Election Machinery

Though it seems that the election year has come and gone, 2009 also promises to hold many elections outside of the U.S. that will affect our futures in ways we may not necessarily perceive. One of the more pressing 2009 elections is being held in the commune of world leaders known as the U.N. The security-council election is to be held in October of this year and is already a growing issue due to the change over of five seats that are non-permanent. These seats are already allocated to various continents, with two allocations for African countries, one for Asia, Eastern Europe and finally one for Latin America and the Caribbean. “The newly elected countries will replace Belgium, Indonesia, Italy, Panama and South Africa.”

A plethora of scenarios can occure with regards to this election, and many issues have begun to develop. Election to the security council is critical for many countries, and though regional endorsement is a common occurrence, (a region endorsing one country to essentially create an uncontested bid for the seat) “because these are elections to a principal organ of the United Nations, formal balloting is required” One of the two largest pressing issues surrounding the coming election are the battle for the West African seat on the security council, leaving Sierra Leone and Nigeria “currently locked in a diplomatic stalemate at the UN Headquarters”. The second, but no less important issue is Turkey’s bid for the Eastern European slot in an effort to “become a regional power through the UNSC”. Though history shows that the USA and Russia dominate the Security Council, these treasured positions can still escalate smaller powers within their respective regions, if not on a global scale.

Though permanent members on the security council enjoy certain luxuries the candidates for these seats do not, the voting system for these seats eliminates that sort of bias: It is very different from the system by which UN resolutions are decided upon, which allows any of the permanent members to effectively veto a resolution with just one negative vote. The influence of the permanent seats (China, France, Russia, U.K. and the U.S.) is essentially removed from the picture, because each seat is voted for by the based on the regional blocks. The election system incorporates a secret ballot for each of the regional voting groups, and allows for multiple rounds of ballots to occur in case there are ties between countries. In the 2008 election, all seats to change over were won on the first ballot, but with the situation in Africa continuing to escalate, whether or not this will be possible during this election year is up in the air. This election to one of the most influential seats in world government, that of the Security Council, the election system seems fairly well-balanced, and simple enough to consistently work. In conclusion, the world looks to New York, where the next General Assembly will be held, to see what will happen.

The Minnesota Canvassing Board in charge of the recount of over 2.8 million ballots yesterday declared Democrat Al Franken the winner of the Senate race, defeating Republican incumbent Norm Coleman by 225 votes. Coleman and his attorneys are expected to challenge the outcome in court before the results are certified. Meanwhile, in Washington, power players in the Senate itself, including Minority Leader Mitch McConnell (R-Ky.) and Majority Leader Harry Reid (D-Nev.), are preparing for a long fight regarding physically seating which man actually received the most votes. Unfortunately, since the 111th Congress officially began on January 3rd, convening for the first time today, and Mr. Coleman has not been certified the winner, his office has been ordered closed.

This Senate race encompasses almost every topic that we have discussed in class and has provided us with something to ponder well after Barack Obama’s anticlimactic (for voting machine enthusiasts, anyway) landslide victory. The Minnesota Canvassing Board is certainly not as comically unprepared as Florida’s was in 2000; they have done well in setting standards and keeping the press informed of their procedures. We can examine, though, the technology that Minnesota uses statewide in their elections (optical scan), and the voting system that the entire nation is so used to (plurality vote).

Statewide, Minnesota employs the ES&S AutoMARK Model 100 precinct-count optical scanner. As we have discussed in class, optical scan ballots are preferable to most other systems because they provide a physical record of each voter’s choices, preserving the possibility of a satisfying recount, and they are able to be tabulated quickly on machines like the Model 100. Unfortunately, optical-scan ballots are also pieces of paper, and, as Michael Shamos has pointed out in his report Paper v. Electronic Voting Records – An Assessment, pieces of paper can be altered and/or misplaced, making an optical scan recount potentially questionable. The Minnesota Canvassing Board has hand-counted every single ballot. The Board does not know, and cannot know, whether some ballots in Coleman’s favor were lost (or “lost”) or altered so that they would not be accepted. Their policy regarding which absentee ballots to accept has infuriated both campaigns at times as well. It is unfortunate that there can still be doubt about the winner after such a laborious recount process, but that is the nature of optical scan ballots.

