The Power of Two: A noble minority on the Civil Rights Commission - commission report, with two dissenting votes, concluded that African Americans were more likely to be disenfranchised in Florida, than whites

In August, when Al Sharpton talked about why he wants to embark on a fantasy campaign for president, he made sure to mention "voter disenfranchisement" in Florida in last year's election. Just about every other Democrat can be expected to do the same between now and 2004. Reverend Al repeated the term like a mantra in a round of interviews, and nodding scribes on the Sharpton beat have buttressed his point with dutiful assertions that blacks were nine times as likely as whites to have their ballots invalidated.

That startling figure promises to become a piece of conventional wisdom, a permanent part of the Florida story that forever questions the legitimacy of President Bush's election; but it's almost certainly false. The only reason it has any currency at all is that the U.S. Commission on Civil Rights concocted it. To ensure that it would stick, the commission even engaged in a little disenfranchisement of its own, suppressing the voices of two GOP-appointed commissioners who hold a different view.

If a group of conservatives had done to liberals what this commission's liberals have done to Abigail Thernstrom and Russell Redenbaugh, the Washington Post would be splashing the details on its front page and the New York Times would be editorializing against right-wing repression. This particular episode, however, has attracted scant attention. That's a shame, because it's a fascinating story that shows how far some so-called civil-rights activists will go in advancing a political agenda.

Fairness, of course, was too much to expect from the commission; its chairman, Mary Frances Berry, had been on an anti-Bush crusade all along (see my article "'A Threat to Our Domestic Institutions'" in the April 2 NR). Her determination to undermine the current administration was on full display in her bombastic report, which accused Florida officials of "a pattern and practice of injustice, ineptitude, and inefficiency." Despite these strong words and many others like them, the report failed to authenticate a single case of actual voter discrimination. It contained plenty of anecdotes of dubious merit, plus a statistical analysis of ballot-rejection rates by Allan J. Lichtman, an American University professor who has done consulting work for Al Gore. It was Lichtman's claim that blacks were nine times as likely as whites to have cast invalid ballots.

Before the commission can issue a report, the individual commissioners must vote to approve it. They don't actually write the report themselves; a large staff compiles it, under the direction of the chairman. Drafts are supposed to circulate before a vote, giving commissioners a chance to review the staff's product. In the case of the 200-page Florida report, however, Thernstrom and Redenbaugh didn't receive their copies until three days before the vote-and one day after it had been leaked to handpicked reporters for the purpose of creating positive spin. This stunt would be bad enough under normal circumstances, but it was particularly cruel with regard to Redenbaugh, who is blind and has an assistant read commission documents to him. (He must have grimaced when he heard the report's line about the necessity of "not discriminat[ing] against blind or visually impaired individuals.")

On June 8, the commission voted 6-2 to adopt the report. Thernstrom and Redenbaugh were the minority, and began to prepare a dissent. This is standard procedure at the commission: Members who don't agree with a report's conclusions are allowed to write dissents, which the commission then publishes as part of the official report. That's how it has worked for as long as anyone can remember, and that's how Thernstrom and Redenbaugh expected it to work this time.

They pored over the document for nearly three weeks and prepared a thorough rebuttal. The commission was uncooperative, refusing to provide them with the machine-readable data and regression tables Lichtman had used in making his calculations. It is customary for scholars to share this type of information-so that they can verify each other's work-but the commission chose to stonewall. "There is no disk, nor was there ever any disk or disks, of Professor Lichtman's data," wrote commission staff director Les Jin in a memo. This was tantamount to saying that the commission's statistical analysis was based on statistics that never existed. Thernstrom and Redenbaugh wound up spending days compiling their own data, even though they could not be sure it was the same information the elusive Lichtman had used. They received assistance from Thernstrom's husband, Stephan, a Harvard historian who pioneered the use of statistics in his own profession, and from John Lott of Yale Law School, who was conducting his own investigation into the Florida voting controversy.

The Thernstrom-Redenbaugh dissent was ready at the end of June. It ran about 60 pages and didn't mince words: The majority report's central claims had "little basis in fact," and the report itself was "a dangerous and divisive document." The dissent was an almost point-by- point refutation of the commission's work. Along the way, Thernstrom and Redenbaugh made fresh contributions to our common understanding of what happened in Florida: They showed, for example, that the incidence of ballot spoilage increased in counties whose election supervisors were Democrats, and rose even further in counties where that Democratic official was black. This datum ought to make the paranoid notion of Jeb Bush's masterminding a racist conspiracy look preposterous, even to those predisposed to believe it. Thernstrom and Redenbaugh also showed that while some predominantly black precincts saw high rates of ballot spoilage, factors such as literacy and voting experience were more useful in explaining the disparity-to the point where race qua race may not have played any role at all.