The Stolen U.S. Presidential Election: A Comparative Analysis

First published as “Election Fraud in America,” Centre for Research on Globalization (30 November 2004), (together with a companion piece, “Evidence of Fraud in the 2004 U.S. Presidential Election: A Reader,” Centre for Research on Globalization [5 December 2004], This essay appeared under the present title at the now defunct website of Autonomy & Solidarity (2 December 2004); under its original title it was also published at six other websites in 2004, and is also available at In the original text, parenthetical references to sources referred to items listed in the “Evidence of Fraud” reader; in the present text, these parenthetical references have been replaced by footnotes.

Two issues raised in this essay call for comment. In referring on pp. 12-13 to a statistical study published on November 18, 2004 by Michael Hout of the University of California at Berkeley that claimed to discover massive irregularities in Bush's Florida vote due to electronic voting machines, I also cited a response to this study published three days later by Alex Strashny of the University of California, Irvine, that criticized Hout's study. Strashny's refutation of Hout's analysis, as Hout himself acknowledged with some embarrassment not long after the appearance of my article, was correct.

On p. 14 I referred to Teed Rockwell's discovery that according to the Cuyahoga County Board of Elections website, 93,000 more votes were cast in 29 communities of that county than there were registered voters. Here again a correction is needed. Rockwell was not mistaken in his reading of the data published by the Cuyahoga County Board of Elections: I read over the same data myself and found, as I noted on p. 14, that he had if anything been conservative in his assessment of the scale of the irregularities that that website revealed.

But as I wrote in “The Strange Death of American Democracy: Endgame in Ohio” (published on January 25, 2005), this was a “very large-scale false alarm.” In that essay, I explained the matter as follows:

The election results published by Cuyahoga County (which includes the city of Cleveland) led a number of commentators in November 2004—myself among them—to believe that there had been massive 'ghost-voting' fraud in the suburbs of Cleveland. But the official lists showing twenty-nine communities with voter turnout figures of more than 100 percent (and hence some 93,000 'ghost votes' in the county) turned out to result from a bizarrely structured software program that grouped communities in the same congressional, house and senate districts, and added the total number of absentee ballots within the combined districts to the voter turnout figures for each community in these districts—though not to the vote totals for candidates or issues. This programming oddity worked, the County's website idiotically declared, in “even-numbered years.”

I have left the text of the present article unchanged: these passages provide an instructive indication of the potential pitfalls involved in work of this kind.


1. ‘Let us compare mythologies’: Presidential Votes in the U.S., Ukraine, Venezuela

Imagine the sensation that would have ensued if a United States Senator had declared, less than three weeks after the 2004 U.S. presidential election, that “It is now apparent that a concerted and forceful program of election-day fraud and abuse was enacted with either the leadership or co-operation of governmental authorities.” The story would have made banner headlines around the world.

As a matter of fact, on November 22, 2004, BBC News attributed these very words to Republican Senator Richard Lugar. However, Lugar was speaking in his capacity as Chairman of the U.S. Senate Foreign Relations Committee—and he was referring, not to the U.S. presidential election of November 2, but to the Ukrainian presidential election of November 21, 2004.1

The primary evidence for Lugar’s charge of electoral fraud is a striking divergence between exit poll data and official vote tallies. As it happens, wide divergences of this kind have been a feature of three important recent elections: the Venezuelan recall referendum over President Chávez’s mandate held on August 15 and the U.S. presidential election of November 2, as well as the Ukrainian presidential election of November 21. In all three cases there is substantial evidence of fraud—though the dishonesty appears to be very differently distributed. In brief: the Venezuelan election was clean and the exit poll flagrantly dishonest; the Ukrainian vote tallies and exit polling seem both to have been in various ways seriously corrupted; the American election, despite the Bush Republicans’ pose as international arbiters of integrity, was manifestly stolen—though the U.S. exit polling was professionally conducted. (The exit polls were subsequently tampered with, but accurate results had in the mean time been made public.)

     a. Venezuela

Hugo Chávez’s landslide victory in August was a surprise only to the hostile U.S. corporate press, which had represented the Venezuelan election campaign as a dead heat: the last opinion poll prior to the referendum in fact showed Chávez leading by a wide margin, with 50 percent of registered voters to the opposition’s 38 percent. In the official tally, Chávez won 58.26 percent of the votes, while 41.74 percent were cast against him. International observers, including the Organization of American States and the Carter Center, declared that the election had been fair: in ex-U.S. President Jimmy Carter’s words, “any allegations of fraud are completely unwarranted.”2

But on election day the leading New York polling firm Penn, Schoen & Berland disgraced itself by releasing (before the polls closed, and hence in violation of Venezuelan law) a purportedly authoritative exit poll, with a claimed margin of error “under +/-1%,” according to which Chávez had been defeated, gaining a mere 41 percent of the vote to the opposition’s 59 percent. The exit polling, it emerged, had been conducted—though not in Chavista neighbourhoods, where the pollsters did not venture3—by an opposition group named Súmate, which had been formed to agitate for a recall referendum, and whose leadership had been implicated in the 2002 anti-Chávez coup. Súmate appears to have been largely funded by the U.S. National Endowment for Democracy (NED), which has been aptly described as “the CIA’s ‘civilian arm’,”4 and by the CIA itself;5 in the period leading up to the election, Venezuelan opposition groups like Súmate received altogether more than $20 million from the U.S., including over $3 million funneled through the NED.6 As had been understood prior to the event,7 fraudulent exit polling was part of a concerted U.S.-backed project of delegitimizing and destabilizing the government of a geopolitically important oil-producing nation. Had the election been less of a landslide, and had it not been conducted with what appears to have been scrupulous correctness, the plan might have succeeded.

