TRB From Washington

Poor Justice O'Connor! For the past three years, she's struggled ineffectually to split the difference between four liberal Justices, who think that the Constitution doesn't prevent the states from drawing voting districts on the basis of race, and four conservative Justices, who think it does. But last week, in striking down majority black congressional districts in Texas and North Carolina, she found an ingenious solution to her dilemma. She wrote an opinion for the Court suggesting that the Voting Rights Act might be unconstitutional, and then filed a separate concurrence to her own opinion reassuring everyone that she thinks the Act actually isn't unconstitutional. At last, O'Connor has found a Justice she can agree with.

But it's hard to be very amused by Justice O'Connor's contortions, which have created a voting rights jurisprudence that's both analytically unintelligible and politically indefensible. Three years ago, in Shaw v. Reno, O'Connor declared for the first time that minority districts with "bizarre" shapes might violate the Constitution because they "segregate" voters on the basis of race. She was widely derided for failing to identify a clear constitutional injury to the individual white voters placed in majority black districts.

As Pam Karlan of the University of Virginia argued, the white "filler people" chosen to integrate black districts are the group least likely to have been segregated, or even personally classified, on the basis of race: in North Carolina, the state didn't care whether the non-black voters were Caucasian, Hispanic, Asian or purple-polka-dotted. Moreover, O'Connor's conclusion, that a white voter can't be adequately represented in a black majority district, seemed impossible to reconcile with O'Connor's premise, that white voters don't think and vote alike.

Shaw v. Reno was a theoretical mess, but O'Connor could have made it easier to administer in practice. As Richard Pildes of the University of Michigan argued, O'Connor might have helped the states avoid the "expressive harms" of weirdly shaped districts by requiring them to draw compact districts, mathematically quantified and objectively defined. But O'Connor refused to take this logical step. Last June, in Miller v. Johnson, she joined the four conservative Justices who held that majority black districts were presumptively unconstitutional, whether or not they looked "bizarre," as long as race was the "predominant factor" motivating the legislature that drew them.

Taken literally, this new test threatened to invalidate the Voting Rights Act. Intended as a remedy for racially polarized black voting, the act, as amended in 1982, has been construed to require states to draw districts where blacks constitute a majority. But O'Connor declared last week that she doesn't want to invalidate the Act after all.

How, then, did she plan to reconcile her conflicting impulses? In the recent Texas case, the state argued matter-of-factly that race wasn't the "predominant factor" influencing the shape of the three black districts in Texas: if it had been, the districts could have been as compact and elliptical as a ten-gallon hat. Instead, the shape of the districts was distorted by political, not racial considerations--namely, the desire of white Democratic incumbents in the surrounding districts to keep enough black voters to protect their seats.

Once again, however, O'Connor abruptly shifted gears: in her opinions last week, she resurrected her original focus on appearances rather than racial motives. A bizarrely shaped district, she announced in an almost indecipherable sentence, "is part of the constitutional problem insofar as it disrupts nonracial bases of political identity and thus intensifies the emphasis on race." But in fact, as Justices Stevens and Souter emphasized in their powerful (and unanswered) dissents, it's O'Connor herself who has "intensified the emphasis on race" in the districting process. In the wake of her latest opinions, black districts have to be elegantly compact, while white districts can be as bizarrely shaped as Jim Carrey.

Is there any way to put Justice O'Connor's contortions in a more charitable light? The truth is that her concern about the aesthetics of districting isn't entirely misguided. But in her overheated rhetoric about "segregation" and "apartheid," O'Connor has missed the real scandal of the modern districting process: partisan gerrymandering, of which racial gerrymandering is only one egregious part. In the 1980s, Richard Pildes discovered, both black and white districts became much more weirdly shaped than they had been in the past. This reflected, to some degree, the increasingly byzantine requirements of the Voting Rights Act; but it also reflected the rise of sophisticated computer technology, which allowed representatives to choose their constituents with Orwellian precision. The Texas legislature, for example, consulted Census data, tabulated block-by-block and sometimes house-by-house, to diagnose the political, ethnic and demographic preferences of individual voters. The result was twenty-nine safe seats out of thirty.

If the Court is determined to continue policing the excesses of racial gerrymandering, it should, as Justice Stevens indicated, begin policing the excesses of political gerrymandering as well. Safe districts created for Democrats or Republicans send the same message as safe districts created for blacks or Germans: voters rightly perceive that their preferences are being tabulated and manipulated by the self-interested maneuvering of partisan politicians. Bizarrely shaped districts are troubling not because, as O'Connor suggested, they send an inaccurate message that all black and white voters think alike. Instead, they send an all-too-accurate message: that representatives are ignoring objective districting considerations, like geography and contiguity, and instead predicting, with the precision of a Johnnie Cochran jury selection, just how individual constituents are likely to vote. These computer-enhanced predictions offend our sense of fairness, democratic autonomy, and perhaps even privacy, not because they're lazy stereotypes, but because they're so chillingly on the mark.

By recognizing that partisan gerrymandering and racial gerrymandering are two sides of the same coin, the Court might put useful pressure on states to take self-dealing legislators out of the districting process entirely. In New Jersey, for example, a citizen initiative recently put the districting process into the hands of a bipartisan commission, whose recommendations have to be accepted or rejected by the legislature as a package. Unfortunately, our least decisive Justice announced ten years ago that she doesn't think political gerrymanders should be second-guessed by judges, no matter how bizarre the districts look, because of the dangers of interjecting courts into the political thicket. Let's hope that, once again, she decides to disagree with herself.