2008

Only in Ohio

by

By Ruth Colker

In a Simpsons episode, Homer Simpson experiences problems with the voting machine and exclaims: “This doesn’t happen in America! Maybe Ohio, but not in America.” My 62 hour saga of trying to get Ohio election officials to accept my properly cast absentee ballot is almost as entertaining as a Simpsons episode. Unfortunately, my story is true. A 1997 clerk-typist error in recording my birth date – which I corrected a decade ago – caused Brian Shinn, the Assistant General Counsel for the Ohio Secretary of State, to declare my absentee ballot “void.” After dozens of hours of advocacy over three days, I finally convinced local officials to accept my ballot. Slavish adherence to hypertechnicality does not protect against voter fraud, wastes the time of voters and election workers, and undermines the public’s confidence in the voting system. These rules must be changed in Ohio and elsewhere.

Here’s my Simpsons episode.

I moved to Ohio in 1997 and registered to vote under the Motor Voter laws. Unnoticed by me, a clerk erroneously typed my birth date as 1958 instead of 1956 on the applicable form. I discovered that error a year later and took my birth certificate to the Bureau of Motor Vehicles (“BMV”), which corrected its records. I was unaware that BMV would not send that correction to the Board of Elections.

Meanwhile, I voted without any difficulty in person or by absentee ballot in nearly every election held in the state of Ohio from 1997 to March 2008. In the fall of 2008, I completed an application for an absentee ballot for the November election and, upon receiving that ballot, completed and mailed it to the Franklin County Board of Elections.

On the morning of October 29, I checked the Board of Elections website to check the status of my ballot and learned that it was not listed as “received.” Concerned that my ballot had been lost in the mail, I immediately called the Board of Elections and spoke to an election worker who said: “What’s your birth date? Ballots are often not shown as received because of birth date errors.” She then found my individual ballot and confirmed that it had arrived at their office but was not recorded as received due to a birth date discrepancy. She told me not to worry, that there was another layer of screening and my ballot would probably be accepted.

I then checked the rules for handling absentee ballots. I learned that Ohio law permits a board of elections to reject an absentee ballot application due to a birth date mismatch. In that case, however, the board is required to notify the voter and provide an opportunity to correct the inconsistency. In my case, an election worker did not notice the discrepancy on my application for an absentee ballot. Instead, an employee caught the mismatch only after the board received my absentee ballot. Under Ohio law, that inconsistency was not a basis for failing to accept a ballot, so long as, like my ballot, it contained the voter’s signature, address, and last four digits of his or her Social Security number.

Because I concluded that Franklin County was not following Ohio law by failing to accept my ballot automatically, I contacted both the Secretary of State and the Franklin County Board of Elections to correct the error. I did not want subjectivity and human error to cause some ballots with mismatched birth dates not to be counted.

On October 30, after two days of haggling with Franklin County and the Secretary of State, I received a letter from Assistant General Counsel, Brian Shinn, who declared that my “ballot is void” because the “birth date that you listed on your absentee ballot request did not match the birth date in the records of the board of elections.” Shinn urged me to work with Franklin County to find a way to vote in the November election.

On the morning of Friday, October 31, Michael Stinziano, Director of the Franklin County Board of Elections, called to tell me that he found Shinn’s letter “persuasive.” I argued that Shinn was incorrectly interpreting Ohio law and Stinziano said he would get back to me later that day after receiving a copy of my birth certificate.

At 10:30 p.m. on Friday, October 31, I received an email from Stinziano saying that my ballot would “be processed as part of the unofficial count because no fraudulent intent existed and you have in fact updated the records with the BMV many years ago.” Sixty-two hours after I first called the Board of Elections, I succeeded in having my properly cast absentee ballot placed in the “accepted” category to be opened and counted on election day. Few voters have the time or resources to engage in such efforts to have their vote counted.

This kind of pointless obsession with birth date mismatches at the absentee ballot voting stage needs to end. All it does is undermine voter confidence. One mother told me that her daughter got her Ohio driver’s license a few days ago and was surprised to learn, when she got home, that her birth date was listed as 1978 rather than 1990! Her daughter had no idea that, after she corrected the record, she could be precluded from voting by absentee ballot under Shinn’s view more than ten years from now.

I hope, in the future, that we can say “this doesn’t happen in America. Not even in Ohio.”

3 Things I’m Watching For

by

As I reflect over this last weekend before November 4, these are the three issues at the forefront of my mind:

1. Will Demand Exceed Supply?

2. Will the Number of Uncounted Ballots Significantly Exceed the Difference in Counted Ballots?

3. Will a Lack of “Paper Trail” Undermine the Outcome in Any Critical State?

As I reflect over this last weekend before November 4, these are the three voting administration issues at the forefront of my mind:

1. Will Demand Exceed Supply?–especially in a “no early voting state” like Pennsylvania. To me, the most worrisome threat on the horizon from a national perspective is the potential that states won’t be able to handle the volume of voters on Tuesday. If this happens, at a minimum we can expect to see a request for further federal court relief in Pennsylvania and Virginia (as is already occurring), and extension of polling hours (perhaps administratively, perhaps judicially–perhaps statewide, perhaps locally). Even worse, what if at the end of the day Tuesday, there is a sense in one or more battleground states that thousands of voters have been unable to cast ballots despite trying–waiting in line for several hours but eventually giving up because of child care and/or work responsibilities–and that these “missing ballots” might have made a difference in the outcome?  One really doesn’t want to contemplate what, if any, remedy might be possible in that situation, so let’s fervently hope it doesn’t happen.  UPDATE: More reports today of the potential for this kind of problem in Pennsylvania.

2. Will the Number of Uncounted Ballots Significantly Exceed the Difference in Counted Ballots? Provisional ballots probably will remain the biggest sub-category of uncounted ballots that could hang over Tuesday night’s reports of who’s ahead in counted ballots, but it is not the only sub-category to look at. Here are two other sub-categories: (a) disputable absentee ballots; and (b) “residual” optical scan ballots for which the readers arguably failed to record the voter’s marking of the oval. Over the last 24 hours or so, reports from various battleground states (Colorado, Florida, Indiana, and Ohio) concerning significant numbers of absentee ballots that appear problematic—and therefore cannot automatically be counted—raise the possibility that this sub-category of uncounted ballots could become an important factor, in addition to provisional ballots, in obscuring the outcome on Tuesday night. Likewise, as the recent case from New Mexico reminds us, it is still possible to fight over “residual” optical scan ballots that the machines did not record as containing a vote. While the residual vote rate for optical scan ballots is lower than for the now-replaced punch card ballots, it is not zero, and it too potentially could become a factor. The comparison of the number of disputable uncounted ballots (in all sub-categories) to the difference between the candidates in the counted ballots is, perhaps, another way of invoking the 2004 question: is the margin of victory larger than the margin of litigation? But it helps me to focus on what the “margin of litigation” might consist of, and here I’m pointing out three distinct sub-categories of uncounted ballots that could be the basis of a dispute.

