Farhad Manjoo:

I appreciate Greg Palast's response to my article criticizing his argument that Kerry won the presidency on Nov. 2. Unfortunately, though, I don't see anything new in his letter to bolster his claims; Palast's theory, as I see it, remains at best just that, a theory that Kerry would be declared the winner if someone would just take the time to count the "spoiled" ballots spit out by punch-card machines in Ohio.

Palast rests his claims on the fact that Republicans have long tried to suppress the votes of minorities -- a point on which I agree, and have documented. Because the GOP has attempted, in the past, to either keep black voters away from the polls or reduce the chances that their votes will be counted, Palast argues that we should assume that the same thing occurred this year in Ohio. Hundreds of thousands of African-Americans went to the polls intending to vote for Kerry, Palast says. But their votes won't be counted, and it's only for this reason that Kerry didn't win the White House.

Alas, Palast's is a theory unencumbered by much rigorous analysis. As I wrote on Tuesday, and as others -- including David Corn, of the Nation, and Daniel Tokaji, a professor of law at the Moritz College of Law -- have pointed out, there is precious little evidence to show that there are enough uncounted ballots in Kerry's favor to have given him the White House.

As Palast explains, there were about 250,000 uncounted ballots in Ohio as of Election Night. About 155,000 of these were "provisional" votes cast as a kind of last resort by voters whose names weren't found on the registration rolls at their polling places, and about 93,000 were "spoiled" ballots that weren't counted by poorly designed punch-card machines. Kerry is about 136,000 votes behind Bush in Ohio. For the senator to win the race, then, he'd need to win more than 77 percent of these 250,000 uncounted votes. Palast is certain that Kerry can meet this threshold. Kerry's only problem, Palast says, is that the "votes won't be counted."

But Palast is wrong on this score. Ohio's provisional ballots, which make up the majority of those 250,000 votes, will be counted. In fact, officials across the state have already begun counting them. As the Cleveland Plain Dealer reported on Saturday, the counting is particularly complex; officials must first determine whether each vote is a legal vote (Was the voter in the correct precinct? Did he cast a ballot elsewhere?), and only then do they tally the vote. Nobody is sure just how many of those ballots will be determined to be legal votes (officials have until Dec. 1 to complete the task), but in 2000 in Ohio, only 87 percent met this threshold.

Going by that rate for this year's votes, you'd get about 135,000 legal provisional votes in Ohio. How many of these will go to Kerry? As I said, we'll know soon enough; those votes will eventually be counted, and it'd be wise for us to wait for the final tally.

But Palast doesn't seem to want to wait. Instead, he's sure the provisional votes will break for Kerry, a belief he bases on a presupposition that they are concentrated in "minority -- that is, Democratic -- areas." So let's assume, generously, that Kerry wins 70 percent of the provisional votes -- that would give him 95,000 more votes, with Bush winning 40,000. That's a net Kerry gain of 54,000 votes, reducing his deficit from 136,000 to about 82,000.

Clearly, then, winning even 70 percent of the provisional votes wouldn't win the White House for Kerry. So how could Kerry make up 82,000 votes? That's where the spoiled punch-card ballots come in. Palast believes that Kerry would win if Ohio launched a hand recount of the 93,000 spoiled votes tossed out by punch-card machines. Everyone knows, Palast argues, that the vast majority of those votes were meant for Kerry, since punch-card ballots are most often thrown out in low-income and minority areas. Citing an ACLU study that revealed "a strong relationship between the racial composition of a precinct and the percentage of discarded ballots in that precinct," Palast says that "it is hardly a wild leap to discern which candidate got punched out by the punch cards."

Palast is right; it's not a wild leap to say that Kerry probably lost more votes than Bush did as a result of faulty punch cards. The Columbus Dispatch recently analyzed the discarded punch-card ballots from the 2000 race, and while it found that "precincts with the highest rate of uncounted votes are heavily Democratic," it also determined that even in those precincts George W. Bush won one-third of the votes cast. So it is a wild leap to say, as Palast emphatically does, that Kerry lost enough votes to have given him the White House.

Now, let's say that Kerry does even better than Al Gore did in those precincts. Instead of winning two-thirds of the votes there, assume he won 80 percent -- that would give him about 74,000 of the 93,000 discarded ballots, with Bush getting 18,600. Assuming Kerry also gets 70 percent of the provisional votes as outlined above, he still loses to Bush in Ohio by about 26,000 votes. For Kerry to win -- still assuming he gets 70 percent of the provisional votes -- he'd need to get something like 94 percent of the spoiled votes. Is that possible? Well, anything's possible. But it is exceedingly unlikely, and Kerry's concession is an indication of how unlikely it is.

I do not disagree with Palast's more general thesis that the American electoral system is deeply flawed. Even if there aren't enough uncounted votes in Ohio to overturn the election, it's disgraceful that 93,000 votes in the state will go uncounted. Just as troubling is that many voters in Ohio -- mainly concentrated in low-income, Democratic areas -- were forced to wait in long lines just to cast their votes. These problems must be addressed, and Salon will continue to press for reforms.

But a broken system doesn't mean the fix is in. At this point, there is simply no evidence to support Palast's claim that Kerry won Ohio.

- - - - - - - - - - - -

Farhad Manjoo is a staff writer for Salon Technology & Business.

Recent Stories