Please note: The next Weekly Comment will appear Tuesday, September 5.
As one watches from a distance, the main problem with Mexico's electoral system seems to be a lack of sufficient trust in its integrity. Felipe Calderón, the conservative candidate, appears to have won a come-from-behind victory by about one-half of one percent (243,000 out of 41 million ballots). But Andrés Manuel Lopez Obrador, his leftist opponent, has suggested defiance of the outcome unless a complete recount of all ballots confirms the result. This unwillingness to accept the result, supported by mass demonstrations, stems in part from a history of election fraud in Mexico and the persistent distrust that past corruption has bred despite recent efforts at reform.
A danger associated with this distrust is that the country cannot peaceably settle its dispute about this election. Suppose a recount cut Calderon's lead in half, to roughly 120,000 votes (the same amount by which Bush beat Kerry in Ohio), but Lopez Obrador still refuse to accept defeat, claiming instead that the recount was rigged. His supporters take to the streets, chanting that the will of the people shall prevail. Violence erupts. The military squashes the protest.
If this occurs, democracy itself will have been defeated. The whole point of democracy is to establish an election process that competing political forces, despite the strength of their passionate ideological disagreements, can accept as a fair mechanism of choosing which side shall prevail for a period (until the time for the next election). The losing side may not like the result, but they can live with it, if they share with their ideological opponents a mutual commitment to abide by the verdict of the citizenry as expressed by the election's outcome. But if the losing side rejects the election's result as irredeemably corrupt — and its resistance to this result can be quelled only by military force — then democracy has failed. In such a circumstance, the society has shown itself to be incapable of functioning democratically.
Whatever danger Mexico currently faces in this regard, it would seem much greater than risk that democracy will fail in Ohio this year. Although Ohio's electoral system had significant problems in 2004, there was no evidence that President Bush's 120,000 victory was tainted by fraud. Nor does Ohio have a history of electoral corruption that would breed the kind of distrust that plagues Mexico's electoral system.
Yet the level of distrust in Ohio is high — and disconcerting. It is, unfortunately, not just fringe conspiracy theorists who continue to believe (incorrectly in my judgment) that Kerry was robbed in Ohio, unless one wishes to classify Robert F. Kennedy, Jr. as part of the fringe. His Rolling Stone article "Was 2004 election stolen?", which asserts that over 350,000 mostly Democratic voters were denied the right to cast a countable ballot, bears little resemblance to reality (see my colleague Dan Tokaji's a point-by-point analysis of Kennedy's claims). Nonetheless, the lesson of this year's election in Mexico, as well as every other disputed vote count (including Florida in 2000), is that public perception is at least as important as reality. Regrettably, the lingering perception of a significant portion of the public is that Bush's 2004 victory in Ohio was questionable because of intentional efforts by Republicans to suppress legitimate Democratic votes.
Moreover, preliminary skirmishes in Ohio this year appear intended to increase public distrust of the electoral process. Democrats have claimed that Secretary of State Kenneth Blackwell, who is running for Governor, has adopted rules that will impede voter registration drives. Led by Senator Chuck Schumer and Representative Rahm Emanuel, both national and local Democrats have urged Blackwell to step aside as Secretary of State, arguing that it is impossible for him to administer the state's election laws impartially at the same time as he is a candidate.
This particular attack on Blackwell seems overwrought, as the Columbus Dispatch observed in a recent editorial. Ohio's legislature, not Blackwell, bears primary responsibility for the new rules on voter registration drives. Likewise, however misguided it may be, Ohio law contemplates that a Secretary of State will run for office while serving in that position: sometimes the elected Secretary of State will be seeking reelection, and sometimes this incumbent may be seeking higher office, as in Blackwell's case. Democrats have been in that position, and indeed a Democratic Secretary of State in Iowa is currently running for Governor. I have argued previously that this situation presents an inherent conflict of interest, but it is one hardly unique to Blackwell, and unless and until the situation is remedied with a structural reform, it is difficult to see that Blackwell uniquely should abandon his current position.
Yet questioning Blackwell's ability to fairly administer the electoral process may be effective in increasing public distrust of Republican victories in November, should they occur by narrow margins. Suppose Blackwell himself wins a come-from-behind victory by less than one-half of one-percent of the vote — say, a mere 15,000 out of a projected 3.5 million cast? How vigorously would Democrats attempt to challenge the result, through litigation and through public demonstrations? To what extent would the claim of illegitimacy parallel the situation unfolding in Mexico?
Another way to consider this issue is to ask how wide would Blackwell's margin of victory need to be for Democrats to accept it as legitimate, without litigation or other forms of public protest? If the margin is under 1,000 — or even under 10,000 — one can be sure that the Democrats will be looking for ways to undermine the result. But what if the margin is somewhere between 10,000 and 100,000?
The answer to that question may depend on how the election process actually unfolds. The more problems that surface that arguably affect the validity of the victory, the wider the margin needs to be. And the enumeration of problems has already begun: while a subsequent claim in November that Blackwell won because he wrongfully impeded voter registration efforts would not stand on its own as a credible attack on the integrity of the election's outcome, that grievance added to a list of others might mean that Blackwell's margin needs to be a little bit larger for it to be publicly acceptable as legitimate (perhaps an extra 1,000 votes or so more than what otherwise would be required for acceptability).