3. Will a Lack of “Paper Trail” Undermine the Outcome in Any Critical State? Probably not, but it’s a vulnerability that’s been known about for a while, which many states have addressed, but some (including Pennsylvania) have not.

Also, when it is comes to voting administration, it is always prudent to expect the unexpected.  Consequently, an additional non-specifc thing to watch for on Tuesday is how well a state or locality handles an unanticipated situation if one occurs.  Has the state or locality put in place some kind of nimble “rapid response” procedure that can prevent the unplanned-for event from becoming worse of a problem that it otherwise might be?

Post-Election Contests: Four States to Watch

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

Recently, friends of mine have been speculating about whether there might be a legal contest to the November 4 presidential election like the litigation that happened in 2000. To me this type of contest seems exceedingly unlikely because, if actual voting even somewhat mirrors the most recent polls, any legal victory in any particular state would be only symbolic and would not change the result of the election. Furthermore, the idea of a series of election contests being brought across the country to change the balance in the electoral college seems facially absurd. Instead, the real legal story that comes out of this presidential year, if a coherent story comes out at all, is likely to center on a series of important, though lower profile, contests for governor, US Senate, or House of Representatives. Here are four states to watch.

Ohio. You probably have already heard all the allegations of fraud in registration and early voting, but what you may not realize about Ohio is that it is home to no less than 5 hotly contested races for US House. One of these races, in the 2nd district, resulted in a recount in 2006, and the same candidates are running again this year. Ohio’s confusing provisional balloting statute almost led to post-election litigation in another close House race in 2006, and as a practical matter that confusion probably still remains, although theoretically resolved (see here and here). A combination of high turnout and wordy ballot issues (election officials in the state’s largest county predict it will take 18-to-23 minutes for the average voter to complete their ballot) could lead to excessively long lines and suits to extend polling hours. Because federal law requires any votes cast after regular hours to be cast on a provisional basis, that will only increase the opportunity for dispute. Provisional balloting is very common in Ohio even without these types of extensions. If a candidate for one of these House races wants to contest the result, it will be messy, because Ohio law currently attempts to deprive the state courts of any jurisdiction over such matters.

Florida. Like Ohio, Florida has 5 House races that are expected to be close. One of those races, in Florida’s 13th district, is a re-match of the election in 2006, which led to a real smoker of a recount and an unsuccessful election contest based on the allegation that touchscreen voting machines had failed to record votes. The lingering bitterness over that fight is probably enough on its own to send the candidates back to the courts if the race is close. Furthermore, at least one Florida House district, the 21st, includes some territory with a history of absentee voting fraud (see this report, starting at page 11). Florida’s well-publicized matching controversy and difficulties achieving adequate throughput in early voting locations may foretell of the kind of election day snafus that could form the basis of an election contest. However, Florida’s provisional balloting rate is low.

Pennsylvania. Pennsylvania, too, has 5 House races predicted to be tight. None of these ended up being close enough to induce recounts and lawsuits in 2006, but there is always a chance. Adding to that chance is Pennsylvania’s heavy reliance on controversial touchscreen voting devices without paper trails and potential confusion flowing from two last-minute legal fights, one seeking to stock dysfunctional polling places with paper ballots and the other seeking to force officials to identify suspicious registrations and force voters to cast provisional ballots. Pennsylvania also has a substantial history of absentee voting fraud, particularly in Philadelphia, though the phenomenon has recently appeared to be in hibernation. See this report, p 8-9 and footnotes 74-76.  There was an important House recount in Pennsylvania in 2006 in the 8th District, but that district apparently this year is not as closely contested.

Georgia. Georgia is expected to have 2 tight House races and, along with Minnesota, North Carolina, and Mississippi, currently has one of the closest US Senate races in the country. Unlike those other states, however, Georgia has experienced a significant amount of elections litigation, including significant lawsuits over matching and voter ID that could affect conduct of the election.

Other states to watch include North Carolina and Washington state, both of which are experiencing close gubernatorial contests (the one in Washington is a re-match from 2004, when it resulted in an ugly lawsuit). However, my impression from research and talking to officials is that North Carolina has one of the better-run elections systems in the country, and Washington’s Secretary of State claims to have vastly improved their system by, among other things, cleaning up voter rolls that permitted many ineligible felons to vote in 2004. Other states with House races expected to be close, and which also have experienced significant administrative controversies, include Colorado, New Mexico, and Virginia.

Is any candidate actually going to file an election contest? Well, election contests are pretty rare, both because very close races are rare and because courts hardly ever grant in an election contest any sort of meaningful remedy even when a very close race is marred by serious problems (the FL-13 litigation and Washington gubernatorial race of 2004 are good examples). Nevertheless, the high concentration of tight House races in states that just so happen to have troubled elections systems seems to suggest a higher than average risk that at least one election contest will be filed in a major race. That is still not a very high chance, but high enough, I hope, to make this article somewhat worth reading.

McCain v. Obama: What’s Next?

by

By Chris Elmendorf

Professor of Law, U.C. Davis School of Law

Ned Foley’s McCain v. Obama simulation has now yielded an opinion that Foley rightly describes as “thoughtful and rigorous”—one worthy of careful consideration by real courts deciding real cases on analogous facts. This Comment briefly addresses the most significant features of the opinion’s legal reasoning, and then turns to the question of what we may learn from the simulation about specialized, bipartisan election tribunals.

The Law of McCain v. Obama

I agree with Foley that the “the most instructive aspect of the Court’s decision” is its Equal Protection analysis, and that will be the focus of my remarks here.

At oral argument, Glen Nager, counsel for McCain, pressed the thesis that accommodating snowstorm-affected voters in Denver County with a polling-hours extension, without giving an equivalent extension to similarly affected voters in other counties, was presumptively (or perhaps conclusively) unconstitutional. The Court did not think this inequality worthy of strict scrutiny. To explain why, it quoted Justice Scalia’s concurring opinion in Crawford v. Marion County Board of Elections, 128 S.Ct. 1610, 1626 (2008): “[W]eighing the burden of a nondiscriminatory voting law upon each voter and concomitantly requiring exceptions for vulnerable voters would effectively turn back decades of equal-protection jurisprudence.” McCain at 7. Like Justice Scalia, the McCain Court thought that the burden of voting requirements should be assessed from an aggregate, society-wide perspective, rather than from the point of view of individual, adversely affected voters. This seemingly arcane distinction will determine whether the federal courts get into the business of crafting “as applied” exceptions to voting laws. In Crawford,Justice Stevens (joined by Roberts and Kennedy) invited such lawsuits; Justice Scalia (joined by Alito and Thomas) said he would not entertain them; and Justices Souter, Breyer, and Ginsberg remained mum. It will be most interesting to see whether state attorneys defending voter ID laws against as-applied challenges trot out the unanimous McCain v. Obama opinion in defense of their position.