The more basic “machine” that helped create this situation is plurality voting system, which we spent the last few class meetings discussing. Plurality voting works fine with two candidates, but, as numerous social scientists have discovered, it becomes imperfect when there are three or more candidates. In this Senate race, Coleman and Franken received 42% of the vote each, while Dean Barkley, an Independent, received 15%. It is Mr. Barkley’s right, of course, to run for office, but his presence on the ballot, as an option for Minnesotans tired of Franken and Coleman, probably skewed the outcome, resulting in an incredibly close tally and launching a statewide recount, political brawls in Minnesota and Washington, and countless lawsuits that will most likely delay any conclusive result for some time.

The lack of public activism and involvement that some feel seems evident in this country – represented by lack of increased turnout for even this past unprecedented presidential election – is a disease that does not affect the health of the global democratic system.

Sometimes if democracy isn’t that impressive day to day in our own country, we can just read the headlines and we’re easily inspired. “Ghana steps back from the brink,” “Wind of change,” and “Opposition leader wins presidency in Ghana,” these are the recent headlines from the global news agencies. As soon as Professor John Atta Mills won the recent presidential election in Ghana, “People danced outside the EC office with party flags, sounded car horns, pulled wheelies on mopeds and one man charged up and down the road on horseback.” (http://news.bbc.co.uk/2/hi/africa/7810191.stm)

Although both parties have traded allegations of voting fraud and intimidation, it’s important to note that the
Ghana’s electoral commission examined the evidence provided by the parties and has maintained the legitimacy of the election. “The Electoral Commission did not seem keen to become not just the referee, but also the police, judge and jury.” (http://news.bbc.co.uk/2/hi/africa/7810191.stm  Also the losing party has not caused any disruptions, nor has it put up any excessive resistance Instead, the losing party plans on taking legitimate pathways, and will contest the election, which was very close (only “the narrowest of margins: barely 40,000 votes out of 9 million, or less than 0.5 percent of votes”) in a court hearing. 

“Some observers say the simple fact the country’s institutions, especially its Electoral Commission, were able to cope with such a tense, tight race and ensure both sides respected the results, is proof of the deep roots democracy has in
Ghana.”

(http://blogs.reuters.com/africanews/2009/01/03/ghana-steps-back-from-the-brink/)

“Analysts say
Ghana’s [election] could help salvage the tarnished image of constitutional democracy in
Africa, after last year’s flawed elections in
Kenya and
Zimbabwe and military coups in
Mauritania and
Guinea. 
Ghana remains a rare example of a functioning democracy in
West Africa.” (http://news.bbc.co.uk/2/hi/africa/7809451.stm) Any time that an opposition leader wins, and there’s a transition of power between parties, it’s a fairly reliable indicator that an electoral system is functioning and legitimate, and also that Democracy is still fit.

My past blogs have been somewhat discordant in ideology. My last blog went as far as to say that in the future it may benefit us to abandon elections entirely. But I would like to address simple yet insurmountable appeal that Democracy has. A single election in any county, granted that a change in power is possible, holds the momentum and potential to change the future of the entire world.

These electoral systems that we’ve studied ultimately decided where the power is distributed and reassigned to. Wherever the power is, is where the decisions are made, and where the decisions are made, is where our future lies. So the decision of every citizen who votes, impacts the decisions that govern our future.

Though any possible tampering with the 2008 presidential election may have been overlooked because of Obama’s huge lead, smaller races are still being scrutinized by America’s dedicated watchdog organizations (i.e. BlackBox Voting) with some disconcerting results. Though error can be found with respect to the Obama election, not many of them have been significant enough to affect the results, even on a county level. One issue that has come to the attention of news-sources and watchdogs alike takes the form of a House of Representative election in, ironically, Florida. With regards to the presidential election, Florida took many precautions to ensure that there was no repeat of the failures of the 2000 election and generally did quite well with gathering information and reporting results. It is a likely hypothesis that all of the focus on the presidential elections left other races within the state exposed to possibly tampering, or simply more prone to unanticipated error.

The issue surrounds the race of Republican Representative David Rivera, initially elected to the House in 2002 and re-elected in the recent 2008 election. He represents Florida’s District 112, which encompasses most of Collier county, with parts of both Miami-Dade county and Broward county. Reports show he had one opponent, a woman by the name of Beatriz Gaffney, a candidate who switched her affiliation from the Republican party to the Democratic to competitively enter the race. “The Florida Democratic Party’s lawyer, Mark Herron, said he thought Gaffney’s candidacy was fake from the start. “ The paperwork surrounding her entrance into the race as well as her switch to the democratic candidate all traces back to Rivera’s party, one of the reasons Herron was so suspicious. When, three days before the race, she withdrew from the election, essentially securing the victory for David Rivera, Herron was convinced. Coincidentally, he could not take any legal stand against her before, could do nothing now that she had withdrawn.