     b. Ukraine

Ukraine is likewise recognized as a country of pivotal geopolitical importance;8 it is a key element in the U.S.’s Silk Road Strategy for domination of central Asia.9 Here the election results were much closer, and have been more vigorously contested. Viktor Yanukovych, the candidate favoured by Ukraine’s Russian neighbours, was declared the winner, with 49.4 percent of the vote to the Western-leaning Viktor Yushchenko’s 46.7 percent. But Yushchenko and his party—supported by a growing chorus of Western commentators and governments—have cried foul, and Ukraine’s Supreme Court has rejected the certified vote tally, declaring that the election must be repeated in late December.

While the Ukrainian exit poll figures publicized in the Western media immediately after the November 21 election do support claims of electoral fraud, the exit polls themselves are not wholly above suspicion. The most widely disseminated claim has been that an authoritative exit poll showed Yushchenko to have won the election with a 6 percent lead; Yanukovych’s governing party would thus have stolen the election, fraudulently swinging the vote by 8.7 percent. According to better-informed reports, however, two distinct exit polls were conducted. One of these, organized by the right-wing U.S. think-tank Freedom House and the U.S. Democratic Party’s National Democratic Institute (NDI), and carried out by the Kyiv Democratic Initiatives Foundation,10 perhaps as part of a group calling itself the Exit Poll consortium,11 found that Yushchenko won 54 percent of the vote to Yanukovych’s 43 percent. (It is perhaps this poll that the University of British Columbia’s Centre for Public Opinion and Democracy is referring to in its claim that “an exit poll conducted by independent research firms” showed Yushchenko to have won by 54 to 42 percent.)12 The other national exit poll, based on interviews rather than questionnaires, was conducted by SOCIS Company and the Social Monitoring Center, and gave Yushchenko 49.4 percent of the vote to Yanukovych’s 45.9 percent.13

It is not my purpose to attempt an unraveling of the complexities of the Ukrainian election. The British Helsinki Human Rights Group has challenged the validity of the exit polls, claiming that the exit pollsters they observed in one city were open Yushchenko supporters and were not following proper procedures.14 While Western observers have reported major irregularities in the government’s conduct of the election, Michel Chossudovsky and Ian Traynor have on the other hand adduced strong evidence of interventions in the Ukrainian electoral process by U.S. governmental and quasi-governmental agencies that resemble the same agencies’ interventions in Serbia, Georgia, Belarus, and Venezuela.15 The voter turnout figures of 96 percent recorded in Yanukovych strongholds in eastern Ukraine are strongly indicative of fraud; so likewise may be “the 90% pro-Yushchenko results declared in western Ukraine,” where the British Helsinki Group observed that Yushchenko’s opposition party “exercised disproportionate control over the electoral process in many places.”16 In what seems to me a well-balanced assessment, Dave Lindorff describes Leonid Kuchma’s outgoing regime as “corrupt and dictatorial,” and sees the opposition’s vehement protests against electoral corruption as being largely “indigenous and heartfelt”—while also noting that “the CIA and various American ‘pro-democracy’ front groups [are] playing a crucial hand in destabilizing the pro-Russian regime.”17 To which Timothy Garton Ash would respond with the proverb about not seeing the forest for the trees: whatever “bad trees” may be visible, “the shape of the wood” is that “An election was stolen.”

I would like merely to suggest that the interview-based exit poll which gave Yushchenko a 3.5 percent lead over Yanukovych—and hence indicated an irregular swing of 6.2 percent in the latter’s favour in the vote tallies—is more likely to have been properly conducted than the exit poll which was organized by Freedom House and the NDI, and which may well have been marked by Súmate-type improprieties.

     c. The United States

Let us turn to the American presidential election, where the same kind of data has encouraged similar suspicions—though thanks to the soothing ministrations of the U.S. corporate media, with nothing resembling the massive public outcry in Ukraine. George W. Bush was hailed the winner on November 2, with 51 percent of the vote to John Kerry’s 48 percent. But there are good reasons to be skeptical of the official vote tallies. The last wave of national exit polls published on the evening of November 2—polls which appear to have been duly weighted to correct for sampling imbalances—showed Kerry, not Bush, leading by 51 to 48 percent.18 A divergence of 6 percent between weighted exit polls and the official numbers is a strong indicator of electoral fraud.

At the decisive point, moreover, the divergence between the exit poll results and the vote tally was wider still.19 Prior to the election, political analysts identified Florida, Ohio, and Pennsylvania as the three key swing states: the candidate who carried these states, or a majority of them, would win the election.