One senses that the McCain Court might have responded more sympathetically to an “individualized” equality claim if Colorado’s regulatory scheme had itself provided for individualized determinations and accommodations. The Court stressed that Colorado’s had given county officials front-line responsibility for dealing with voting emergencies, and that these officials “can deal only in general responses that affect the majority of voters in their districts.” McCain at 6. The county-level granularity, as it were, of practicable emergency accommodations under the state’s regulatory scheme made the Court extremely reluctant to find an equal protection violation based on the fact that individual snowstorm-affected voters ended up confronting different polling hours in different counties. Colorado’s scheme was not arbitrary, and any remedy mandating the coordination of polling-hour extensions across counties would have seriously disrupted it—arguably “flout[ing] the Constitution’s express commitment of the task [of election administration] to the States.” Id.at 7 (quoting Crawford, 128 S. Ct. at 1626 (Scalia, J., concurring in the judgment)).

If Colorado’s had provided for individualized accommodations to deal with emergencies (for example, a right to cast one’s ballot on the day after Election Day for voters who were unable to get to the polls on Election Day due to unforeseen circumstances), it might have been possible for a court to remedy cross-county inequalities by tweaking substantive standards or procedures in a manner that leaves the gist of the state’s regulatory scheme intact.

One last feature of the McCain equal protection analysis is worth mentioning: the judges had no use for Nager’s distinction between polling-place logjams caused by natural disasters, and problems caused more transparently by “state actors”—for example, pollworkers not showing up to work. See id. at 6. (At oral argument, Nager maintained that long lines and other inequalities caused by natural disasters could never give rise to an equal protection violation.) Nager’s approach represents a superficially plausible way of limiting the extent of federal court involvement in election administration, but as the McCain judges seem to have recognized, it is neither conceptually nor pragmatically sound. Conceptually, burdens on the exercise of fundamental rights in highly regulated environments are always the joint, interactive product of state law and natural and social conditions. The extent to which “acts of God” hinder the exercise of a right depends on the legal framework that the state has established (e.g., polling hours). Pragmatically, it would be odd to distinguish between, for example, voters who face long lines due to weather conditions and voters who face long lines because a poll worker overslept. To be sure, it is worth distinguishing between evanescent (one election only) and systemic, persistent problems, but as the oversleeping poll worker example suggests, this distinction is not well captured by Nager’s state-actor/act-of-God line.

What Does McCain v. Obama Teach Us About Election Courts?

Foley considers the trial run of his model for bipartisan election courts a clear success. He writes, “The unanimous decision in McCain v. Obama indicates that any resolution of a debatable Equal Protection (or Article II) issue in a contentious election case will seem more acceptable, and indeed worthy of respect, if it is demonstrably non-partisan[, i.e., decided unanimously by a court composed as this one was].” (Foley does allow, parenthetically, that this conclusion “must remain tentative until it is subjected to further evaluation.”)

My reaction is perhaps a bit more circumspect, for two reasons. First, I wonder whether the apparent ease with which these three judges arrived at their unanimous opinion was related to the fact that the 2008 presidential election did not, in fact, hang in the balance. The hypothetical nature of this exercise may have made it much easier for the judges simply to follow the law as they saw it, and to reach compromises where they saw things differently. Post-election litigation may be one setting where the mythically salutary properties of an “actual case or controversy” prove quite unhealthy. One may hope that real judges (whether on ordinary Article III courts or specialized, bipartisan election courts) who appreciate the risk of subconscious partisan bias when an election does hang in the balance will treat it as a good reason to follow the reasoning of bipartisan bodies that have issued opinions concerning analogous but hypothetical disputes.

Second, the McCain v. Obama simulation has not, as yet, provided us with evidence corroborating Foley’s hypothesis that “any resolution of a debatable . . . issue in a contentious election case will seem more acceptable, and indeed worthy of respect” to the extent that it comes as the unanimous decision of a bipartisan court composed as this one was. We may learn something about this by watching the reaction of real courts to McCain v. Obama. One measure of whether the opinion “commands respect” is whether real courts treat it as an important piece of persuasive authority.

Foley’s larger thesis—that the general public would find post-election dispute resolution by bipartisan election courts more acceptable than dispute resolution by the Supreme Court—might be put to the test using the “experimental vignette” research technique developed by political psychologists. Between McCain v. Obama, Brunner v. Ohio Republican Party,and Bush v. Gore, we now have ample raw material for the construction of vignettes that would test variation along three dimensions: (1) the consensual character of the decision (fractured or unanimous); (2) the partisan balance of the decisionmaking authority (even or tilted); and (3) the “name brand” of the decisionmakers (Supreme Court Justices, versus unfamiliar election court judges). The outpouring of interest in the McCain v. Obama simulation suggests that the results of the proposed study could substantially influence the development of institutions for post-election dispute resolution.

Waiting for a Presidential Winner: The Lessons of 1884 and 1916

by

This year is the second presidential election since 2000. 1884 was the second presidential election since 1876. Can we learn anything useful from this historical parallel, as we prepare for Tuesday Night—and potentially beyond—wanting to know who won the presidential election?

Yes, and from 1916 as well: both involved a delay in determining the outcome of the presidential election, but neither involved kind of calamity that occurred in 2000 (or 1876).

This year is the second presidential election since 2000. 1884 was the second presidential election since 1876. Can we learn anything useful from this historical parallel, as we prepare for Tuesday Night—and potentially beyond—wanting to know who won the presidential election?

As all election lawyers know, and others may recall from their study of history, 1876 was our nation’s first and only predecessor to the debacle of 2000: an all-out brawl over the counting of ballots cast for presidential electors, with the fight not subsiding after the certification of official results by state authorities. Instead, the fight went to Washington, D.C., needing a national institution of government to resolve the bitter dispute, and even the resolution—while decisive in declaring which candidate was entitled to take office—never achieved the key sense of closure that comes from the other side accepting result as a fair resolution.