The end results of the election had were as such: “though voters were provided with a notice of Beatriz “Betty” Gaffney’s withdrawal, it wasn’t enough to keep the 30-year-old political unknown from winning the Broward County portion of HD 112 by 4,424 votes. In Collier County, she came within 496 votes of beating Rivera, according to the Collier supervisor of elections office. In Miami-Dade, Rivera won by 9,901, according to the supervisor’s office.” With a little more support, this phantom candidate could very well have defeated Rivera in the Rep. election. Whether or not this was an elaborate hoax or a ploy by Rivera’s campaign to make the race seem competitive still remains unproven, and one must decide for themselves whether or not this story is ultimately too suspicious to overlook.

Last week, I took a break from walking in Princeton’s winter wonderland to speak with Mr. Robert Dwyer about his experience with voting fraud.  Here is part two of my interview with him.  (You can find part one here)

If Michigan politics was ugly, Illinois politics was infinitely worse.  Results from the north would sometimes be delayed until the results from the south came in.  When questioned about the delay, Mr. Dwyer told me the officials were sometimes surprisingly candid.  “They said they waited for the results so they would know how many votes they had to switch to make up for the votes the south stole.”

Blatant election fraud was not limited to the counting side of the procedure, however.  Mr. Dwyer told me about a process known as “voting the cop” that was practiced in Chicago.  In this procedure, the police officer who was on duty to protect the opening of the polls would vote first.  Instead of submitting his vote, he would leave with his ballot and give it to a political boss.  The boss would then fill out the ballot as he saw fit and send it in with the next voter, who would vote the ballot the boss filled out and bring out the blank one he was given in the polls.  In this way, one could completely control how people voted.

Mr. Dwyer has considerable experience with the election process, so he has an invaluable insight into the actual practice of elections.  He has been involved in recounts in Michigan, served as the chair of a congressional district, been county chair twice, worked on state-wide campaigns, and served as a member of a state central committee.  He has seen enough of American politics to know that fraud does occur, so it is unsurprising that he also has some opinions about how one might correct our current situation.

For Mr. Dwyer, our current election protocols are less than ideal.  How should we conduct elections, then?  Paper ballots are the best solution, he says.  In Mr. Dwyer’s opinion, a paper-based system works well and is the simplest method of voting.  Electronic voting, on the other hand, is a flawed solution to a problem that does not exist:  “We are consumed with the idea that we have to know at 8:30 who won.”  And what does he think of absentee ballots? “The absentee ballot is not a safe ballot; it’s not a secret ballot.”

Earlier in the semester we read the Paper Vs. Electronic Voting Methods assessment written by Michael Shamos. Although in class we were able to briefly analyze some of the many valid points made by Shamos, I re-read his piece and felt it could be revisited. Shamos concluded that although there were many security risks involved with DRE voting machines, paper methods did not adequately address these issues.

One of Shamos’s most interesting points comes in his comparison of a plane to a voting machine. One of the arguments against DRE machines is that “Voting machines are ‘black boxes’ whose workings are opaque to the public and whose feedback to the voter is generated by the black boxes themselves. Therefore, whether or not they are operating properly cannot be independently verified and the machines should not be used.” However, as Shamos explains, the software in many airplanes used today could in fact have malicious codes intended to cause harm, yet people still fly today. He continues to explain that a company that decides to build airplanes wants to manufacture safe planes. Similarly a DRE manufacturer would want to make reliable machines that don’t crash. He also explains that planes can be tested, and so can voting machines. As I have read in a previous Shamos piece, in this assessment he makes the claim that voting machines should be held to strict engineering regulations, similarly to how planes must meet certain safeguards and engineering standards. The issue of creating a well-designed voting machine should be seen as an engineering issue.