What additional problems might emerge that could cloud a Blackwell victory? There are many possibilities, but perhaps the most likely is difficulties in the administration of the new voter identification rules at polling places on election day. These rules, together with complexities added to the provisional voting process, may cause confusion among poll workers, leading to variation among precincts in the administration of these rules. Some precincts may adhere to strict enforcement of the new ID rules, or perhaps erroneously adopt an ID requirement more exacting that the law itself demands, while other precincts may opt for a more lenient approach, perhaps even abandoning any ID requirement as long as the would-be voter's signature matches the signature in the poll book. Any such discrepancy could give rise to an Equal Protection lawsuit modeled after Bush v. Gore, with the requested remedy being an obligation to count all provisional ballots that were rejected for lack of ID. If it turns out that 30,000 provisional ballots are rejected for this reason, it would seem that Blackwell would need to win by at least 20,000 votes for Democrats to accept his victory without going to the courts and perhaps to the streets as well.
Moreover, it might happen that confusion over the new voter identification rules causes delays at some polling places, resulting in long lines. If there is any evidence that voters abandoned the effort to cast a ballot because of these delays, allegations about this disenfranchisement will be added to the list of problems that supposedly prevented the legitimate candidate from winning. Depending on the strength of the evidence, such allegations could require Blackwell to win by several more thousand votes for Democrats to accept the result.
Problems might also arise in the context of absentee ballots. In past elections in Ohio, as in other states, there have been allegations of improper partisan tampering of absentee ballots. If any evidence of such wrongdoing turns up this year, it will fuel the claim of fraud.
There could also be a repeat of the disasters that occurred in Cuyahoga County during this year's primary election in May: polling places hours late in opening, requiring court order to extend their operations; equipment failures caused by improper procurement and inadequate training; and the loss of memory cartridges containing vote totals from a large number of precincts. Any missing votes attributable to similar mismanagement in the November election would further extend the margin necessary for an unimpeachable victory.
One could go on, but the point is clear: an accumulation of various problems that prevent ballots from being cast or counted, or that cause improper ballots to be included, would serve as the basis for contesting the election's result. If there are enough of these different problems, and if the total number of erroneously excluded or included votes becomes large enough, it is conceivable that there could be a cloud over a Blackwell victory even if the margin were 50,000 votes or more.
To be sure, any of these problems could cloud the result if the winner of Ohio's gubernatorial election is Blackwell's opponent, Democratic Congressman Ted Strickland. Similarly, even if the Blackwell-Strickland race ends up not that close, it is conceivable that the statewide U.S. Senate race between Republican incumbent Mike DeWine and Democratic challenger Representative Sherrod Brown could be tight. Then, these kinds of problems could taint a DeWine or Brown victory. Even so, Blackwell needs to win by a wider margin than any other of these Ohio candidates in order to avoid allegations of an inaccurate result, simply because of the significant distrust (whether deserved or not) among the public concerning the fairness of the election as long as Blackwell serves as Secretary of State.
Likewise, there could be close races in other states that lead to charges of erroneous or fraudulent races. Several other midwestern states — Michigan, Minnesota, Wisconsin, and Iowa — have gubernatorial elections that are projected to be tight. Each of those states has had some questions raised previously about the integrity of its electoral process, and again Iowa has a "Blackwell-type" situation of the Democratic candidate being the current Secretary of State. (In Michigan and Minnesota, Republican Secretaries of State are running for reelection.)
Elsewhere around the country, Florida, Pennsylvania, Maryland, Tennessee, Missouri, and Colorado, all have gubernatorial or senatorial elections that currently are projected to be close. Obviously, Florida is well known for its electoral difficulties, but these other sates have suffered problems in the past as well. (Colorado's Republican Secretary State is also running for reelection.) Therefore, they could become the scene of an electoral meltdown, like the one that befell the State of Washington in connection with its governor's race in 2004.
But all eyes are — and should be — on Ohio. That is where the combination of public distrust, risk of problems, and likelihood of a photo finish is the highest. It is this combination that presents the gravest danger that the public will not accept the validity of the result. Even if a recount occurs, a significant portion of the public may refuse to acknowledge the legitimacy of the recount. While one hopes that this portion grows no larger than it was in the aftermath of the 2004 election, one must be ready for the possibility that it will increase substantially and also become more virulent.
Is there anything that can be done now to reduce this risk? Improved poll worker recruitment and training, as well as public awareness campaigns about the new ID rules, are obvious, albeit expensive, measures. Replacement of Cuyahoga County's election officials, as urged by the Cleveland Plain Dealer, would be another prudent step. But much of the system for administering this November's election is already in place: the new ID and provisional ballot rules have already been adopted by the legislature, as has the move to "no excuse" absentee voting. And even with new management, one cannot expect Cuyahoga County to have turned things around entirely by election day.
Thus, if the election is close and problems occur, one can only hope for some measure of forbearance on the part of the losing candidate and his supporters. In other words, rather than rushing to condemn the result as illegitimate even when the evidence of outcome-determinative errors is thin, the losing candidate instead might refrain from contesting the result unless the evidence of an inaccurate outcome is overwhelming. But it is probably too much to expect such magnanimity, especially in this climate of distrust.
Indeed, as the situation in Mexico illustrates, the worst attribute of this distrust is that, like a cancer, it can grow according to its own destructive dynamic, with little ability of intervention to stop it.
Thus, we must wait and watch as the 2006 election in Ohio unfolds. Let us hope it is entirely uneventful and that the races for Governor and Senate are won by indisputable margins (whether they be Democratic or Republican victories). Otherwise, if 2006 compounds the distrust generated by 2004, one can only dread what the public perception of the electoral process might be like in 2008.