In the immediate aftermath of 1876, no one wanted a repetition of that situation—in the same way that no one today wants a repetition of what we went through in 2000. But the nation nearly had a repeat situation, at least more nearly than was entirely comfortable at the time. In some ways, the election of 1880 was comparable to the election of 2004. New York ended up being the one swing state in 1880, just as Ohio ended up the one swing state in 2004. And the 1880 presidential election in New York was close: only about 20,000 votes separated the winner, James Garfield, from the runner-up, Winfield Hancock. But it wasn’t close enough to repeat a fight like the one that had occurred four years earlier. It took a few days for Hancock to concede, even longer than it took Kerry to concede in 2004 (because the margin of victory in 1880 was considerably narrower than the roughly 120,000-vote margin in 2000). And even though there were allegations of problems in New York’s voting process that Hancock could have complained about—reports of residents from neighboring states (and even Canadians) coming to New York to cast unlawful ballots there, as well as the claim that 5,000 Hancock ballots had been dumped into the Hudson River—Hancock did not want to put the country through the kind of ordeal that it had suffered in 1876. He and his leading political advisers seriously entertained the idea of contesting the election result in the immediate aftermath of Election Night, but they quickly abandoned that idea.

Four years later, in 1884, New York again was the key swing state, and this time the result was even closer: a roughly 1000-vote margin between Grover Cleveland and James Blaine. Because this outcome was an order-of-magnitude narrower than the immediately previous election, the chance of a contest like the one that had occurred eight years earlier was much greater. Both sides “lawyered up” in 1884, as they had in 1876, as they would do over a century later in 2000, and as they are once again preparing to do this year. Blaine and his lawyers looked for reasons to challenge the result that so slimly favored Cleveland. Bipartisan teams throughout New York reviewed the tallies from each precinct. There was evidence of some problems here and there, including more stories about citizens of other states coming to New York to cast ballots. Ultimately, however, there was not enough evidence of wrongdoing to undermine even the tight 1000-vote margin.

It took two weeks to settle the matter in 1884, an amount of time that might seem like an eternity to us in our era of instantaneous internet and cable news. But it is important to observe that the way in which the election was resolved after two weeks was entirely different from what had happened eight years earlier, in 1876. After the completion of the two-week process of inspecting the vote tallies, Blaine and his supporters accepted the legitimacy of the count. They did not challenge it, claiming that they had been denied their rightful victory, as had occurred in 1876. The dispute, in other words, was not taken to that next, and qualitatively different, polity-wrenching level. That is why, even though the nation retains some historical memory of 1876 (especially after 2000), virtually no one is familiar with the circumstances of the two-week delay in 1884.

Ultimately, it was no big deal: the nation was able to achieve closure successfully, even if took a while longer than would be desirable. This point is significant. It is not merely that a close election of the kind that occurred in 1880 or 2004 is able to be resolved successfully, even when there is no apparent winner on Election Night itself, and it takes a day or so for the concession to occur. The point is rather that even an election as exceptionally close as 1884—where there is but a 1000-vote difference in a single state that will be decisive in determining the Electoral College winner—does not necessarily degenerate into a debacle like the ones that occurred in 1876 and 2000. The two-week delay, while inevitably unsettling at the time, quickly faded into historical obscurity once the concession was made that the count of the ballots was demonstrated to be valid (and thus not susceptible to protracted litigation).

To be sure, there were some Blaine supporters who never accepted the outcome in 1884, just as there may be some partisans who can never reconcile themselves the result this year, whichever way it turns out. But, again, it is an order-of-magnitude point. In 1884, the vast majority accepted the outcome as accurate after the careful two-week scrutiny of the ballots by teams of lawyers on both sides, whereas after 1876 many citizens continued to feel wronged by the eventual outcome, just as would occur again after 2000. Consequently, even if we were to suffer another two-week delay this year (and that scenario does not appear likely as of this writing), it does not need to degenerate inevitably into a repeat of the divisiveness that occurred in 2000. Eight years after 2000, this year could instead be another 1884.

Moreover, it may be of added comfort that 1884 is not the only year in which our nation has avoided the kind of debacle that occurred in 1876 and 2000, despite the exceptional closeness of the vote in a critical swing state. The same sort of situation occurred in 1916. This time the decisive swing state was California. Woodrow Wilson, in his bid for reelection, was ahead of Charles Evans Hughes by about 3000 votes there. Again, it took a couple of weeks to review the counting of the ballots before Hughes issued a concession (on November 22).

If the lawyers had found something to fight over in California, they likely would have—turning that year into another 1876. A recount in New Hampshire had already flipped the state from Hughes’s column to Wilson’s (with him winning it by only 56 votes). And if Hughes had fought over California, Wilson probably would have tried to put Minnesota back in play, where Hughes had won by only 389 votes. But in the end the lawyers couldn’t find enough to fight about. Isolated incidences of counting errors in California were found, but appeared to be random and cancel each other out. Ultimately, California’s electoral process that year was clean and accurate enough to withstand the intense scrutiny of even a 3000-vote margin.

Thus, 1916 was another year (like 1884) in which, despite the considerable delay, the result was an election whose legitimacy and fairness was accepted by both sides. Accordingly, it was another successful closure to the counting process. Hughes’s supporters did not end up feeling like Gore’s supporters, thinking that the wrong candidate was permitted to take office by a system that had failed. As a result, the closeness of 1916 (like that of 1884) is largely lost to history. Given the successful closure, the delay ultimately didn’t matter.

Given this history, two concluding observations. First, it is unlikely that any given presidential election is close enough even to fall into the category of a delayed outcome like the ones of 1884 and 1916. Most close elections are more like 1880 or 2004, if they are even that close. 1948, 1960, and 1976 come to mind as close presidential elections that did not occasion two-week delays before a concession from the losing candidate. Thus, odds are that even if this year ends up being close, there will be a recognized winner within a day or so, with the losing candidate conceding defeat in that quick timeframe.

Second, however, in the unlikely event that we must endure a protracted wait of two or more weeks before knowing who won the presidency this year, as lawyers from both sides review the count of the ballots, it still does not mean that we must suffer a repeat of 2000. Instead, as undesirable as the delay would be, we could hope for a successful resolution along the lines of what occurred in 1884 and 1916. As long as the defeated candidate and the bulk of his supporters accept the review of the count as showing it to be fair and accurate—so that the eventual concession is recognition of the outcome’s democratic validity—then the counting process of 2008 could happily become as historically obscure as these two previous instances of a delayed result.

This piece is drawn from research for a book on the history of disputed elections in the United States that I am writing with my Moritz co-author Steve Huefner.

The Ohio Ballot Experiment of 2008

by

By David Kimball

Associate Professor, Department of Political Science, University of Missouri-St. Louis

In preparation for this year’s election, I have been working with Larry Norden and Margaret Chen of the Brennan Center, and Whitney Quesenbery of the Usability Professional Association to focus more attention on improving ballot design, to reduce voting errors and minimize the number of disputed ballots in post-election recounts. Our report, Better Ballots, summarizes research on ballot design and identifies several confusing ballot features that should be avoided. One of the main recommendations in the report is to list candidates for the same office in a single column on a single ballot page or screen. Some voters see the second column as another contest and mistakenly make a selection from both columns, thus overvoting the ballot.

