The best time by far to solve election problems is after the election. Not on election day, not one month before, and not three months before. You start, says Denise Lieberman of the Advancement Project, two years before an election–in other words, beginning the day after. Because if you wait until three months before the election, all the decisions have been made and all the systems have been put in place. Anyone who waits until a month ahead of time to complain about, say, e-voting machines would be better off going to the movies. Even if that activist could convince a local election board of the dangers of DREs, it would be too late to do anything about it.

The Advancement Project is a national non-profit civil rights organization based in D.C. It promotes research, analysis, and advocacy, all of which are activities that don’t lend themselves to the hustle and bustle of an election. Lieberman, who is the Project’s Missouri state coordinator, has been a civil rights lawyer in St. Louis since 1995 and is a political science professor at Washington University. She is experienced in the areas of research, analysis and advocacy. In fact, she just spent the last two years putting those skills to work by focusing on provisional ballots.

Her superiors chose that area for her and others to home in on because such ballots only came into common use recently and many wrinkles need ironing out. After the disenfranchisement of thousands of voters in 2000, when the names of voters were inexplicably missing from the rolls or deliberately purged from the rolls–not only in Florida but right here in Missouri as well–the Help America Vote Act (HAVA) of 2002 mandated the use of provisional ballots so that, in theory anyway, voters who faced such problems could cast a provisional ballot and officials could sort out later whether they should have been allowed to vote.

But the best laid plans, you know, gang aft agley, and so it is with provisional ballots. The theory has some flaws. One is that these ballots sometimes disenfranchise people who’d have gotten regular ballots under the old system and who’d have had their ballots counted. And besides, even the people who would not have gotten regular ballots may vote provisionally and still not get their votes counted when they ought to. Early in 2007, Denise and other Advancement Project attorneys from around the country were assigned to go down in the trenches and file public records requests so that they could get copies of every envelope containing provisional ballots in five critically important counties, including Cuyahoga County in Ohio. They did not want to rely on what state election officials said about the ballots. As she pointed out, of course they’re going to say that they’re only rejecting ballots of people that aren’t eligible. Her team wanted to know exactly why each ballot was rejected.

They worked with political science professors who were experts in geographic mapping. First, they mapped locations where provisional ballots were cast and then overlaid those maps with maps of which provisional ballots actually counted. Next, they overlaid those maps with census data and racial data.

“It turned out that provisional ballots were being disproportionately cast and disproportionately rejected in certain areas of the counties, areas that had higher minority and higher core low income populations. That was something that we were able to then prove, not just sort of speculate–you know, say ‘Well, it’s probably true that ballots of poor black people get rejected more often.’ And there’s some reasons we could easily speculate about why that’s likely to be true, but here we were able, we were, like, no, let’s actually sit down and read these ballots and then look at exactly why they didn’t count. And when you can look at the ballot, you can see, ‘Oh, this person actually was eligible to vote. The reason his ballot was rejected was because one of the election judges didn’t sign it.’ Right? And that’s just one example of the reasons ballots were rejected. There are half a dozen more.

(…..)

By combing through these boxes of documents, analyzing them in an academically credentialed, viable way, you know putting that analysis together, you then have an incredible picture of how this particular election policy is playing out.”

Lieberman would have liked to include St. Louis County in that seminal two year study. Unfortunately, she was unable to obtain the data. The St. Louis County Board of Elections refused to release the ballots. Nevertheless, she has been able to use the study in advocating for changes to the way provisional ballots are handled here–successfully, too. Since 2006, the County Board of Elections has reduced by half the number of provisional ballots issued, largely as a result of suggestions from Lieberman and the people she works with.

Furthermore, Lieberman sat down last summer and presented the results of the study to Robin Carnahan, who was interested in learning what the study showed–though she and Lieberman parted company on whether the state’s open records law requires that local boards of election turn over provisional ballots.

Sure, there’s still wide room for improvement, not only in how provisional ballots are handled in this state but on a host of other election issues. Lieberman doesn’t look like she’ll be losing a job for lack of work that needs doing. So just because her work isn’t being splashed across the front page of the Kansas City Star and the Post-Dispatch, don’t assume that progress in election protection has been stymied. It’s happening steadily, behind the scenes.