VOTE  (CONT)

"The system kept running through the early morning hours as users from around the world looked to Ohio for their election results."

All the facts are not in, but enough is known to warrant a serious congressional inquiry. Beginning with a timeline on Election Night after a national media consortium exit poll predicted Democrat John Kerry would win Ohio, the first Ohio returns were from the state's Democratic urban strongholds, showing Kerry in the lead.

This was the case until shortly after midnight on Wednesday, Nov. 3, when for roughly 90 minutes the Ohio election results reported on the Secretary of State's website were frozen. Shortly before 2am EST election returns came in from a handful of the state's rural Republican enclaves, bumping Bush's numbers over the top.

It was known Bush would carry rural Ohio. But the vote totals from these last-to-report counties, where Karl Rove said there was an unprecedented late-hour evangelical vote giving the White House a moral mandate, were highly improbable and suggested vote count fraud to pad Bush's numbers. Just how flimsy the reported GOP totals were was not known on Election Night and has not been examined by the national media. But an investigation by the House Judiciary Committee Democratic staff begun after Election Day 2004 and completed before the Electoral College met on Jan. 6, 2005, was first to publicly point to vote count fraud in rural Ohio.

That report, "Preserving Democracy: What Went Wrong in Ohio," cited near-impossible vote totals, including 19,000 votes that were mysteriously added at the close of tallying the vote in Miami County. The report cited more than 3,000 apparently fraudulent voter registrations - all dating back to the same day in 1977 in Perry County. The report noted a homeland security emergency was declared in Warren County, prompting its ballots to be taken to a police-guarded unauthorized warehouse and counted away from public scrutiny, despite local media protests.

In our book, "What Happened in Ohio: A Documentary Record of Theft and Fraud in the 2004 Election" (The New Press, 2006), we go beyond the House Judiciary Democratic report to analyze precinct-by-precinct returns and we print copies of the documents upon which we base our findings. We found many vote-count irregularities based on examining the certified results, precinct-level records and the actual ballots.

The most eyebrow-raising example to emerge from parsing precinct results was finding 10,500 people in three Ohio's 'Bible Belt' counties who voted to re-elect Bush and voted in favor of gay marriage, if the official results are true. That was in Warren, Butler and Clermont Counties. The most plausible explanation for this anomaly, which defies logic and was not seen anywhere else in the country, was Kerry votes were flipped to Bush while the rest of the ballot was left alone. While we have some theories about how that might have been done by hand in a police-guarded warehouse, could full Republican control of the vote-counting software and servers also have played a role?

The early returns on the Secretary of State's website suggest Blackwell's vote-tallying and reporting system could manipulate large blocks of votes. Screenshots taken during the early returns in Hamilton County, where Cincinnati is located, gave Green Party presidential candidate David Cobb 39,541 votes, which was clearly incorrect. Similarly, early return screenshots in Lucas County, where Toledo is located, gave Cobb 4,685 votes, another clear error. (The screenshots are in our book). Were these innocent computer glitches or was a GOP vote-counting and reporting system moving and dumping Kerry votes?

There's more evidence the late returns from Ohio's Republican-majority countryside were not accurate. During the spring and summer of 2006, several teams of investigators associated with Freepress.org, notably one team led by Ron Baiman, a Ph.D. statistician and researcher at Chicago's Loyola University, examined the actual election records from precincts in Miami and Clermont Counties. These records - from poll books where voters sign in, to examining the actual ballots themselves - were not publicly accessible until last year, under orders from Ohio's former Republican Secretary of State. Baiman compared the number of voters who signed in with the total number of votes attributed to precincts. He found hundreds of "phantom" votes, where the number of voter signatures was less than the reported vote total. That discrepancy also suggests vote count fraud.

There was other evidence in the observable paper trail of padding the vote, including instances in Delaware County where in one precinct, 359 of the final punch-card ballots cast on Election Day contained no Kerry votes, which means the day's last voters all were Bush supporters, which also is improbable. In another Delaware County precinct, Bush allegedly received the last 210 votes of the day. Were partisan local election workers trying to mask what was happening electronically to tilt the vote count?