During the run-up to Nov. 7, my 7-year-old son, David, kept asking whether someone would win by just one vote. It happened once, David insisted. I patiently explained that this was inconceivable. Millions of votes are cast and it simply isn't possible to win by just one vote. And no, I stressed a bit haughtily, it has never happened in history.
Well, not only does this election seem to be coming down to just a few hundred votes, pretty close to what Mom confidently assured her son was impossible, but David was also correct, in a way, about the historical record.
In 1876, Rutherford B. Hayes, the Republican governor of Ohio, failed to win the popular vote against Samuel Tilden, the Democrat, though he did win the Electoral College. In that race, three Southern states, Louisiana, South Carolina and Florida, were contested. The disputed election went to the House of Representatives, which appointed a special commission to decide the outcome with delegations from the House, Senate and Supreme Court. After deliberations, Hayes was selected on a vote of 8 to 7.
So, in a sense, Hayes did win by just one vote. I think it was a mistake to teach David to read.
What have we provisionally learned from this bizarre election?
One: Every state in the union should move to computer machine voting -- no cards, thank you very much. Florida's system, in which the voter uses a machine to make holes in a punch card, is clearly open to abuse when subjected to a hand count. These ballots were never intended to be read by hand. The whole idea that election canvassers are attempting to divine the intentions of voters based upon the look of the chads is absurd. Some counties are ruling that "pregnant chads" are votes, others will only accept a chad that is detached on three sides. Still other counties will count only three-pointed chads.
Now suppose a nice middle-aged gentleman entered the voting booth, began to depress the Gore key and then changed his mind and voted for no one? The counters will more than likely add his card to the Gore pile, based on the indentation. This is a farce, and the only way to prevent its recurrence is to change the voting machines. (It would be nice if we could change the Democrats, but that's probably too much to hope for.)
Two: The television networks should be shamed into restraint about calling states for one candidate or another before there is certainty about it -- and certainly before all the polls in that state have closed.
On election night, the networks famously called Florida for Gore early in the evening, before the polls had closed in the panhandle and when the vote was clearly too close to call. Honest mistake? Perhaps. But it leaves dangling the question: Why did the networks delay calling a number of states for Bush that were not nearly as close?
Rep. Billy Tauzin assembled a list that makes interesting reading. Bush won Alabama, for example, by 15 points. The polls closed at 8 p.m., yet the state wasn't called for Bush until 25 minutes later. Bush won Georgia by 12 points. The polls closed at 7. Yet the state was not called for Bush until 7:59. Bush won Colorado by 9 points, Louisiana by 8, North Carolina by 13 and West Virginia by 6 -- yet none of these states was called for Bush until later in the evening, not until 10:46 in the case of West Virginia. Meanwhile, Michigan, Pennsylvania and Florida were being called for Gore, giving the impression of a building win for the vice president.
The networks and all the cable news channels should resolve never to call a state before all of the polls have closed in that state. Also, they should have one standard, not two, for calling states at all.
Three: States should reform their rules on absentee ballots to require that they be mailed at least one week, or preferably 10 days, before an election. That would obviate the problem of waiting, after a close election -- and who can say whether this is the last one we'll see? -- for the absentee ballots to filter in.