Brooks Jackson examines mechanics of voting machines
CNN Correspondent Brooks Jackson discusses what he calls the "dirty little secret" of elections: That vote-counting machines have been known to have inaccuracies through the years.
Q: Is it common to see problems with machine counts in elections?
JACKSON: (There are) inherent inaccuracies in any sort of machine count. Like all machines, there are problems with those used to count votes. I can find nobody -- not the people who sell the machines, not the people who observe elections -- who will say that machine counts are 100 percent accurate every time. It's quite the contrary.
There are a number of ways that inaccuracies can creep in, especially with punch-card ballots. Some people recommended getting rid of these punch-card ballots a dozen years ago, because they are inherently prone to problems like we are seeing in Palm Beach County.
Q: To many Americans, this whole revelation has been surprising -- that without a voter even knowing it, his or her vote may not be getting tallied. What are we to take from this?
JACKSON: It has been one of the dirty little secrets of election administration over the years. These problems are always there. They only matter in very close elections, and so the problems crop up rarely.
One expert quoted me what he called the 'Election Administrator's Prayer' and it goes like this: Oh, Lord, let whoever wins win in a landslide.
That's because when you have a landslide, if there are problems with one-tenth of one percent of the vote, it doesn't matter. But now it does matter, and it matters more than it has ever mattered in any election certainly in my lifetime and maybe ever. I mean this is for the most powerful office on Earth.
Q: Are manual recounts the answer? There are many who would argue that hand recounts are open to fraud, whereas you're not going to get that from a machine even with occasional glitches.
JACKSON: There's no question that humans can commit fraud and try to steal ballots and taint results. There's always that danger. I think that's probably pretty remote in this case because the whole world is watching. And certainly, the whole Republican party is watching the Democratic observers and vice versa.
The area of subjectivity is very interesting. In these punch-card ballots, judges - often state Supreme Court judges - have taken a more liberal standard than even the Palm Beach canvassing board has taken on trying to determine the intent of a voter. What's at issue in this case is something along the lines of 10,000 ballots where the machines detected no vote for president. Maybe it's true that 10,000 people in Palm Beach didn't like either of these guys, but typically you skip voting for a county assessor or lower offices that you really don't care about. Most everybody votes for president.
There is now some good evidence that people actually tried to punch a hole in one of these punch cards, and the chad dangled by one or two strings. (In my research of cases in other states), the fact is that in several states Supreme Court justice have counted bulging and pregnant chads in some recounts. One county level race in South Dakota got judged a tie on the basis of one disputed bulging chad.