Bush won Florida, with 52.1 percent of the vote to Kerry’s 47.1 percent. (This tally, by the way, diverges by 4.9 percent in Bush’s favour from the state exit poll, which gave Bush a paper-thin 0.1 percent lead.) Kerry won Pennsylvania, with 50.8 percent of the vote to Bush’s 48.6 percent. (Here again the vote tally differs in Bush’s favour from the exit poll results—this time by 6.5 percent.)

That left Ohio as the deciding state, the one on which the national election results depended. George W. Bush won Ohio, according to the official vote tally, with 51 percent of the vote to John Kerry’s 48.5 percent. The divergence in this case between the vote tally and the exit poll, which showed Kerry as winning by 52.1 percent to Bush’s 47.9 percent, is fully 6.7 percent.

Is it possible that these three divergences in Bush’s favour between exit polls and vote tallies could have occurred by chance? I wouldn’t bet on it. Dr. Steven Freeman of the University of Pennsylvania’s Center for Organizational Dynamics has calculated that the odds against these statistical anomalies occurring by chance are 662,000 to 1.20

Or are exit polls perhaps just not as reliable as people think? Dr. Freeman has an answer to this question as well. In the last three national elections in Germany, the differential between the exit polls and the vote tallies was, on average, 0.27 percent; and in the last three elections to the European Parliament, the differential in Germany was 0.44 percent.21 Exit polls conducted professionally and without political bias are highly accurate—which is why they have been used (more honestly in most cases, it would seem, than in the Penn, Schoen & Berland poll in Venezuela and the Freedom House-NDI poll in Ukraine) as a measure of electoral integrity in situations where improprieties have been anticipated. The U.S. exit polls were conducted by Mitofsky International, a survey research company founded by Warren J. Mitofsky, who as the company’s website proclaims “created the Exit Poll research model” and “has directed exit polls and quick counts since 1967 for almost 3,000 electoral contests. He has the distinction of conducting the first national presidential exit polls in the United States, Russia, Mexico and the Philippines. His record for accuracy is well known.”22

The fact that Mitofsky International systematically altered the U.S. presidential exit poll data early on the morning of November 3, contaminating the exit poll figures by conflating them with the vote tally percentages, has quite rightly become a matter of controversy.23 But there seems no reason to doubt that the Mitofsky exit poll data made available by the CNN website on the evening of November 2 was professionally gathered and weighted.

Mightn’t one propose, as a last resort, that Bush’s election-winning divergence of 6.7 percent between the Ohio exit poll results and the Ohio vote tally was, at any rate, somewhat less scandalous than the 13.7 percent swing Yanukovych’s party was blamed for by the Freedom House-NDI exit poll? (Ignore, if you like, the lesser 6.2 percent swing indicated by the SOCIS and Social Monitoring exit poll—which, if accurate, shows the Freedom House-NDI poll to be skewed in Yushchenko’s favour by fully 7.5 percent.) But if stealing elections is like knocking off banks, the fact that one practitioner can dynamite the vault of the central bank and get away with it, while his less fortunate compeer draws unwanted attention by blowing out all of the windows of the neighbourhood Savings-and-Loan, doesn’t make the former any less a bank robber than the latter.

The parallels between the Ukrainian and the U.S. presidential elections extend beyond the exit poll divergences. Ballot-box stuffers appear to have achieved a 96 percent turnout in parts of eastern Ukraine, with turnout figures in some areas exceeding 100 percent. There is evidence of similar indiscretions on the part of Bush’s electoral fraud teams. Twenty-nine precincts in a single Ohio county reported more votes cast than there are registered voters—to a cumulative total of over 93,000 votes.24 And in six Florida counties the total number of votes reported to have been cast exceeded by wide margins the total number of registered voters.25 Senator John McCain, manifesting the same stunning lack of irony as other Republican spokesmen, has weighed in on the issue: “IRI [the International Republican Institute] found that in a number of polling stations, the percentage of votes certified by the Central Election Commission exceeded 100% of total voters. This is simply disgraceful.”26 McCain is of course referring to eastern Ukraine; when it comes to Florida or Ohio, he keeps his eyes wide shut.

The question of advance indications of electoral fraud offers a final point of comparison. In the United States, as in Ukraine (where international observers described the polls and vote-counts in previous elections as deeply flawed), electoral fraud was widely anticipated prior to the 2004 presidential election.

As the materials itemized in the first three sections of the Reading List that is being published to accompany this article make clear, the electronic voting technologies in use in the U.S. were widely denounced by electronic security experts months and even years in advance, as permitting, indeed facilitating, electoral fraud; there is clear evidence that the 2000 election and the 2002 mid-term elections were marked by large-scale fraud on the part of the Bush Republicans; and U.S. computer scientists and informed analysts warned insistently that fraud on an unprecedented scale was likely to occur in this year’s election.27

How has it been possible for the massive ironies arising out of the similarities between the elections in the U.S. and Ukraine to pass unobserved in the corporate media. Have the media been simple-mindedly buttering their bread on both sides? If so, it is a habit that makes for messy eating. On November 20, an article in The Washington Post informed those who might question the U.S. election that “Exit Polls Can’t Always Predict Winners, So Don’t Expect Them To.”28 Two days later, The Washington Post carried breaking news of the early election results from Ukraine—and quoted a purported election-stealer who holds exactly the same opinion of exit polls: “‘These polls don’t work,’ said Gennady Korzh, a spokesman for Yanukovych. ‘We will win by 3 to 5 percent. And remember, if Americans believed exit polls, and not the actual count, John Kerry would be president’.”29


2. Key Issues and Evidence in the U.S. Presidential Election

Mainstream media assessments of the integrity of the 2004 U.S. presidential election have tended to focus on particular and local problems—computer errors or ‘glitches’ for the most part—that came to light on the day of the election or shortly afterwards. Naturally enough, the fact that these problems were noticed, and in some cases corrected, works if anything to enhance public confidence in the integrity of the electoral system.

