What the Supremes don't know about Florida.

What can you think, though, when the U.S. Supreme Court hands down a ruling so transparently incoherent that one does not even need any special understanding of law to grasp its vacuity? The central rationale underlying the Court's decision to forbid Florida from counting ballots by hand is " equal protection." Counting different ballots according to different standards, the majority argues, unfairly dilutes the power of some citizens. Forget the legality of this finding--which every lawyer not affiliated with either George W. Bush's legal team or the Supreme Court seems to consider absurd. As a purely analytical matter, the Court's logic holds together only if you know virtually nothing about how the Florida vote was conducted.



The starkest violation, of course, lay in the different kinds of voting equipment used across the state. As anybody who has read the newspaper accounts now knows, voters in counties that used optical-scan ballots were far less likely to have their votes go unrecorded than voters who reside in counties using punch cards. In Florida, three out of every 1,000 optically scanned ballots recorded no presidential vote (that is, they were "undervoted" ), compared to 15 of every 1,000 punch-card ballots. A small portion of this differential could be explained by demographics--voters in punch-card counties may be less experienced, or have less physical strength, or face more chaotic voting stations--but statistical analysis shows that it's overwhelmingly the fault of the machines.

This discrepancy more than accounted for Bush's margin of victory. If you imagine that the punch-card counties had the same rate of undervoting as the optical-scan counties and that the extra punch-card votes broke down along the lines of each county as a whole, then Gore would win Florida by some 1, 500 votes. (This, incidentally, belies the cliche that "we'll never know who won" since the vote differential "fell within the margin of error." It's not true, because the margin of error--namely, bad machines--disproportionately disadvantaged one candidate.)

Here's where the hand count came in. By taking ballots that the machines did not read and turning them into actual votes, it cut down the disparity between punch-card and optical-scan counties. In a statewide hand count, all counties would gain votes, but the counties with the most undervotes would, rightly, gain the most. The leveling effect was slight. Palm Beach had more than 10,000 undervotes--2.2 percent of its total. The manual recount turned only about 800 of them into votes, reducing the county's undervote rate to just under 2.1 percent. Even Broward County, which used the loosest standard for reading ballots, identified only one-quarter of its undervotes. After its permissive manual recount, then, Broward still had almost three times as many undervotes as a typical optical-scan county. Broward voters might have ended up with more voting power than Palm Beach voters, but they still had far less than, say, Seminole County voters.

Was it fair that some counties gained more than others? Of course not. The question is what to do about it. If you were to apply the equal protection standard, you would probably want to make every county use the permissive Broward County standard--which would best alleviate the discrepancy in voting power between voters of different counties. Or maybe, reaching a bit, you would decide that dimpled chads don't have a sufficiently clear precedent in Florida law--or don't indicate a sufficiently clear preference--so you would instead impose the strict Palm Beach standard statewide.

But the Court did neither. Instead, it intervened with the least equitable solution possible. The majority--with unfathomable moral obtuseness--applied the equal protection standard not to the underlying inequality but to the remedy for the inequality. Some of the orphans are receiving more porridge than others: Let's cut them all off!

Of course, as a matter of constitutional stability, you have to respect the legitimacy of a Supreme Court ruling you disagree with. But you'd like to think that when the justices rule, even tendentiously and wrongly, they have based their decision on a coherent theory. That didn't happen this time. Maybe the justices in the conservative majority simply groped about for any plausible rationale for handing the election to their preferred candidate. Or maybe they didn't closely follow the media reports of the Florida recount and the abbreviated testimony never imparted to them a decent understanding of the facts of the case. The first possibility is frightening; the second, merely depressing. I'll try to believe number two.