Where There's Smoke, There's Fire

Electronic Voting and the Legitimacy of the 2004 Presidential election

November 30, 2004
By Bruce W. Fraser

I'm not one for fantastical explanations or conspiracy theories, so when murmurings about election fraud in the recent presidential election began to make their way into the public dialogue, I didn't think much of it. However, when a research team at UC Berkeley released the results of its statistical analysis of voting patterns in Florida recently, [1] my skepticism gave way to a healthy curiosity about the factual evidence behind the study.

As it turns out, there are a number of reasons to be skeptical about the official results of the recent presidential election, reasons that transcend party politics and raise questions about the integrity of the democratic process in this country. My purpose here is to provide some general observations that link the Berkeley study with other, equally important information about the accuracy of the recent election results.

The press release that introduced the Berkeley study to the public is unequivocal in its implications: "UC Berkeley Research Team Sounds 'Smoke Alarm' for Florida E-Vote Count; Irregularities may have awarded 130,000 - 260,000 or more excess votes to Bush in Florida."

While the projected numbers of votes referenced in the title are below the margin of victory for President Bush in Florida (Bush won the state by 380,000 votes), the study took into account only those counties that used electronic voting machines; it did not analyze the results in the counties that used optical scanners to tabulate election results, scanners that have had a history of technical problems and false results. [2]

Moreover, the results of the study are robust: "No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," says Michael Hout, the lead investigator and member of the National Academy of Sciences. Simply put, something went terribly wrong in Florida on Election Day, and the integrity of the election results in the state now seem to be in question.

Relying on a single study to justify such a bold claim about Florida's election results is problematic, but the Berkeley study does not stand on its own. Exit polls in Florida showed Bush winning about 14 percent of the Democratic vote, a number almost identical to, and statistically indistinguishable from, the number of Democrats Bush won in 2000, but he somehow managed to win the state despite the fact that independents favored Kerry by a margin of 57 to 41 percent. In 2000, by contrast, Gore won the independent vote by a margin of 47 to 46 percent, [3] and would have won the state had it not been for interlopers like Katherine Harris.

Such results are statistically stunning to say the least. When one also takes into account the exit polls showing Kerry in the lead in Florida before the majority of voters in largely Democratic areas had made their way to the polls, [4] suspicions about the accuracy of the final vote tallies in Florida cannot be summarily dismissed.

By now, though, the major news organizations, along with members of the Republican Party, have spun enough skepticism about the reliability of exit polls into the American consciousness that inferences about election results on the basis of exit polls appear uninformed. This is unfortunate.

Historically, exit polls have been so reliable that they are used in emerging democracies to determine whether or not election fraud has taken place. In the former Soviet Republic of Georgia, for example, this year's parliamentary elections were monitored using exit polls sponsored by international foundations. When the official results of that election did not jibe with the exit poll predictions - predictions that indicated the opposition party was victorious - Eduard A. Shevardnadze, the prime minister, was forced to resign under pressure from the US government. [5]

More recently, exit polls of voters in Ukraine provided information that motivated Republican Senator Richard Lugar to offer a scathing review of the Ukrainian presidential election: "A concerted and forceful program of election-day fraud and abuse was enacted with either the leadership or cooperation of governmental authorities." [6]

Dick Morris, a Republican polling consultant and regular on Fox News, put the reliability of exit polls in perspective immediately following our own Presidential contest: "Exit polls are almost never wrong... To screw up one exit poll is unheard of. To miss six of them is incredible. It boggles the imagination how pollsters could be that incompetent and invites speculation that more than honest error was at play here." [7]

Morris's reflections are backed up by Steven F. Freemen of the University of Pennsylvania, who, in a paper released to the public on November 10, [8] notes that the statistical likelihood of exit polls being as far off the mark as they were on election night in three of the battleground states - Ohio, Pennsylvania, and Florida - is 250 million to 1 (Freemen used polling data that were not adjusted by the networks to fit with the official tally [9]).

When one considers that exit polls failed to predict the correct outcome in 9 out of the 11 battleground states, and that in every case the error showed that Kerry would be the victor, the statistical likelihood that the exit polls were off by chance becomes a practical impossibility.