The stance of the mainstream media is inadequate in at least two respects. First, some of the ‘problems’ were not mere accidents, but open and flagrant violations of democratic principles. Prominent among these was the election-night ‘lockdown’ of the Warren County, Ohio administrative building, on wholly spurious grounds of a ‘terrorist threat’: as a result, the public, the press, and the local legal counsel for the Kerry-Edwards campaign were prevented from witnessing the vote count.30 This maneuver generated widespread outrage: Warren County’s Republicans may perhaps have ‘misoverestimated’ the degree to which previous conveniently timed ‘terror alerts’ and Osama bin Laden’s late-October Jack-in-the-Box act had tamed the electorate.

But more importantly, while ‘problems’ and ‘glitches’ have commonly been covered by the corporate media as local issues, they can be recognized as belonging to a larger pattern. As James Paterson’s compelling analysis of “The Theft of the 2004 US Election” makes clear, Republican intentions were evident well before the election.31 And as Joseph Cannon has remarked, “An individual problem can be dismissed as a glitch. But when error after error after error favors Bush and not a single ‘accident’ favors Kerry, we’ve left glitch-land.”32

There is widespread evidence, which goes well beyond any mere accumulation of local problems, that “glitch-land” is indeed far behind us. The landscape to which the 2004 U.S. presidential election belongs includes the murky swamps of Tammany Hall-style election-fixing—and the still more sinister morasses of ‘Jim Crow’ as well.

It has been reported that Republican-controlled counties in Ohio and elsewhere sought to reduce the African-American vote by deliberately curtailing the numbers of polling stations and voting machines in working-class precincts: large numbers of would-be voters were effectively disenfranchised by line-ups that were many hours long.33 The Republican Party’s purging of African Americans from voters’ lists gained the 2000 election for George W. Bush;34 as informed observers had anticipated,35 this shameful illegality was repeated in 2004 on a wider scale.

Large-scale polling-station challenges were used to further slow the voting, and to turn the new provisional ballots into a mechanism for effectively disenfranchising minority voters. In the swing state of Ohio this year, it appears that fully 155,000 voters—most of them African-Americans—were obliged as a result of polling-station challenges to cast provisional ballots.36 Although it is becoming clear that the great majority of these citizens were legally entitled to vote,37 the likelihood that their votes will be fairly counted, or that Ohio’s Republican Secretary of State Ken Blackwell will permit them to be included in the official tally, remains slender.

The effect of this Jim Crow mechanism appears to be compounded by racially-biased judgments of ballot spoilage. As Greg Palast reports, 54 percent of all ballots judged ‘spoiled’ in the 2000 election in Florida were cast by African-American voters, and similarly scandalous percentages are expected in key states this time round. Nor have African Americans been the sole victims of these tactics: it appears that in New Mexico, where Hispanics’ ballots are five times more likely to be laid aside as ‘spoiled’ than those of white voters, 13,000 Hispanics were effectively disenfranchised by means of provisional ballots.38 Bush won New Mexico by less than half that number of votes.

But it is the co-presence of other forms of corruption, in addition to all these, that establishes the difference between an election dirtied by illegalities, and one that was not merely soiled and distorted by fraud but actually stolen. The evidence presented within the texts listed in my “Evidence of Fraud” bibliography suggests with gathering strength that the Karl Rovian maneuvers alluded to above were supplemented on November 2, 2004 by less conspicuous—and yet decisive—manipulations of the machines that recorded and tabulated the votes.

How precisely this apparent manipulation may have been carried out in different jurisdictions—by rigging machines in advance to mis-record or delete votes, by configuring proprietary software so as to allow ‘back-door’ access for unrestrained vote-tampering, or by hacking into the notoriously insecure vote-tabulation systems—remains as yet undetermined. However, the evidence has been coming to light with surprising rapidity.

As observers and analysts noted at once, troubling discrepancies were apparent between the exit poll results published by CNN on the evening of November 2 and the official vote tallies.39 No less disturbing, as I observed in my article on the subject, is the fact that the exit poll data was systematically tampered with early on November 3 to make the figures conform to the vote tallies. At 1:41 a.m. EST on November 3, for example, the Ohio exit poll was altered: Kerry, who had previously been shown as leading Bush by 4 percent in that state, was now represented in the revised exit poll as trailing him by 2.5 percent. And yet the number of respondents in the poll had increased from 1,963 to only 2,020. An additional 57 respondents—a 2.8 percent increase—had somehow produced a 6.5 percent swing from Kerry to Bush. At 1:01 a.m. EST on November 3, the Florida exit poll was likewise altered: Kerry, who had previously been shown in a near dead heat with Bush, now trailed him by 4 percent. In this case, the number of respondents rose only from 2,846 to 2,862. A mere 16 respondents—0.55 percent of the total—produced a 4 percent swing to Bush.40

