for a “pro-corporate bias” by the Roberts Court. Counting victories and defeats by corporate litigants or interested amici simply cannot demonstrate what is claimed here. The whole world understands “bias” to mean “favoring a party that does not deserve to win.” The fact that the Chamber of Commerce “won” 13 of 16 cases in which it filed briefs doesn’t tell us anything about whether it deserved to win all 16, or none of them, or something in between. The CAC “study” makes as much sense as one noting that the Chicago Cubs have not won the World Series since 1908, and therefore we can conclude that major league umpires have an “anti-Cubs bias.” Maybe they need to play better baseball.

If justice were randomly distributed, then the Chamber should have won 8 of the 16 cases (assuming, which we cannot, that the Chamber and its legal counsel took no account of the probabilities of victory in expending their intellectual resources). So if the “pro-corporate” rulings had come in just 4 or 6 of the 16, would the leftist number-crunchers at CAC tell us that the Roberts Court was “anti-corporation”? Not likely. By their lights, probably any decision in favor of corporations is evidence of the “bias” they claim to see.

But, to repeat, the real idiocy here is the notion that merely counting Supreme Court decisions, and categorizing which were “pro” or “anti” certain interests, can tell us anything about “bias” or about whether the Court is doing its job well or badly. Only an analysis of the arguments in all the cases can begin to move us toward such a judgment. “Bias” is a conclusion one reaches about a Court behaving badly, but bad behavior by the Court means the employment of bad reasoning to reach unjust results. A court that reaches conclusions favoring corporate interests 100% of the time might be right every time; then there would not be “bias” but, instead, justice. Ditto if it favored corporations none of the time. “The Court favored parties of Type A over parties of Type B in X percent of cases” tells us exactly . . . nothing of any interest whatsoever, absent a persuasive analysis of the issues in each and every case under consideration.

I belabor the point here because the vacuity is so maddeningly common, and not just in lazy commentary or in “studies” by political advocacy groups. In my own discipline of political science, an amazing amount of intellectual effort has been wasted over the last half century in studies of the “ideology” of judges that suffer from the same flaw that we see here–studies that eschew all interest in the law except as a dependent variable in behavioral science. There is a surprising number of people who think that counting can do the work that only legal reasoning can do. But there are no statistical shortcuts.

Of course, if you believe that all legal reasoning is a fraud, a mere cloak for ideological decision-making, then the shortcut makes sense. Just make up your ideological categories, and start counting. But then there would be all sorts of interesting questions about why one would want to teach law . . .