Mass. bests Florida, but it's not perfect

17 November 2000

For a state that counts for so much, Florida sure can't count. But don't laugh too quickly, Massachusetts. We're better, but we're sure not perfect.

I speak from some experience on this matter. In 1993 I won an election by a scant 27 votes. There was, of course, a recount. Now admittedly, this was for a Boston City Council seat, not for president of the United States. At stake was not the right to be leader of the most powerful nation in the world, but instead a really excellent parking space at City Hall.

Nevertheless, the election had been fiercely fought and the recount was filled with tension. There were errors, as it turned out, with each of us getting two more votes than originally counted. As luck would have it, I had preserved my 27-vote margin.

Others have not been so lucky. Most famously, perhaps, was Phil Johnston, who seemingly won his Democratic primary for Congress in 1996 by 288 votes. His chief opponent, William Delahunt, then forced a recount, which Johnston still won by 175 votes.

But it didn't stop there. The registrar of the town of Weymouth had ruled that 900 ballots were blanks. The voting system used in that election, called Votomatics, relied on the same punch-card ballots that have been at the center of Florida's current controversy.

Delahunt persuaded a Superior Court judge that any indentation on the ballot - a so-called "pregnant" chad - should be counted as a vote. A bunch of once-blank ballots were suddenly turned into Delahunt votes. He ended up winning by 108.

After the Johnston-Delahunt imbroglio, the Massachusetts secretary of state's office banned the Votomatics. The real flaw with that system, as we have now learned, is that a lot of ballots don't completely punch through.

Today cities and towns in Massachusetts rely on three systems for voting. About three-quarters use paper ballots, which in turn are read by optical scanners. Except for Lawrence, most everyone else, including Boston, Brookline, Everett, Waltham and Woburn, uses machines. Lawrence still uses a version of punch cards, called DataVote, that election officials say doesn't produce as many errors as the now-forbidden Votomatics.

Paper ballots dominate in Massachusetts because they are cheap and easy. The optical scanners that read them weigh only about 20 pounds and are easy to store between elections. The scanners are supposedly more accurate than the mechanical readers that scan punch cards. Plus, voters like paper. There's an old-time feel to it that harkens back to the elections we had in high school for class president.

Voting machines, on the other hand, are big, clumsy and expensive. They weigh 900 pounds each. They're hard to move around. And finding space to store them between elections is often difficult. They also can be confusing, with voters always worrying whether they have thrown the right lever. Still, machines are, in theory at least, unfailingly accurate. Unlike punch cards or paper ballots, voters can't accidentally vote for two candidates when they should have only voted for one. And there's no mistake as to whether a vote was intended or not. With punch cards, officials argue over whether a slightly distended chad was intended as a vote. With paper ballots, a stray pen mark or a crossout raises similar questions. Those problems don't happen with machines.

But while the machines may be accurate, people are less so.

After machine voting concludes, one pollworker will open the back of a machine, find the counter that corresponds to a particular candidate or ballot question, and read off the results. Another will write them down.

Mistakes are common. Pollworkers, exhausted after a 14-hour day, misread or mishear numbers. Three years ago in Boston, an at-large City Council election was recounted and the number of votes changed markedly. Even worse, pollworkers can look at the wrong mechanical counter. That appears to have happened this year when half of the ballot questions in Boston were incorrectly tabulated, meaning perhaps 20,000 votes were misreported. The secretary of state is now insisting on a new count.

And, of course, when it comes to handwritten absentee ballots, the potential for error grows.

The truth is, an error-free voting system has yet to be devised. Massachusetts, having banned the Votomatic machines, is probably in better shape than Florida. But mistakes still occur - and, scarily, they occur with no one knowing, especially because Massachusetts, unlike Florida, does not automatically require recounts in close elections.

When the election-that-won't-end finally ends (right now, there's the same level of pessimism about that occurring as there is about the Big Dig ever being finished), it will be time for a clear-eyed assessment of what works, what doesn't and how it can be improved. Until then, one imagines, most local election officials are eyeing Florida and thinking, "There, but for the grace of God, go I."