However, the key exit-poll issue remains the divergence between the November 2 exit polls and the vote tallies. Steven Freeman concluded, in the first draft of his judicious study of the November 2 exit poll data, that “Systematic fraud or mistabulation is a premature conclusion, but the election’s unexplained exit poll discrepancies make it an unavoidable hypothesis, one that is the responsibility of the media, academia, polling agencies, and the public to investigate.”41

Other evidence points toward a strengthening, indeed to a substantial confirmation of this “unavoidable hypothesis” of systematic fraud. Some of this evidence has been emerging from the swing state of North Carolina, and from the two key swing states of Florida and Ohio—either one of which, had John Kerry won it, would have made him the acknowledged President-elect.

In North Carolina, the tell-tale marks of electronic electoral fraud have been brought to light by an analyst who publishes at the Democratic Underground site under the name of ‘ignatzmouse.’ (“Ignatz,” remember, is the name of the mouse who in the Krazy Kat cartoons smacks the unhappy cat with the inevitable brick. That pesky mouse is once again on target.)

What gives the game away in the North Carolina election data is the disparity within the presidential and senatorial vote-counts between the so-called “absentee” votes—a category that apparently includes the early voting data as well as votes cast by citizens living abroad and military personnel—and the polling-day votes cast on November 2.

In the race for Governor, 30 percent of the votes cast for the Republican and the Democratic candidate alike were absentee votes; the other 70 percent were cast on November 2. The Democrat won with 55.6 percent of both the absentee and the polling-day votes. In most of the other statewide races in the North Carolina election there were similarly close correlations between absentee and polling-day votes. For example, Democrats won the post of Lieutenant Governor, with 55.7 percent of absentee and 55.5 percent of polling-day votes; the post of Secretary of State, with 58 percent of absentee and 57 percent of polling-day votes; and the post of Attorney General, with 56.7 percent of absentee and 55.2 percent of polling-day votes. In three other statewide races, and in the voting for three constitutional amendments, the correlation between absentee and polling-day votes remains very close (though tight races for three other positions in the state administration were won by Republicans with polling-day swings in favour of the Republican candidates of 4.2, 5.2, and 5.4 percent respectively).

Given the close correlations between absentee and polling-day votes in ten of the thirteen statewide races, the senate result looks suspicious: the Democrat’s narrow lead in the absentee voting became a clear defeat on November 2, with a 6.4 percent swing in the polling-day votes to the Republican. And the presidential results look more seriously implausible. In the absentee votes, Kerry trailed by 6 percent, a result that ‘ignatzmouse’ remarks “is consistent with the pre-election polls and most importantly with the exit polls of November 2nd.” But in the election day voting, there was a further swing of fully 9 percent to Bush. Bush led in the absentee votes (30 percent of the total) by 52.9 percent to Kerry’s 46.9 percent; but on polling day he took 57.3 percent of the remaining votes, while Kerry received 42.3 percent. In the absence of any other explanation, these figures point to electronic fraud—and, more precisely, to “a ‘date-specific’ alteration in the software, a hack, or a specific [software] activation just prior to the election.”42

The Florida evidence is, if anything, more flagrant. On November 18, Professor Michael Hout of the University of California at Berkeley released a statistical study indicating that electronic voting technology had produced a very substantial distortion of the presidential vote tally in Florida. According to the analyses conducted by Hout and his team, irregularities associated with electronic voting machines accounted for at least 130,000 votes in Bush’s lead over Kerry in Florida—and possibly twice that much. (The uncertainty stems from the fact that the machines may have awarded Bush “ghost votes” which increased his tally without reducing Kerry’s, or they may have misattributed Kerry votes as Bush votes. As Hout explains, the disparities “amount to 130,000 votes if we assume a ‘ghost vote’ mechanism and twice that—260,000 votes—if we assume that a vote misattributed to one candidate should have been counted for the other.”)43

Hout’s results have not gone unchallenged;44 obviously enough, the validity of statistical analyses depends on the extent to which all possible causal factors have been accounted for. But other data indicates that the ‘haunting’ of Florida’s electronic voting tabulators was if anything more serious than Hout and his associates believe. As I have already noted, in six Florida counties the number of votes purportedly cast exceeded the number of registered voters—by a cumulative total of 188,885.45 These are apparently “ghost votes,” and unless we’re willing to assume a level of electoral participation resembling those claimed by totalitarian states like Ceaucescu’s Romania or Saddam Hussein’s Iraq, a significant percentage of the other votes cast in these counties must also represent the electoral choice not of human beings but of Republican hackers.