Ohio has eight candidates on the ballot running for president this year. With this many candidates, some Ohio counties using optically scanned ballots have listed the candidates in two columns on the ballot. By the most recent count, twelve counties in Ohio using optical scan ballots are listing the candidates for president in two columns. The counties are Ashtabula, Athens, Auglaize, Champaign, Delaware, Lawrence, Logan, Madison, Ottawa, Seneca, Shelby, and Wyandot, and they include more than half a million registered Ohio voters. Larry Norden has already blogged about this issue, with links to images of many of the county ballots. A directive from the Secretary of State’s office to county election officials included a ballot template that also listed the presidential candidates in two columns (see page 12 of the directive). We communicated with the Secretary of State’s office and with county election officials, urging them to avoid the two-column layout for the presidential contest, but we were not persuasive enough. In addition, at least three Ohio counties using electronic voting machines have placed the candidates for president on two separate pages. These are Hancock, Portage, and Wayne counties, including more than 230,000 registered voters. Larry Norden has also blogged with more details about these ballots.

We have seen ballots like this before, and the results are not pretty. One way to measure the impact of ballot design on voting behavior is the “residual vote rate” – the difference between the number of ballots cast and the number of valid votes cast in a particular contest. In 2002, ballots in Kewaunee County, Wisconsin listed gubernatorial candidates in two columns (see an image of the ballot here). The residual vote rate for the gubernatorial contest in Kewaunee County was almost 12%, substantially higher than the residual vote rate of 1% in the rest of the state where the contest was listed in one column. A similar problem contributed to the presidential election controversy in Florida in 2000. While a lot of attention focused on the “butterfly ballot” and punch cards in Palm Beach County, there were fourteen other Florida counties with optical scan systems that listed the presidential candidates in two columns. The residual vote rate for president was much higher in the two-column counties (6.9%) than in other optical scan counties in Florida that listed the candidates in one column (0.9%).

Voting in the 2008 general election has already begun in Ohio, so it appears that it is too late the change these ballot designs. In effect, Ohio is conducting what social scientists call a “natural experiment” with respect to the arrangement of presidential candidates on the ballot. We will have a chance to see the results after Election Day. If there is a recount in Ohio, expect some extra attention for the ballots in the counties listed above.

Unanimous Decision in McCain v. Obama (hypothetical)

by

The specially structured 3-Justice Court has released its unanimous decision in the McCain v. Obama simulated adjudication.  The Court ruled that the political question doctrine was inapplicable in the context of the particular case and thus did not deprive the Court of jurisdiction to consider the Equal Protection and Article II questions presented.  On the merits, the Court rejected both the Equal Protection and Article II challenges to the (hypothetical) Colorado Supreme Court decision.

This unanimous decision stands as a model for how “the rule of law” can operate in highly polarized election cases.

As one involved in developing the McCain v. Obama exercise, writing immediately upon the release of the Court’s decision, I am aware that this initial analysis inevitably lacks the benefit of critical distance and reflection. Nonetheless, the following aspects of the decision are striking.

First, its unanimity and anonymity. It was by no means inevitable that the three-Justice Court would be unanimous. As explained previously, the panel was structured intentionally to have two Justices each of whose background showed an affiliation with one of the two major political parties. (These two then mutually selected the third, neutral Justice.) These two Justices were chosen because of their reputation for judicial integrity and intellect. Consequently, there was the hope that, in their effort to decide the case according to law and not politics, they might agree. But there was no guarantee. The facts of the hypothetical were constructed in an effort to be balanced between the two sides of the case, giving each side a fighting chance to prevail—and thus the possibility of each side getting one vote in a 2-1 split decision from the Court.

That is not, however, how it turned out. We do not know the internal dynamics among the members of the panel, but there is no dissent, and the opinion does not read as if it was difficult for the three Justices to come to an agreement. Consequently, there is no way that the Court’s decision can be considered a political rather than legal one.

Moreover, and perhaps this may be reading too much into the “per curiam” designation, but the fact that no single Justice is identified as the author of the opinion might be an extra step to make the decision impartial. The opinion speaks equally for all the Justices, for no single Justice more than the others. That anonymity increases its neutrality—at least in effect, if not by design. Given the impression that one sometimes has that the most important fact about an election case is the identity of the judge (or judges) to whom it is assigned, it is a welcome sign of this McCain v. Obama decision that it conveys the opposite message: the law is the same regardless of the identity of the individual jurists who happen to sit on the panel that issues the decision.

Second, its judicial craft. The opinion is relatively short: ten single-spaced pages. But it is thoughtful and rigorous. The Justices went beyond the precedents cited to them. They articulate clear and cogent views on the three legal issues they discuss: (1) the political question doctrine; (2) Equal Protection; and (3) Article II. Were this a real case, there is no doubt that this opinion—simply by virtue of the power of its reasoning and exposition—would be cited frequently as a well-respected precedent in subsequent cases. It remains to be seen whether, even as an analysis of a hypothetical set of facts, its reasoning will be cited as worthy of emulation in future actual cases raising similar issues.

Third, its “political question doctrine” holding. Walter Dellinger, on behalf of respondents (including Barack Obama), pressed first and hardest on the argument that the Court should dismiss the case as “improvidently granted” on the ground that, involving a presidential election, the dispute belonged exclusively to Congress and thus was beyond the Court’s power. The Court emphatically rejected this proposition—although it ultimately did no damage to Dellinger’s position, since he won on the merits.

The Court’s analysis of the political question doctrine’s potential applicability to this case is a careful consideration of the relevant precedents, pointing to the precise posture of the legal issues in the particular case. The Court emphasizes that it does not decide “the question of who should win the presidential election or who should be on Colorado’s slate of presidential electors.” Instead, the legal matter before the Court is “whether certain votes cast in Denver pursuant to a localized poll-hour extension can be counted in the Secretary of State’s certification of results under federal law.” Moreover, the Court repeatedly notes that it does not know which presidential candidate will prevail if these provisional ballots are counted: “It bears repeating that we are rendering our opinion without knowledge of whom the provisional ballots ultimately will favor in the presidential election as well as the many other election [races] included on the Denver November ballot.” This language, it seems, is designed to underscore the legal, rather than political, nature of the Court’s consideration of the case.

The Court sees its resolution of the Equal Protection and Article II issues on the merits as a straightforward exercise of its power to interpret the Constitution in the context of a conventional legal dispute under the doctrine of Marbury v. Madison. The Court implicitly acknowledges that there might be some specific circumstances in which it would be inappropriate for there to be judicial interference with congressional proceedings concerning the transmission of Electoral College votes from the states, just as the Court may not interfere with the U.S. Senate’s authority to seat one of its members—and its citation to Roudebush v. Hartke, 405 U.S. 15 (1972), for that analogy is an insightful one. But preliminary matters concerning the counting of ballots cast by citizens are appropriate for judicial resolution, and indeed the Court describes as “astonishing” the contention that it could not consider an Equal Protection claim arising in that context.