Another interesting point brought up in the assessment, is his opinions on voter verified paper trails. Many people have suggested that by adding a voter verifiable paper trail to DRE machines, that perhaps any suspected cheating could be detected and a reliable recount could be administered if need be. Shamos refutes this argument by stating if a machine cannot be trusted to a point in which a paper trail is required, then how can that same machine be trusted with handling the paper? Potentially, a machine could mark any ballot void and print out a different one, therefore defeating the entire purpose of the paper trail. He makes it clear that he is “firmly opposed to any audit trail requirement.” However, he does make it a point to acknowledge that if a company is able to effectively incorporate a paper trail into their system, he has no problem with it, as long the paper record does not “govern” over the electronic one.

From our discussions and other readings in class I, myself, feel that there is no real problem with having some form of an effective paper trail within DRE machines. As seen by some of the issues we addressed after our meeting with Kevin Chung of Avante, an effective paper trail may not be the easiest thing to produce. However, this may be an “engineering” problem. With more research and development an effective paper trail could adequately serve its purpose to act as an audit trail and potentially catch any problems within DREs (if they exist). Once an effective method is developed it should become a requirement to include such safeguards.

In the assessment, Shamos briefly addresses the idea of parallel testing. Although he himself called the purpose of the test “limited” he does feel that it can be used to ensure that the “nightmare scenario” does not occur. The nightmare scenario, as he defined it, would be that somehow, most of the DREs within an area have been tampered with. In Shamos’s rebuttal of the case brought forth by Andrew Appel concerning the safety of DRE machines, he stated that parallel testing could be used to ensure that machines are properly running. As we have discussed in class, parallel testing can be defeated through a few interesting methods. In a number of previous blog entries by R.S. Callowe, he explains numerous different methods of still avoiding detection; even in a scenario in which parallel testing is used. However, the real concern should be how often the practice of parallel testing is actually used. R.S. Callowe again addressed this issue in another blog piece, “Parallel Testing is nearly nonexistent in practice.”

Throughout the term, I have come across many different pieces written by Shamos. Although he does make many valid points, on some issues I have slightly differing opinions. However, his argument that voting technology should be perceived as an engineering issue is definitely one in which I agree. Just as research is done to create stronger bridges, or faster cars, or better computers, more research can be done to improve our voting machines. As I have discovered from this class, there will never be a “perfect” machine. However, more can be done to bring us closer to creating one.

Professor Ronald Rivest has solved the unsolvable problem.

How does one create a voter-verifiable audit trail, a receipt of one’s vote, without risking voter coercion? Any written proof of vote could be demanded by someone offering to buy or threatening in order to gain a specific vote, but no proof of vote means that a voter has no way to check that what he plugged into the machine was what actually went into the final tally– a clear conundrum.

Rivest has posed an answer: his end-to-end voting. Three ballots are provided to a voter, and <i>randomly</i>, the voter places two votes for his candidate of choice and one for the candidate he does not wish to win the election. One receives a copy of one of these three ballots at random. This copy will have either a vote for the candidate of choice, the opposite candidate, neither candidate, or both candidates. Hence, no one can use that copy of proof of the way in which someone voted, and it cannot be used as evidence for the sake of bribery or coercion.

End to end voting also has additional voter verification properties. The receipt has a unique ID number, and once all ballots are posted online, one can check to make sure one’s receipt matches what is posted. Voter verifications organizations can also make sure that the totals that would result from the posted ballots match the announced election totals.

Sounds good, right?

However, complications do ensue. First of all, there’s the difficulty of the process itself. In a sample procedure, a classroom of nine Stanford students armed with nothing but their wits and their sunny dispositions managed to become thoroughly confused. So how likely is it that voters who struggle so much with the current, simpler systems will be able to pull it off without trouble?

Princeton Professor Andrew Appel further critiques Rivel’s system. Unless one can memorize the serial number of all three ballots, only one can be double-checked. Additionally, although there cannot be direct vote-buying, there can be receipt-buying. If Candidate A purchases receipts with Candidate B’s name, he can change the serial number to the serial number of a receipt with his name on them, then challenge the accuracy of the vote. On a larger scale, Candidate A could bribe or coerce voters to produce receipts which state Candidate A. Because a majority of voters in precinct then have a receipt which shows Candidate A, if Candidate A’s campaign changes a small but decisive fraction of votes from Candidate B to Candidate A, those who voted for Candidate B but kept a Candidate A receipt to receive a bribe cannot prove their vote has been changed. Assuming the majority of those not affect by the bribe within Candidate B’s  supporters do not bother to check the results against their receipts, only a tiny fraction of people could possibly notice and draw attention to the switch.

On a final note, so the receipt is wrong- what can we do? Who can we hold accountable? How many incorrect receipt will it take until we decide to re-do an election?