Of course, there is always the possibility that there was some systematic problem with the exit polls, that too many women or not enough Republicans were surveyed, but to attribute this kind of blatant error to polling agencies is itself problematic. Not only are such errors easy to spot and correct, but the corporate bodies that make polling their bread and butter (in this case, Edison Media Research and Mitofsky International) would have filed for Chapter 11 long ago as the result of such blatant incompetence. Such attributions smell more like ex post facto rationalizations than they do genuine explanations.

Even with the polling data cast in the appropriate light, the tendency on the part of most Americans will be to dismiss such conspiratorial reflections as half-baked or poorly informed. After all, how could such a thing happen in the United States? How could something so sinister in its implications get by the media and our well-intentioned election officials?

Such questions presuppose two fundamental beliefs: first, that the press is still capable of challenging the power structure in American in a nontrivial way, and, second, that election officials are knowledgeable enough about the potential pitfalls of electronic voting to identify and correct problems. Both assumptions are, regrettably, false.

The declining quality of press coverage in the United States has been documented and bemoaned by countless authors and institutes. For example, T.E. Patterson of the Joan Shorenstein Center on the Press at Harvard University has tracked the steady decline in the quality of news coverage over the last decade. "Evidence suggests that soft news [is] weakening the foundation of democracy by diminishing the public's information about public affairs and its interest in politics," [10] says Patterson.

Elliot D. Cohen, media ethicist, concurs: "The mainstream corporate media in this country are simply not living up to their responsibility to serve as [the] Fourth Estate. Instead of protecting the public against governmental abuse of power, these corporate giants, driven by an insatiable appetite for profit, are in business with government. For example, it would be na�ve to expect General Electric, parent company of NBC, to jeopardize a military contract by exposing widespread election fraud." [11]

Granting for the moment the inadequacies of the corporate media, wouldn't our well-intentioned election officials have spotted election fraud? Although most of the country's elections officials are competent and well-meaning, it would be premature to conclude that voting irregularities would be identified, let alone corrected, by those officials - at least when it comes to electronic voting.

Electronic voting machines currently used in the US are not auditable: There is no way to determine whether or not a technical error is assigning too many votes to one candidate, no way to tell whether or not every vote is being counted, and no way to tell whether what appears on the voting screen is actually what is being recorded in the black box. To make matters worse, the manufacturers of voting machines require that their clients (e.g. the State of Florida) sign 'proprietary contracts' to protect against possible theft of their trade secrets, thereby eliminating any oversight of the product by governmental officials or voter advocacy groups.

In an environment with no meaningful oversight and no paper trail, vote tampering can occur virtually undetected. As M.K. Dooner, founder of Florida-based Deep Sky Technologies and an expert on the technical aspects of electronic voting puts the point: "There are virtually no limits to the way votes can be distributed... Programmers can write code that is self-erasing and redistributes votes to conform to whatever outside interests are operative, and without any kind of traceability." [12] Dooner's reflections are reinforced by a Johns Hopkins University study in which researchers note: "Software flaws in a leading US electronic voting system could be used to subvert the outcome of an election." [13]

This is why the Berkeley study is so important and why exit polls matter. Without the data provided by these types of empirical studies, there is no way to evaluate the reliability and accuracy of voting machines.

In the weeks following the presidential election, pundits used the official results to challenge the accuracy of the exit poll outcomes, but in light of the historical accuracy of exit polls and the scientific legitimacy of the Berkeley study, one can with equal justification use the information to cast doubt on the accuracy of the machines.

The most reasonable conclusion to draw from the evidence is that the election results are skewed, and that there is a very real possibility that widespread election fraud took place during this year's presidential election. Such reflections bring us to the most nefarious aspect of the election picture, namely, the connection between electronic voting, corporate interest, and the Republican Party.

As it turns out, only a handful of corporations produce the voting machines currently used in US elections, and federal law requires no background checks on company employees or oversight of the production process. [14] The potential for conflicts of interest in this environment is obvious.

