So any more, the idea of a fixed election in 2004 strikes me as completely plausible. After all, if it were at all possible, Bush and Rove would be precisely the people who would try it. Along these lines, Where We're Bound recently linked to a GAO report. The report is discussed elsewhere, inlcuding in the ever-reliable Consortium News. The findings are startling to say the least:
- The GAO report confirms that:
- Some electronic voting machines "did not encrypt cast ballots or system audit logs, and it was possible to alter both without being detected." In other words, the GAO now confirms that electronic voting machines provided an open door to flip an entire vote count. More than 800,000 votes were cast in Ohio on electronic voting machines, some seven times Bush's official margin of victory. "It was possible to alter the files that define how a ballot looks and works so that the votes for one candidate could be recorded for a different candidate." Numerous sworn statements and affidavits assert that this did happen in Ohio 2004.
- "Vendors installed uncertified versions of voting system software at the local level."
- Falsifying election results without leaving any evidence of such an action by using altered memory cards can easily be done, according to the GAO.
- The GAO also confirms that access to the voting network was easily compromised because not all digital recording electronic voting systems (DREs) had supervisory functions password-protected, so access to one machine provided access to the whole network. This critical finding confirms that rigging the 2004 vote did not require a "widespread conspiracy" but rather the cooperation of a very small number of operatives with the power to tap into the networked machines and thus change large numbers of votes at will. With 800,000 votes cast on electronic machines in Ohio, flipping the number needed to give Bush 118,775 could be easily done by just one programmer.
Access to the voting network was also compromised by repeated use of the same user IDs combined with easily guessed passwords. So even relatively amateur hackers could have gained access to and altered the Ohio vote tallies.
- The locks protecting access to the system were easily picked and keys were simple to copy, meaning, again, getting into the system was an easy matter.
- One DRE model was shown to have been networked in such a rudimentary fashion that a power failure on one machine would cause the entire network to fail, re-emphasizing the fragility of the system on which the Presidency of the United States was decided.
- GAO identified further problems with the security protocols and background screening practices for vendor personnel, confirming still more easy access to the system. In essence, the GAO study makes it clear that no bank, grocery store or mom & pop chop shop would dare operate its business on a computer system as flimsy, fragile and easily manipulated as the one on which the 2004 election turned.
The exit polls showed Kerry winning in Ohio, until an unexplained last minute shift gave the election to Bush. Similar definitive shifts also occurred in Iowa, Nevada and New Mexico, a virtual statistical impossibility.
A few weeks prior to the election, an unauthorized former ES&S voting machine company employee, was caught on the ballot-making machine in Auglaize County
Election officials in Mahoning County now concede that at least 18 machines visibly transferred votes for Kerry to Bush. Voters who pushed Kerry's name saw Bush's name light up, again and again, all day long. Officials claim the problems were quickly solved, but sworn statements and affidavits say otherwise. They confirm similar problems in Franklin County (Columbus). Kerry's margins in both counties were suspiciously low.
A voting machine in Mahoning County recorded a negative 25 million votes for Kerry. The problem was allegedly fixed.
In Gahanna Ward 1B, at a fundamentalist church, a so-called "electronic transfer glitch" gave Bush nearly 4000 extra votes when only 638 people voted at that polling place. The tally was allegedly corrected, but remains infamous as the "loaves and fishes" vote count.
In Franklin County, dozens of voters swore under oath that their vote for Kerry faded away on the DRE without a paper trail.
In Miami County, at 1:43am after Election Day, with the county's central tabulator reporting 100% of the vote - 19,000 more votes mysteriously arrived; 13,000 were for Bush at the same percentage as prior to the additional votes, a virtual statistical impossibility.
In Cleveland, large, entirely implausible vote totals turned up for obscure third party candidates in traditional Democratic African-American wards. Vote counts in neighboring wards showed virtually no votes for those candidates, with 90% going instead for Kerry.
Prior to one of Blackwell's illegitimate "show recounts," technicians from Triad voting machine company showed up unannounced at the Hocking County Board of Elections and removed the computer hard drive.
In response to official information requests, Shelby and other counties admit to having discarded key records and equipment before any recount could take place.
In a conference call with Rev. Jackson, Attorney Cliff Arnebeck, Attorney Bob Fitrakis and others, John Kerry confirmed that he lost every precinct in New Mexico that had a touchscreen voting machine. The losses had no correlation with ethnicity, social class or traditional party affiliation---only with the fact that touchscreen machines were used.
In a public letter, Rep. Conyers has stated that "by and large, when it comes to a voting machine, the average voter is getting a lemon - the Ford Pinto of voting technology. We must demand better." But the GAO report now confirms that electronic voting machines as deployed in 2004 were in fact perfectly engineered to allow a very small number of partisans with minimal computer skills and equipment to shift enough votes to put George W. Bush back in the White House.
Personally, I find the statistical discrepancies particularly odd. Having studied statistics, I know that the chance of large bodies of data running completely against statistical probabilities for no apparent reason is virtually impossible. (In a technical sense of course, it's always "possible" but so is someone's statistical ability to walk through a brick wall according to quantum physics.)