Yet maybe the importance of Rivest’s suggestion isn’t whether or not ThreeBallot voting could be implemented practically. Maybe the importance lies in the idea that what seems to be an unsolvable problem could potentially have a solution. Rivest’s idea may have holes, but the next may not.

I recently had the opportunity to sit down and talk with Ed Kessler, a long term resident of New Jersey, and an eye-witness to some very startling local New Jersey voting fraud.

How would you define voting fraud?

“As restricting the voters vote for other purposes; diverted or stolen completely, or purchased.” 

How prevalent is voting fraud?

“If you asked me that six months ago I think I would have given you a different answer. I don’t believe there was much in this past election. The Obama Campaign had teams of people country wide to monitor polls.” 

Are there other ways of causing voter inaccuracies or other ways of affecting “the vote”?

“Yes, restricting people from voting in the first place. This is done through scare tactics and trying to disenfranchise people, generally based on profiling.” 

What kind of election abuses have you witnessed in the past?

Mr. Kessler was an eye-witness to some very interesting instances of voting fraud. The first instance was as a young child in the 1930’s when he went to the polling place with his father. There were no voting machines during this time. “I watched people personally take the paper ballot into the closed area to vote and then fold up their ballot, leave the booth, and deposit it into a ballot box. This is what my father did. However, some people would come out of the booth with their ballot unfolded, would have it inspected, and then would put it in the ballot box. When I asked my father why? I was informed that these people either worked for the city or had relatives who did.” The inspection was to ensure that the voters had marked their ballots ‘correctly’.

The second instance of voting fraud that he was a witness to occurred when Mr. Kessler was a challenger at a polling place during the June 16, 1970 Newark mayoral election between Addinizzo (a white male) and Gibson (an African-American male). During this election, as a precaution, nearly 600 policemen were deployed to various polling places to help protect the polling places from voting fraud. Mr. Kessler stated clearly that the law enforcement officers were of very little help. When he needed their assistance, the officers taunted him by calling him “chicken” and provided no help. Another display of voter intimidation was observed by Mr. Kessler when he first arrived at the polling place early that morning. He was welcomed by a presentation of seemingly unlimited Black Power pamphlets glued to the windows of automobiles aligning all sides of the street. This act was the creation of Addinizzo supporters and was meant to significantly disenfranchise Gibson voters and create a negative environment for any voter inclined towards a Gibson vote. The next act of suspicious behavior occurred when Mr. Kessler asked the election workers if he could inspect the counters on the machine. The election workers responded to the request by bombarding Mr. Kessler with questions about where he was from. They claimed that Mr. Kessler was not allowed to inspect the counters because he was not a local resident. Mr. Kessler insisted that he was permitted to because the election was run by Essex County, which he was a resident of. The election officials reacted by calling Mr. Kessler a “Carpetbagger” and a “Commie”. More challengers arrived and they were welcomed with the same type of treatment that Mr. Kessler received. All of the challengers were eventually herded into the middle of the gymnasium and were surrounded by the ‘goons’. A young woman challenger identified herself as being from the district and she was firmly denied the right to challenge. At this moment, the ‘goons’ started to get rowdy and began jabbing and kneeing at the challengers; while the police officers, whom were deployed to prevent such situations, stood by only observing. Mr. Kessler was able to escape from the mess –alive- and returned home to his family (it was still very early in the morning). He contacted the U.S. Marshalls and informed them of the situation at the polling place. They at first ignored the threatening situation, but after much persuasion on Mr. Kessler’s part, they sent a Marshall to investigate. Later in the day, at around 3 P.M., Mr. Kessler received a phone call from the U.S. Marshall’s office. They reported that they found “gross irregularities” at the poll and that if the “election is a close race, all the votes from Vailsburg HS will be thrown out and not counted.” Gibson won the election by 12,000 votes (55,000 to 43,000).

How does our government create confidence with our voting processes? “By running a FAIR election procedure.”

This blog will be devoted to looking into whether or not it would make sense to adopt the Australian practices of compulsory voting and IRV in US elections.

Compulsory Voting:
Since 1924, Australia has been requiring eligible voters to vote in elections. Because of this, voter turnout has increased from around 40% (in the 1920s) to 95%. Voter turnout for most national American elections is usually under 60% (2008 had an exceptionally high turnout) of the eligible voting age population; so making voting mandatory could potentially increase the amount of voters by 50%. This increase would result in an additional 50-100 million votes, which could potentially change the outcome of elections.