Fourth, its Equal Protection holding. Perhaps the most instructive aspect of the Court’s decision will prove to be its Equal Protection analysis. As I have written previously, this is an especially murky area of election law, which could greatly benefit from increased clarity. This decision provides a good measure of that, which if followed in actual cases would be applicable to non-presidential as well as presidential elections.

The Court rejected a formalistic approach to Equal Protection analysis, looking instead at the functional equivalence of voting opportunities among citizens. It gives flexibility to state and local election administrators to implement their applicable election codes in a way that avoids micro-management from the federal judiciary pursuant to the Fourteenth Amendment. The unanimous opinion cited Justice Scalia’s concurrence in Crawford (the Indiana voter identification case), as well as the longstanding notion that there must be “a little play in [the] joints” in the administration of election law.

This is not to say that the Court’s approach would give the states a carte blanche to treat their voting citizens however they wished. Rather, the Court’s approach calls for a careful attention to the particular facts of each case. But when the extension of polling hours in one specific location was “a reasonable response to an unanticipated and location-specific natural phenomenon” that ended up “merely in bringing to par” voters in the affected locality—rather than “conferring a preferential impact or disadvantage”—the Court would not invalidate the administrative conduct in question.

This opinion, of course, does not mean that states cannot do better to provide clear guidance in their election codes on how to handle emergency circumstances. The Court observes that “uniform voting rules within a state are highly desirable.” But, as the Court also notes, emergencies can and do happen. And if state law leaves to localities the responsibility for responding to emergencies, federal constitutional law will not invalidate those responses when they can be seen “to serve the overarching goals of equal access to the ballot box and the facilitation of maximum voter participation.”

Interestingly, the Court’s opinion consigns Bush v. Gore to a single footnote. It writes that Bush v. Gore involved a distinct and different “sub-category of election cases,” where “large numbers of local officials applied an indeterminate standard throughout the state over a period of time.” By contrast, the single decision to keep the polls open in Denver for two extra hours was “one unambiguous rule . . . for one district and no further discretion was permitted.” Thus, according to the Court, even if Bush v. Gore had not been explicitly limited to its own circumstances (as it was), it still “would have little precedential force” in this separate context. This distinction is an altogether reasonable one. Nonetheless, it remains noticeable that, both during the oral argument of this case and now in this written decision, there seems to be a desire to make references to that precedent as inconspicuous as possible. Others have observed that Bush v. Gore sometimes seems the “Lord Voldemort” of Court decisions—one almost “dares not speak its name”—and this McCain v. Obama adjudication tends to support that observation to a considerable degree.

Fifth, its Article II holding. Without explicitly saying so, the Court’s opinion appears to embrace the position of the Bush v. Gore dissenters on this issue. The Court accepts, for sake of argument, that there is some Article II constraint on what a state supreme court may do in relationship to interpreting a state’s election code. But the Court here has little difficulty in finding that the (hypothetical) majority opinion of the Colorado Supreme Court did not exceed that outer limit, whatever it may be.

This result is potentially significant if it were followed in subsequent cases. It seems unlikely that the exact same approach used in Chief Justice Rehnquist’s concurrence in Bush v. Gore would accept the state supreme court’s decision here. The Colorado election code says that polls close at 7 p.m. and explicitly adds: “Any person arriving after 7 p.m. shall not be entitled to vote.” Nonetheless, the three-Justice Court here unanimously concluded that the state supreme court was entitled to view this language as “merely ‘a general rule,’ not a blanket prohibition against local initiative to protect voter access in exigent circumstances.” The Court pointed to the “liberal construction” provision of the Colorado election code as a basis on which the state supreme court was entitled to rely for this flexible interpretation of the superficially rigid poll-closing language. Although the Court could have rejected reliance on this “liberal construction” provision had it wished to do so, it evidently did not want to second-guess state court rulings too closely: “We cannot say that this interpretation was so novel or so strained as to fall short of constituting a ‘fair reading’ of state law.”

In sum, it seems fair to say that this non-partisan unanimous three-Justice ruling reflects “lessons learned” in the aftermath of Bush v. Gore, even though the opinion contains no explicit critique of that decision. For one thing, this McCain v. Obama decision’s rejection of the political question doctrine argument is entirely consistent with Bush v. Gore and, in some sense, is a reflection on the fact that there is no going back to a world (if it ever existed) where the federal judiciary lacked power to consider legal issues decided by state supreme courts in this kind of context. Nonetheless, the McCain v. Obama decision reflects a return to the more generally prevalent relationship of federal judicial involvement in the administration of state election law prior to Bush v. Gore—a relationship that sees federal judicial oversight as a narrowly limited exception in keeping with respect for state autonomy. But perhaps most significantly—and in accordance with the hypothesis that this experimental simulation was designed to test (but a conclusion that must remain tentative until it is subjected to further evaluation)—this unanimous decision in McCain v. Obama indicates that any resolution of a debatable Equal Protection (or Article II) issue in a contentious election case will seem more acceptable, and indeed worthy of respect, if it is demonstrably non-partisan.

This unanimous decision surely meets this criterion of demonstrable non-partisanship. It stands as a model for how “the rule of law” can operate in highly polarized election cases. It remains to be seen how well this model can be replicated in real cases.

The Latest from the Land of [Election] Litigation

by

UPDATE (Oct. 24, 5:30pm): Ohio Secretary of State Brunner issued a new directive to govern the process of verifying provisional ballots in order to resolve this dispute for the 2008 election.

Even as Ohio Attorney General (and former Moritz dean) Nancy Rogers attempts to mediate the dispute between Republicans and Secretary of State Brunner over what role if any database “mismatches” should have in the process for determining the eligibility of voters who have cast absentee ballots, these state officials are confronting the resurrection of a separate lawsuit originally filed in 2006. That lawsuit, entitled Northeast Ohio Coalition for the Homeless (NEOCH) v. Brunner, challenges the state’s voter identification and provisional voting rules. A preliminary injunction hearing in federal district court in Columbus has been scheduled for tomorrow (Thursday).

In 2006, NEOCH got up to the U.S. Court of Appeals for the Sixth Circuit, which vacated a TRO that the district court had entered to block enforcement of the voter ID rules specifically for absentee voting. Relying on the then-“brand new” Purcell decision from the U.S. Supreme Court, the Sixth Circuit majority—it was a split 2-1 decision—said that the TRO was unduly disruptive of the voting administration process coming so close to the election and that the interests of any voters harmed by invalid ID laws could be protected adequately by counting their provisional ballots after Election Day. After the case was remanded to the district court, the parties worked out a Consent Order to govern the review of provisional ballots cast in 2006. That order, however, has expired. When Jennifer Brunner replaced Kenneth Blackwell as Secretary of State in 2007, it appeared as if this lawsuit—like many of the others that had been filed in 2006 or even earlier—would be settled. But those settlement negotiations broke down, and now NEOCH has been reactivated.