Further evidence which may help to identify the agents involved in Florida’s electronic voting fraud has in fact begun to emerge. Brandon Adams, for example, has noted striking divergences among Florida voters according to the makes and models of the voting machines they used in different counties;46 and a heavy hacking of vote-tabulation systems used in conjunction with the older optical-scan voting machines is now well-established.47

Moreover, statistically-based work is being complemented by acquisitions of direct material evidence. In Volusia County, one of Florida’s six most seriously ‘haunted’ counties, where 19,306 more votes were cast than there are registered voters, Bev Harris’s BlackBoxVoting team caught county election officials red-handed on November 16 in the act of trashing original polling-place tapes which BlackBoxVoting had asked for in a Freedom of Information request. In addition to filming the behaviour of county officials, her team was able to establish that some copies of the tapes that officials had prepared to give them in response to the Freedom of Information Act request had been falsified in favour of George W. Bush—in one precinct alone by hundreds of votes.48 The Volusia County materials provide proof, moreover, that the GEMS central vote-tabulation system, which was supposedly “stand-alone” and non-networked, was remotely accessed during the election.49

Ohio, remember, was the deciding state. John Kerry conceded the election after calculating that the some 155,000 provisional ballots cast in Ohio would not suffice—even if they were properly counted, and even if, as expected, they were very largely cast by Kerry supporters—to overturn the tallied results, according to which Bush had won the state by 136,483 votes.

However, the exit poll data indicates that it was Kerry who won the state, and by a comfortable margin. Once again, there is substantial evidence of electronic electoral fraud. Teed Rockwell found, after careful study of the Cuyahoga County Board of Elections website, that twenty-nine precincts in this county “reported votes cast IN EXCESS of the number of registered voters—at least 93,136 extra votes total.” The same website he studied also repays further study, for Rockwell’s tallying of ‘ghost votes’ is in fact conservative.50 To cite just one example, Brook Park City is listed as having 14,491 registered voters, of whom it is claimed that fully 14,458 exercised their civic duty and cast ballots—for a turn-out rate of 99.4 percent. I leave it to the curious to discover how many of these high-minded but possibly nonexistent citizens supported their incumbent President.

Those who want to pursue the questions of vote fraud and suppression in Ohio may also want to consult the studies carried out by Richard Philips, whose work, together with the data available on the websites of Cuyahoga and other counties, provides depressing evidence of successful vote suppression in urban precincts. (It has been estimated that vote suppression tactics may have cost Kerry 45,000 votes across the whole state of Ohio.)51

The Green Party and Libertarian Party presidential candidates, belatedly followed by the Kerry/Edwards campaign, have called for a recount in Ohio. But if Ohio’s Republican Secretary of State Blackwell permits no more than a recount, without a rigorous audit of the electronic voting machines and tabulators as well, the numbers for a reversal of the election results are probably not there. On the optimistic assumption that a fair count of the 155,000 provisional ballots would result in 10 percent of them being disqualified and 70 percent of the remainder being validated as Kerry votes, those ballots might reduce Bush’s lead in Ohio by as much as 55,800 votes. However, it seems unlikely that a recount, including a re-examination of the more than 96,000 Ohio votes (most of them cast on old punch-card machines) that were discarded as spoiled, would turn up the almost 81,000 additional Kerry votes that would still be needed.

Together with the principle that every duly cast vote must be counted, advocates for democracy need to assert another complementary principle: the principle that votes cast not in polling booths, but in the hard drives of voting-tabulation machines; and not by citizens, but rather by ghosts summoned into existence by Republican hackers’ nimble fingers, have no business getting counted, and should be removed from the tally.

The effect of turning a ‘Ghostbuster’ computer-auditing team like Bev Harris’s BlackBoxVoting organization loose on the Ohio results, to carry out a serious audit of any polling precinct and computer-log data that hasn’t already been quietly destroyed, might well be startling. For while a simple recount would probably leave Kerry trailing by several tens of thousands of votes, a thorough computer-audit ‘exorcism’ of the vote tallies, should such a thing ever be permitted, might well lead to a reversal of the national election results.

Whatever the finally certified results may be, a larger informing context should not be forgotten. The regime of George W. Bush has made no secret of its scorn for the American Constitution and Bill of Rights, its hostility to any notion of international law, its contemptuous dismissal of the decent opinion of humankind both at home and abroad, its contempt, in the most inclusive sense, for truth.

Bush has claimed that the 2004 election gave him “capital”—which he now will not hesitate to spend. An early instance of this expenditure has been the assault on the city of Fallujah, and a compounding of the manifold war crimes of which Bush and those who serve him are already guilty.

But what is this “capital”? As the evidence is revealing with growing clarity, the 2004 presidential election was not in fact a victory for Bush, but rather the occasion for an insolent usurpation.

A ‘president’ who takes office through fraud and usurpation can make no legitimate claim to exercise the stolen power of his office.

As the knowledge of his offence becomes ever more widely disseminated, he may yet come, like Shakespeare’s Macbeth, “[to] feel his title / Hang loose upon him, like a giant’s robe / Upon a dwarfish thief.”




