You may remember that in the battle over Pennsylvania's voter ID law, state attorneys acknowledged that they knew of no investigations, prosecutions, or even incidents of in-person voter fraud in the state.

But maybe they missed one.

When I interviewed State Rep. Daryl Metcalfe, the prime sponsor of the law, for Fresh Air last week, I asked him about the state's filing, and if he believed people were really impersonating others at polling places to steal votes.

"Yes," he said.

"Yeah, actually, State Representative Bernie O'Neill serves in the Pennsylvania House of Representatives, and during our debate on the floor, he testified to the body that he went to vote as an elected official even, and when he showed up to vote, somebody had already voted in his place and signed on his line and signed his name in there," Metcalfe said. "So yes, it happens."

A criminal at large

I've called Rep. O'Neill's offices four times since last Thursday, trying to get him on the phone to talk about the theft of his vote.

It strikes me as pretty brazen for someone to walk into the home polling place of a state representative and pretend to be the man himself.

This criminal was somehow certain that the local election board wouldn't recognize O'Neill. Or were they in on the heist, making this a conspiracy to commit in-person voter fraud?

Or is it possible that a poll worker made a clerical mistake, and had some other voter sign in on the wrong line in the election materials?

I'm dying to ask Rep. O'Neill what election this occurred in, and what he did about it. Did he ask the county board of election to investigate? Did he call 9-1-1, the district attorney's office or the FBI?

Or was the only remedy to enact a law that will now require the Pennsylvania Department of State to issue special voting ID's to thousands of citizens?

If you know Rep. O'Neill, please ask him to call me, so we can get to the bottom of it.

Does it even matter whether vote fraud happens?

I've written several pieces about this controversy already, and I hold to the notion that in election law, you have to weigh the benefits of increased ballot security against its potential to limit ballot access.

And the evidence seems compelling that there's precious little in-person voter fraud happening, and potential for affecting a lot of qualified voters' access to the ballot – not as many as critics say, perhaps, but thousands.

But the courts apparently don't regard the question of how much voter fraud is happening as terribly relevant.

On this question, Commonwealth Court Judge Robert Simpson relied heavily on the U.S. Supreme Court's reasoning in it's 6-3 decision upholding the Indiana voter ID law.

From Judge Simpson's opinion:

"To that end, the Court found the statue was a reasonable means of preventing the occurrence of in-person voter fraud. The Court also rejected the argument that the state's interest in preventing voter fraud was "illusory" because there was no significant evidence of such fraud. The state legislature was not required to "prove" that significant in-person voter fraud existed before it could permissibly act to prevent it. Id. at 458. "The United States Supreme Court has explicitly stated that 'elaborate, empirical verification of the weightiness of the State's asserted justification is not required. Rather, a state is permitted to take prophylactic action to respond to potential electoral problems..."