The NEOCH plaintiffs are specifically asking for a preliminary injunction that would require the Secretary of State to do more to mandate uniform standards for the evaluation of provisional ballots throughout Ohio. In essence, given the Consent Order applicable in 2006, the plaintiffs are asking for something equivalent, or perhaps even more specific in light of evidence that emerged from 2006 concerning variations among Ohio’s 88 counties in the rates at which they deemed provisional ballots to be eligible for counting. Disclosure: in seeking this preliminary injunction, the plaintiffs cite From Registration to Recounts, a book that my Moritz co-authors (Steve Huefner, Dan Tokaji, and Nate Cemenska) and I wrote. That book discussed some of this variation among counties in the rates at which they disqualified provisional ballots for particular reasons. I am not here expressing any view (one way or the other) on the merits of plaintiffs’ request for a preliminary injunction, or their use of this or related provisional voting data as part of their legal claim. (I am also not involved in this litigation, apart from this citation to Moritz’s scholarly work.)

The Secretary of State and the Attorney General are opposing the request for a preliminary injunction. (For complicated reasons that apparently relate to a dispute between these two offices in 2006, two separate oppositions to the preliminary injunction have been filed.) They say that adequately uniform and specific standards have been put in place for the counting of provisional ballots across Ohio. Therefore, they argue, there is no risk of a Bush v. Gore violation in this counting process.

In a reply, the plaintiffs say that they will show at tomorrow’s hearing evidence of county variation in the procedures used to evaluate provisional ballots (and not just the data of different rates of disqualifying them), claiming that this variation constitutes a Bush v. Gore violation. They claim that Secretary of State Brunner has encouraged this local variation in an internal memorandum that states: “The process that a BOE [Board of Election] uses to verify the voter’s ID and qualifications is up to the Board—and there should be a policy in place regarding this.” It is not clear to me, at least at this point based on a quick initial examination of the preliminary injunction papers, the extent to which the plaintiffs are complaining about variations in substantive standards for reviewing provisional ballots, or the procedures used to apply those substantive standards, or both—and what the plaintiffs would have the State do instead. But perhaps the hearing will clarify these points.

The plaintiffs also renew their complaint about the voter ID rules, at least insofar as they are applied to homeless individuals. They argue that the ID requirement, unlike Indiana’s in Crawford, operates as an unconstitutional poll tax. Finally, plaintiffs say that they are not too late in renewing their request for a preliminary injunction applicable to the 2008 election.

For word on the hearing, and what might happen in its aftermath, stay tuned . . .

Understanding the Supreme Court’s Decision in Brunner v Ohio Republican Party

by

By Dale A. Oesterle

There is a great deal of bad commentary in the media on the United States Supreme Court’s recent decision holding in favor of the Ohio Secretary of State, Jennifer Brunner, and against the Ohio Republican Party. The holding did not sanction Brunner’s conduct nor was it a “normal” judicial decision not to intervene in elections at “a late date.” Both claims are false.

The Background of the Case

In 2002 Congress passed a statute, the Help American Vote Act (HAVA), that requires the Secretary of State in all states to compare voter registration lists with other readily available computerized lists, most prominently the state motor vehicle registrations, in order to verify names and addresses.

Ohio’s Secretary of State, running the names of over 660,000 new voter registrations in Ohio, has found close to 200,000 of the registrations to not match the motor vehicle registrations. The 2002 act does not detail what she should do with the list of mismatches.

Brunner’s position, the rationale for which changed during the litigation, was to keep the information on mismatches in her office until after at least after the election and not to make it available to the 88 county election boards that are primarily responsible for gathering and counting votes. Under her system, if poll workers decided to challenge a voter’s credentials, they could call her office and get information on specific names on the list. But the mismatch itself would not be the basis of the challenge.

A challenge would result in the voter cast a provisional ballot that would give the board time to check out the credential issues. It would not “purge” voters or block voters from casting ballots.

In the court papers, her justification was twofold: First, the computer technology was of the state, and the counties did not match, and the state’s list would be hard be subdivide. And second, that most of the errors would prove to be technical and not disqualifying.

Republicans wanted county elections boards to have the mismatch information before the election so challenges could be made at the time people on the list voted.

What’s At Stake

President Bush carried Ohio by a scant 119,000 votes in 2004. In other words, if 60,000 voters had voted for Senator Kerry instead of President Bush, Senator Kerry would have won the election.

Democrats, and organizations sympathetic to the Democratic Party, such as the very controversial not-for-profit group ACORN, have been very aggressive in registering new voters that will vote for Senator Obama. The super majority of the 660,000 new registrations and, therefore, the 200,000 mismatches in issue are likely Democratic voters.

Brunner ran successfully on the Ohio Democratic ticket, which swept all state wide positions except one, and is a strong and loyal Democrat party member. Republicans suspect she is protecting a significant number of problematic new voter registrations from challenges.

The Judicial Opinions

A federal district court judge ruled that Brunner had to make the list of mismatched information to the country election boards. A three judge panel of the Sixth Circuit Court of Appeals overruled the district court judge. The Sixth Circuit then took the case en banc (before all the judges of the court) and ruled 10-6 in favor of the district court judge and reversed its own panel.

Of the judges on the Sixth Circuit, a Republican President had appointed nine of them and a Democratic President had appointed seven. All nine Republican appointees and one of the Democratic appointees were in the majority.

The United States Supreme Court, in a one and one-half page unsigned, unanimous (per curiam) opinion, reversed the Sixth Circuit. The Supreme Court held, very simply, that the Republican Party, a private party, would, if the Court proceeded to a full hearing, likely be held to be unable to sue (would not have “standing” to sue) for violations of HAVA.

The Supreme Court was careful to note that it was not reaching the merits of whether Brunner was complying with the dictates of the statute.

Two errors need to be cleared up. First, of the total number of federal judges on all levels that did reach the issue on the merits, eleven of seventeen federal judges did decide that Jennifer Brunner’s policies were likely to be in violation of HAVA. This alone should be sobering to Ms. Brunner, who has a clear official duty to comply with the statute.

Second, the Supreme Court was not electing to decline to get involved in an election “at a late date” or at any date. The Court strongly implied, in essence, that the plaintiffs did not have standing and would not have had standing had they sued two years earlier.