1  See “Ukraine cities defy poll result,” BBC News (22 November 2004),; and “In quotes: World concern at Ukraine election,” BBC News (23 November 2004),

2  David Rosnick, “Polling and the Ballot: The Venezuelan Referendum,” Center for Economic and Policy Research (19 August 2004),

3  Jonah Gindin, “Venezuela's Opposition Resorts to Phony Exit Polls,” (15 August 2004),

4  Michel Chossudovsky, “IMF Sponsored 'Democracy' in The Ukraine,” Centre for Research on Globalization (28 November 2004),

5  See “Súmate,”

6  For documentation derived from Freedom of Information Act requests, see

7  See Philip Stinard, “Governor Enrique Mendoza: Greenberg's man for the Venezuelan Presidency,” Vheadline (4 July 2004),

8  See Anders Aslund, “Ukraine's Future and U.S. Interests: Testimony before the U.S. House of Representatives Committee on International Relations, Subcommittee on Europe,” Carnegie Endowment for International Peace (12 May 2004),; Larry Chin, “Cold War Crisis in the Ukraine: Control of Oil: Key Grand Chessboard 'Pivot' at Stake,” Centre for Research on Globalization (26 November 2004),; and Olga Oliker, “Ukraine and the Caspian: An Opportunity for the United States,” Issue Paper 198, RAND Center for Russia and Eurasia (2000),

9  Michel Chossudovsky, War and Globalization: The Truth Behind September 11 (Shanty Bay, ON: Global Outlook, 2002), pp. 65-75.

10  Alexsandar Vasovic, “Ukraine's Early Results, Exit Polls Differ,” Associated Press (21 November 2004), available at Findlaw,

11  John Kubiniec, “Election Fraud in Ukraine Presidential Vote,” Freedom House (22 November 2004),

12  “Ukraine,” Centre for Public Opinion and Democracy, The University of British Columbia,

13  SOCIS and the Social Monitoring Center had previously collaborated in exit polling; see “Press Release” [Exit Polls in the March 2002 Election to be conducted by the Kiev Institute of Sociology, SOCIS Company, and the Social Monitoring Center, coordinated by the Democratic Initiatives Foundation],

14  “Ukraine: 2nd Round of Presidential Election,” British Helsinki Human Rights Group,

15  See Chossudovsky, “IMF Sponsored 'Democracy' in The Ukraine”; and Ian Traynor, “US campaign behind the turmoil in Kiev,” The Guardian (26 November 2004), available at the Centre for Research on Globalization (28 November 2004),

16  See “Ukraine: 2nd Round of residential Election.”

17  Dave Lindorff, “Double Standards on Exit Polls: Hypocrisy Sans Irony,” CounterPunch (24 November 2004),

18  See 'Mystery Pollster' [Mark Blumenthal], “Exit Polls: The NEP [National Election Pool] Report,” Mystery Pollster (5 November 2004),

19  Steven F. Freeman, “The Unexplained Exit Poll Discrepancy: Part I,” Working Paper #04-10, Center for Organizational Dynamics, Graduate Division, School of Arts and Sciences, University of Pennsylvania, Buzzflash (21 November 2004),; also available at Scoop (23 November 2004),

20  Ibid.

21  Ibid.

22  “National Election Pool (ABC, AP, CBS, CNN, Fox News, NBC News), Conducted by Edison/Mitofsky,”

23  See Michael Keefer, “Footprints of Electoral Fraud: The November 2 Exit Poll Scam,” Centre for Research on Globalization (5 November 2004),; and Keith Olbermann, “Zogby vs. Mitofsky,” Bloggerman (24 November 2004),

24  Teed Rockwell, “93,136 EXTRA Votes Found in ONE Ohio County,” (19 November 2004), [As indicated in the note at the head of this article, the indication of large numbers of 'ghost' voters in Cuyahoga County was subsequently revealed to be mistaken; it arose because of the misleading manner in which election results were posted by the Cuyahoga County Board of Elections. My essay “The Strange Death of American Democracy: Endgame in Ohio,” Centre for Research on Globalization (24 January 2005),, reported other cases of impossibly high certified voter-turnout figures in two pro-Bush Ohio counties, Perry County and Miami County—as well as impossibly low certified turnout figures in predominantly African-American precincts in Cuyahoga County.]

25  Stirling Newberry, “The Voters are Restless: Election Fraud Story circulates the Internet,” BOPNews (n.d.),; and “Not 'Was It Stolen', but 'Was it Stealable',” BOPNews (11 November 2004), In the state of Wyoming the number of votes cast likewise exceeded the number of registered voters—but in that case there are no evident grounds for suspicion. As the Wyoming Secretary of State’s website explains, “There were 232,396 registered voters in Wyoming prior to the [2004] General Election and 245,789 voted. This was possible because Wyoming state statute allows voters to register and vote at the polls on General Election Day.” (Wyoming’s current voting age population numbers 376,359, of whom 232,366 [62 percent] were registered; 245,789 people voted [106 percent of registered voters], for an overall turnout rate of 65 percent of eligible voters.) In Florida and Ohio, however, election-day registration is not permitted. According to Florida’s rules, “The registration books will be closed on the 29th day before each election and will remain closed until after that election. You must be registered for at least 29 days before you can vote in an election.” Ohio’s requirements are similar: “You must be a U.S. citizen and a resident of Ohio for at least thirty (30) days before the election and you must have registered to vote in Ohio at least thirty (30) days before the election.” Within these states, therefore, a surplus of votes cast over and above the number of registered voters is a clear indicator of electoral irregularities.