Compulsory voting would make America a more democratic place since more people are voting, right? The answer is not so simple. Yes, more people would be forced to vote, but the fact that they were forced may violate democratic principles or even the Constitution. A New York Times article asserts that compulsory voting would be a tough sell in the US because “we value the freedom not to vote”. Many Americans just simply choose not to vote, and although in a compulsory system they could just go to the polls and vote “none of the above”, it seems as though the have lost some liberty by being required to vote.

Another issue with the potential of American compulsory voting is the logistics behind it. Australia spends a great deal of resources going over excuses of people who didn’t vote to make sure they were legitimate. The US is a great deal larger than Australia, so the logistics of requiring everyone to vote could get messy.

IRV:
IRV seems like a legitimate improvement to US’s plurality-voting system. IRV stops the spoiler-effect (for the most part), and allows voters to express their support for a third (minor) party candidate without their vote “not counting”. However IRV is not a perfect system. IRV is non-monotonic; meaning that voting a candidate higher can potentially cause that candidate to lose. Here is an example of the problems with IRV. When voters are dishonest, and “game” their vote in order to help a candidate win (instead of ranking their candidates honestly), IRV becomes a potentially disastrous system.

Would IRV work in the US? For the amount of resources and time it would take to change current voting logistics (machines, procedures, education of voters, etc.) to allow IRV, I don’t believe it would be worth it. IRV (though at most times an improvement over plurality voting) is far from a perfect system. Its complexity allows for more convoluted paradoxes.

Group Voting Ticket:
The majority of Australians vote along a GVT (Group Voting Ticket). Since Australia uses IRV, the GVT automatically ranks candidates from every race along a preset list that the political party decides on. North Carolina employs a similar form of “ticket voting”, which actually sparked controversy as it led to a big under-vote percentage for presidential races. However due to awareness of this issue and voter education, this trend was reversed in the 2008 presidential election.

I don’t think we should adopt a plurality GVT. As seen in North Carolina, this could potentially confuse a substantial amount of voters and lead to an under-vote. The bottom line is that the convenience is not worth the trouble. GVT seems to be more useful in Australian IRV, because voters may not know about certain candidates and thus not know where to rank them.

I recently interviewed Professor Frank Askin, Professor of Law and Director of the Rutgers Law School in Newark. He also works with the Constitutional Litigation Clinic.

Q: How busy was the 2008 Presidential Election for you?
A: This past Election Day was very busy. Sixty students worked at the Essex County Court House, with eight faculty supervisors. We were able to obtain 150 orders to vote for people.

Q: What was the most common reason people were unable to vote?
A: The bulk of the people were ones who had mailed their registration or filled one in on the street. Many of these were never processed. The New Jersey Board of Elections was so overwhelmed this year with new registrations that many fell between the cracks. Many people came to courthouse, but we don’t know what percent of total problems this represents- maybe 10%? Once they arrived at the courthouse, my students took them to a judge. If judge was satisfied they had made an effort to register, he gave them an order to vote.

Q: Have you seen much progress since you started?
A: We started 30 years ago. Our busiest year was 2004. We’ve seen progress since then. Some of the holes have been plugged up. The second largest problem comes with people who went to the DMV to get or change their driver’s license. Under federal law, when someone comes to the DMV, the DMV has to ask if they want to be registered to vote at their address. If someone is unregistered to vote, but says “I went to DMV and they didn’t ask me”, they have a viable court case because the DMV failed to do their duty. This year, for the first time, there was an agreement between the DMV and public advocates. The DMV promised that, starting in June, that they’d ask everyone if they had registered. As a result, there are already fewer cases of this kind than in the past. However, other public agencies, such as welfare food stamps, or unemployment, have the same responsibility to voters as the DMV, but have not made an effort to inform people. Cases with these agencies similar to the ones can be presented before a judge.

Q: Have you had many problems with provision ballots?
A: Under HAVA, when someone moves within the same county but fails to notify the BOE, they must be given a provisional ballot. This has cured some problems, but has created others. For people who have already mailed in their registration, they’re not yet officially in the books. They’re allowed to vote by provisional ballot, but if their registration is not processed correctly their provisional ballot is thrown out. The provisional ballot is really a placebo- About 1/3 of all provisional ballots in the country were discarded.