For example, Senator Chuck Hagel of Nebraska, who in 1996 pulled out two improbable victories during his meteoric rise to the Senate, had been Chairman of American Information Systems (AIS) right up until two weeks before he announced his candidacy for the Senate. AIS was the company that produced the voting machines that calculated as many as 85% of the votes cast in the senatorial race, [15] a race in which Hagel defeated a better known Republican candidate in the primary on his way to a landslide victory over a popular Democratic governor who had led in the polls the entire race. [16]

Then there is Wally O'Dell, [17] the CEO of Diebold Election Systems, whose now-infamous comment about helping deliver Ohio to the GOP in 2004 raised the ire of those already concerned about corporate administration of the election process. O'Dell is a staunch political supporter of the President and has given generously to Republican candidates for years.

People should not be so naive as to believe that, in the post Enron and WorldCom climate, there is no possibility of vote manipulation by those who both control the voting process and stand to gain. At a time when the United States is marketing its foreign policy by touting the virtues of democracy, the tawdry state of the election process in this country is more than an embarrassment. With corporate influence over press coverage and governmental policy, and with the very real possibility that the voting process is being controlled by the ruling elite, American democracy is in danger of being replaced by a form of government that looks remarkably like tyranny.

What makes this political shift so difficult to detect, and so dangerous, is that the illusion of democracy can still be maintained by exploiting the ignorance of the voting public about the technical problems with electronic voting, and by using the media to incite and reinforce suspicions about exit poll results. [18]

No doubt, conservative think tanks will come out with 'studies' at odds with the UC Berkeley findings, and the political headwaters will then be sufficiently muddied that most Americans will chalk up suspicions about the election to disgruntled Democrats and leftist conspiracy theorists who are part of the 'Liberal Elite'.

When all is said and done, however, such complacency will only allow the collective voice of the American people to be drowned out by the special interest groups and corporate advocates who have hijacked the democratic process. Such remarks are alarmist, to be sure, but they are not divorced from the facts, and if the facts still mean anything, it is time for Americans to reclaim the democratic process. Indeed, if Justice is truly blind, the issue of election fraud and corporate malfeasance must be pursued to its logical conclusion.

Bruce Fraser is a freelance writer who earned his Ph.D. in Philosophy from Boston University in 2001. He resides in Florida.

1. http://ucdata.berkeley.edu
2. Harris, Beverly, and David Allen, Black Box Voting: Ballot Tampering in the 21st Century, Plan Nine Publishing, 2003
3. http://www.msnbc.com/m/d2k/g/polls.asp?office=P&state=FL.
4. Zogby International, www.zogby.com
5. Steven F. Freeman, Ph.D., "The unexplained exit poll discrepancy," available online at http://www.buzzflash.com/alerts/04/11/The_unexplaind_exitpoll_discrepanc y_v00k.pdf
6. New York Times, November 23, 2004, "Republican challenges presidential election based on exit polls."
7. The Hill, November 4 2004. "Those faulty exit polls were sabotaged," by Richard Morris.
8. Steven F. Freeman, Ph.D., "The unexplained exit poll discrepancy," available online at http://www.buzzflash.com/alerts/04/11/The_unexplaind_exitpoll_discrepanc y_v00k.pdf
9. The unadjusted data were posted on CNN's website on election day, but were pulled around 1:30 am and replaced with the data adjusted to fit the vote tallies.
10. T. E. Patterson, "Doing Well and Doing Good: How Soft News and Critical Journalism are Weakening our Democracy," available at www.ksg.harvard.edu/presspol/Research_Publications/Reports/softnews.pdf
11. Personal correspondence. For an in depth look at issues relating to compromised press coverage, see Elliot Cohen's News Incorporated: Corporate Media Ownership and Its Threat to Democracy, forthcoming, Prometheus Books. Noam Chomsky's work also gives detailed analyses of the role of the press in a democratic society (e.g. Manufacturing Consent)
12. Personal correspondence.
13. NewScientist.com, July 25, 2003, "E-voting systems flaws 'risk election fraud'"
14. Harris, Beverly, and David Allen, Black Box Voting: Ballot Tampering in the 21st Century, Plan Nine Publishing, 2003
15. Ibid.
16. Business Week, July 10, 2000, "Chuck Hegal.Landslide Upset."
17. Harris, Beverly, and David Allen, Black Box Voting: Ballot Tampering in the 21st Century, Plan Nine Publishing, 2003
18. The Hill, November 4 2004. "Those faulty exit polls were sabotaged," by Richard Morris.