26  “McCain Statement on Elections in Ukraine,” International Republican Institute (23 November 2004),

27  See Michael Keefer, “Evidence of Fraud in the 2004 U.S. Presidential Election: A Reader,” Centre for Research on Globalization (5 December 2004),

28  Richard Morin, “Surveying the Damage: Exit Polls Can't Always Predict Winners, So Don't Expect Them To,” The Washington Post (20 November 2004),

29  Peter Finn, “Partial Vote Results Show a Tight Race in Ukraine Runoff,” The Washington Post (22 November 2004),

30  Erica Solvig and Dan Horn, “Warren Co. defends lockdown decision; FBI denies warning officials of any special threat,” The Cincinnati Enquirer (10 November 2004),  ; and Keith Olbermann, with Richard Engel and Jim Miklaszewski, “Did Your Vote Count? The Plot Thickens,” Countdown, MSNBC (8 November 2004), transcript and video stream available under the title “November 2nd: Voter Fraud and Homeland Security Terror Threat 'Advisories' in Ohio and Florida: Fraud on a massive scale is now corroborated by Network TV,” Centre for Research on Globalization,

31  James Paterson, “The Theft of the 2004 US Election,”,

32  Joseph Cannon, “The empire strikes back: Data and disinformation,” Cannonfire (12 November 2004),

33  Bob Fitrakis, “None dare call it vote suppression and fraud,” The Free Press (7 November 2004),; Fitrakis, “Document reveals Columbus, Ohio voters waited hours as election officials held back election machines,” The Free Press (16 November 2004),; Fitrakis, “How the Ohio election was rigged for Bush,” The Free Press (22 November 2004),

34  John Conyers, Jr., and Democratic Investigative Staff, House Committee on the Judiciary, How to Make Over One Million Votes Disappear: Electoral Sleight of Hand in the 2000 Presidential Election. A Fifty-State Report Prepared for Rep. John Conyers, Jr. (Ranking Member, House Committee on the Judiciary; Dean, Congressional Black Caucus (Washington, DC: U.S. House of Representatives, August 20, 2001).

35  Greg Palast, “Electoral Fraud, Ethnic Cleansing of Voter Rolls, An Election Spoiled Rotten,” (1 November 2004),; available at the Centre for Research on Globalization (4 November 2004),; Martin Luther King III, and Greg Palast, “Jim Crow Revived in Cyberspace,” The Baltimore Sun (8 May 2003); available at

36  Greg Palast, “Kerry Won Ohio: Just Count the Ballots at the Back of the Bus,” In These Times (12 November 2004),; also published as “Most voters in Ohio chose Kerry; here's how the votes vanished,” OpEdNews (15 November 2004), See also David Solnit, “Massive Vote Suppression and Corruption in Ohio,” Centre for Research on Globalization (3 November 2004),

37  Mark Williams, “Ohio Provisional Ballots Seem Legitimate,” Associated Press (17 November 2004); available at Truthout (18 November 2004),

38  Palast, “Kerry Won Ohio.”

39  See Faun Otter, “Vote Fraud—Exit Polls Vs Actuals,” Scoop (4 November 2004),; Jonathan Simon, “47 State Exit Poll Analysis Confirms Swing Anomaly,” Scoop (11 November 2004); available at; David Dodge, “Response to MIT/Caltech,” (13 November 2004),; Sara S. DeHart, “Something is Rotten in Denmark: Exit poll data in former Soviet Republic of Georgia vs. USA,” Online Journal (17 November 2004),; and Steven Freeman, “The Unexplained Exit Poll Discrepancy.”

40  Keefer, “Footprints of Electoral Fraud.”

41  Steven F. Freeman, “The Unexplained Exit Poll Discrepancy” [First Draft], Buzzflash (11 November 2004),

42  'ignatzmouse,' “Unfficial Audit of NC Election: Comprehensive Case for Fraud,” Democratic Underground (12 November 2004),

43  Michael Hout et al., with the assistance of the UC Berkeley Quantitative Methods Research Team, “The Effect of Electronic Voting Machines on Change in Support for Bush in the 2004 Florida Elections,” Working Paper, UC Data (Data Archive & Technical Assistance), UC Berkeley Survey Research Center (18 November 2004),

44  Alex Strashny, “Working Paper: The lack of effect of electronic voting machines on change in support for Bush in the 2004 Florida elections,” Institute for Mathematical Behavioral Sciences, University of California, Irvine, available at (21 November 2004),

45  Newberry, cited in n. 25 above.

46  Brandon Aams, “An Examination of the 2004 Elections,”

47  Paterson, “The Theft of the 2004 US Election”; Bev Harris, “The Tampering of Electronic Voting Systems on November 2nd,” BlackBoxVoting (7 November 2004),, available at the Centre for Research on Globalization (8 November 2004),

48  Bev Harris, “Vote Fraud, Volusia County on Lockdown. County election records just got put on lockdown: Duelling lawyers, election officials gnashing teeth, film crew catching it all,” Scoop (16 November 2004),; Thom Hartmann, “'Stinking Evidence' of Possible Election Fraud Found in Florida,” Scoop (19 November 2004),

49  Bev Harris, “Dems Pocket $52 Million, CNN Ignores Evidence, and Officials Stonewall … What Vote Fraud?” (24 November 2004),

50  The website is that of the Cuyahoga County Board of Elections,

51  David S. Bernstein, “Questioning Ohio: No controversy this time? Think again,” The Boston Phoenix (12-18 November 2004),