The eleven federal judges who do not like the Brunner policy note that it is unreasonable to expect local election officials to challenge voters on random grounds in hopes of accessing the mismatch information one voter at a time. Her policy of withholding the mismatch information until after the election will likely result in very few polling place challenges that result in checks of the list. The bottom line: There will be fewer provisional ballots being cast and checked later.

So Where Are We?

Eleven federal judges think Brunner is in violation of the statute. So who can sue Brunner to force compliance? Selected government legal officers may sue. The State Attorney General, another Democrat, and the United States Attorney General come to mind. Neither look to have the stomach for it.

The Ohio Republicans have tried another gambit, suing in state court, the Ohio Supreme Court. If the state court takes the case and decides on the merits it could be appealed to the United States Supreme Court, on the merits, insofar as the state court relies on the violation of federal law found by the eleven federal judges. (If, as the new state suit also alleges, Brunner is also violating an independent state-law obligation, and if the state court chooses to rely solely on that ground for its decision, then there would be no further U.S. Supreme Court review, unless somehow the state court decision itself contradicted federal law.)

If Brunner’s decision stands there will be, in the end, more unchallenged votes cast by new registrants. In other words, new registrants will be asked to cast far fewer provisional ballots cast that are subject to subsequent verification.

ORP v. Brunner: What is at stake?

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

HAVA arguably requires states to compare information contained on incoming voter registration applications against information in state motor vehicles databases and the federal Social Security Administration database to confirm the accuracy of the personal information on the applications. The Ohio Secretary of State has been performing a confirmation process of this kind, but has not been sharing the information with local boards of election (or with the public) in the most accessible way. However, a recent TRO issued by an Ohio district court and upheld by the 6th Circuit would compel the Secretary to supply this information in an accessible form. The Secretary appealed and the case is now in front of Justice Stevens of the US Supreme Court.

What is really at stake? On a superficial level, not much. The case does not directly affect any substantive voting rights, only how information is shared. However, the case could indirectly affect substantive voting rights depending on how people use this information.

Specifically, people can use a failed confirmation as some evidence to suggest that the registrant is not a qualified voter and is not entitled to vote. There is no black-letter law that would be binding in Ohio that supports the proposition that a failed confirmation is in itself sufficient to disqualify a voter (see here) but, perhaps together with other pieces of information, a failed confirmation might be enough to persuade election officials or a judge that a person is not qualified to vote. The law does not go into the details of how much evidence would be necessary to permit election officials or a judge to conclude that a voter is not qualified, but by omission seems to imply that these actors should rely upon common sense and their own best judgment. See, e.g., R.C. 3505.20, 3509.06, 3509.07.

The mismatch list could be used to attempt to disqualify votes in at least two ways. First, people could use the information to attempt to disqualify absentee ballots that have already been cast. At the discretion of the county Board of Elections, these ballots may be counted centrally or distributed to the precincts and counted there. R.C. 3509.06. If they are counted in the precincts, the poll workers who count the ballots will determine whether to challenge each ballot. Id. If the ballots are counted centrally, “special election judges” appointed in the same manner as poll workers will decide. Challenged absentee ballots should be rejected, among other reasons, when election officials find the voter’s statement accompanying the ballot is “insufficient” or “the applicant is not a qualified elector in the precinct.” R.C. 3509.07; 3509.05. If the poll workers decide to reject the ballot, they endorse it as “not counted” and put it off to the side with its accompanying paperwork, making it possible to go back and reconsider whether to count the ballot later. R.C. 3509.07. Although current black-letter law does not afford voters any notice or opportunity to be heard regarding the validity of their absentee ballot, Secretary Brunner has ordered these kinds of due process protections in other contexts and may order them to occur in this one.

The other thing that could happen is that the information could be used to challenge voters at the polls. Ohio law used to permit appointed challengers to make these kinds of challenges, but was recently amended to prohibit challengers. Nevertheless, regular poll workers are still able to challenge voters at the polls, and the Ohio Supreme Court has even hinted that it is improper for them to refrain from doing so when they subjectively believe something is amiss. Some of the poll workers in each Ohio precinct are required to be Republican (see here) and, although the Ohio Republican Party has stated that it will not directly challenge voters at the polls, that statement is not necessarily inconsistent with informally encouraging Republican poll workers to make such challenges. The upshot is that, if the mismatches cause poll workers to subjectively question the eligibility of a voter– regardless of whether that questioning is reasonable– we can expect at least a good handful of challenges on election day. Furthermore, if a majority of poll workers are not satisfied with the voter’s response to these challenges, the voter may cast only a provisional ballot and attempt to cure any perceived deficiencies by providing information to the Board of Elections under R.C. 3505.181(B)(8). R.C. 3505.20. It is also possible, theoretically, that county Boards, after investigation, could identify certain voters in the poll book as having problematic registrations and order poll workers to require those voters to cast a provisional ballot. However, as stated earlier, a mismatch should not be enough to authorize this unless it causes the Board to conclude that the registration suffers from one of the underlying problems enumerated in Ohio law that require the casting of a provisional ballot. See R.C. 3505.181. A mere mismatch is not one of those enumerated grounds.

One positive note about all this is that, if the letter and spirit of the law are followed, no legitimate votes will be discarded. Rather, the ultimate consequence of all this litigation should just be to slow the process down so that society can take a closer look at these issues and deal with them in a way that has a greater likelihood of being accepted by the public. Officials will carefully sort through the provisional ballots cast by voters challenged at the polls, as well as the absentee ballots that remain in their identification envelopes, and make a fair and lawful determination of which ones to discard, if any. Then they will declare a winner and everyone will accept it. This is what should happen.

What will happen? Well, there is risk of strategic behavior on the part of both political parties that would probably reduce the likelihood that the ultimate result will be accepted. If Ohio is close for any important race, both parties can be expected to attempt to influence the disposition of the provisional and absentee ballots in accordance with narrow conceptions of self-interest that will do damage to the already low level of faith that some Ohio voters currently have in the elections system. Furthermore, because Ohio will probably be close, both parties might act in this selfish manner even before the election. For instance, Republicans might encourage poll workers to be overzealous in their challenges to voters at the polls. While this theoretically should not lead to the rejection of anyone’s ballot, it could certainly lead to long waiting times that could discourage people from casting ballots in the first place. For their part, Democrats might, among other things, pressure election officials to begin removing absentee ballots from their identification envelopes at the earliest possible time (October 25) and mix them together so that it becomes a practical impossibility to identify problematic ballots and investigate them. In 2005, the Ohio Supreme Court suggested that premature opening of absentee ballot envelopes is harmless error unless there is some further suggestion of fraud or other impropriety. Harmon v. Baldwin, 107 Ohio St.3d 232, ¶ 32 (Ohio, 2005). While this tactic might increase the number of votes counted for Democrats, it might also cause Republicans to question the accuracy of the vote in the same way that some Ohio Democrats continue to question the result of 2004.