A Repeat of Election Fraud 2004 Must Be Prevented in 2006 and 2008

[Time for change]

A Repeat of Election Fraud 2004 Must Be Prevented in 2006 and 2008 –
Discussion of Different Types of Electronic Election Fraud in 04 & Means of Prevention

Though election fraud has undoubtedly played a part in every national election since the founding of our country, it has probably never posed as much threat to our democracy in the past as it does today. I say this because of the enormous potential for massive election fraud posed by the computerization of our elections (using secret software code) and because of the evidence indicating that it was used in 2000 and 2004 to install and maintain in office the most disastrous Presidential administration in our nation’s history.

There are many different types of election fraud, and most of them were used in the 2004 Presidential election. The main point of this post is to discuss three of these types, for which widespread computerization of our elections seems to have opened up the possibility of fraud on a scale never seen before: 1) Vote switching; 2) Central tabulator mediated fraud; and, 3) Voter registration purging fraud.

All three of these types of fraud must be prevented. If we are able to prevent vote switching fraud but fail to prevent the other two, we very well may end up with Jeb Bush as President until 2016.

That is not to say that these were the only types of fraud perpetuated in the 2004 election. Voter suppression, for example, was also perpetrated on a massive scale against the citizens of Ohio in 2004 – targeted exclusively to likely Democratic voters. Two sources of evidence for this are John Conyers’ Report of the Democratic Staff of the House Judiciary Committee and the transcripts of hearings such as this one. The types of voter suppression that were committed included failing to provide provisional ballots according to law (page 334 of Conyers Report), targeting minority voters for legal challenges (p337), failing to provide absentee ballots upon request, and then refusing to let those voters vote on Election Day (p 340), and a myriad of dirty tricks involving misinformation on such essential topics as where or when voters are supposed to vote. One especially effective means of voter suppression was the withholding of sufficient numbers of electronic voting machines from heavily Democratic precincts that tens of thousands of voters didn’t get to vote. This was done to such an extent in Columbus that a thorough analysis of this situation showed that it probably cost the Kerry/Edwards ticket about 17,000 net votes.

Because I feel that election fraud is such an important issue, and because I have limited expertise on most election fraud issues, I had some DUers (Bill Bored, OTOH, liam_laddie, adagiopop) with expertise in various election fraud areas review and provide me their comments on this before I posted it.

Now, back to the three types of election fraud that are the main subject of this post. For each of these types of fraud I will describe what it is, discuss the evidence for its use in the 2004 Presidential election, especially in Ohio (where the election was decided), and then discuss ideas on how it might be prevented in 2004:


Electronic vote switching

Description
Although vote switching can occur in many different ways, electronic vote switching with DRE machines poses the greatest danger for this type of election fraud. DRE stands for direct-recording electronic. The great danger with these machines is that they have the potential to switch a voter’s vote without being noticed by the voter. In other words, someone tries to vote for John Kerry, and the machine registers a vote for George Bush instead. What makes matters worse is that most of these machines don’t even produce a piece of paper with the vote on it, which can then later be used for a recount. In other words, recounts are impossible with these machines, so if fraud is suspected there is no recourse. And worse yet is the fact that most of these machines use proprietary (secret) code to determine who the voter voted for.

Evidence of fraud from the 2004 Presidential election
We know for a fact that vote-switching did occur in the 2004 election. One study, based on voter reports to the national Electronic Incident Reporting System (EIRS), showed that vote switching incidents favored Bush over Kerry by a ratio of 12 to 1 nationally. A similar study showed that these vote switching incidents that favored Bush were 9 times as common in the heavily contested “swing states” than in non-swing states. To make the point that the EIRS reports represent only a small fraction of actual Election Day problems, an investigation by the Washington Post identified about 25 electronic voting machines in Youngstown, Mahoning County, Ohio, that were said to have been switching votes all day long. Yet only eight incidents of this nature from Mahoning County (all in favor of Bush) were reported to EIRS that day. And here’s a specific kind of vote switching where votes for a straight party Democratic ticket were “accidentally” programmed to go to the Republican candidate.

Why did all this vote switching occur, why did almost all of it favor Bush, and why was it so much more prevalent in the swing states, especially in Florida? A clue to the answer to this question comes from Clint Curtis, a computer programmer working in Florida prior to the 2004 election. In testimony before the Democratic staff of the House Judiciary Committee, Curtis said that he was requested in 2000 by Tom Feeney to “develop a prototype of a voting program that could alter the vote tabulation in an election and be undetectable”. Tom Feeney, currently a U.S. Congressman from Florida, was Speaker of the Florida House of Representatives when Al Gore was fighting to have the Florida vote recounted. And it was Feeney who threatened to use his power as Speaker to submit an alternate slate of Florida electors if a recount of the Florida vote showed Al Gore to have won Florida’s electoral votes. So when Curtis testified that Feeney told Curtis’ supervisor that the computer program he asked Curtis to develop was needed to control the south Florida vote, you can bet that Curtis was telling the truth – especially since it was South Florida where half of the vote switching incidents in the country were reported from. And if that isn’t suspicious enough, consider the death of Raymond Lemme, the investigator who was looking into Curtis’ allegations, and who appeared to have made great headway in his investigation shortly before he unfortunately committed “suicide”.

NEVERTHELESS, I have to say this: It appears to me from the preponderance of evidence that of the three types of electronic fraud that I discuss in this post, vote switching was the least important in changing the results of the 2004 election, at least in Ohio, where the election was decided. The reason that I want to make a big point about this is that so much relative attention is being given to this mechanism for fraud that the other mechanisms are being relatively neglected. Electronic vote switching is a very scary thing, and we are right to be very concerned about it and work hard to prevent it. But in the process we must not neglect other important kinds of fraud.

What evidence do I speak of as suggesting that vote switching was less important in 2004 than the other two types of fraud that I discuss in this post? First there is the evidence FOR the other types of fraud, which I discuss below. And also, there are two major studies on the Ohio 2004 election that cast serious doubt on the likelihood that vote switching played a major role in that election. One was conducted by Professor Walter Mebane, whose study played a prominent role in the report of the Democratic National Committee on the Ohio election, and which concluded (on the basis of strong correlations between the Kerry 04 vote and other voting results, such as voting for Eric Fingerhut for Senator) that there was no massive fraud committed in the 2004 Ohio election. I took issue with that conclusion in a letter to Howard Dean, in which my main stated objection to the DNC report was that only vote switching fraud was considered in coming to the conclusion that there was no massive fraud in the 2004 election. And the other evidence against a major role of vote switching in the 2004 Ohio election comes from a study by the Election Science Institute (ESI), which showed that there was no correlation by precinct between exit poll discrepancy from the official vote and discrepancy between the Gore 2000 vote and the Kerry 2004 vote.

I am not saying that these studies prove that vote switching type fraud alone could not have accounted for the Bush victory in Ohio in 2004 (although the authors of these studies essentially DO say that). What I AM saying is that we better give some attention to mechanisms of fraud other than vote switching.

Prevention of electronic vote switching fraud
The means of preventing electronic vote switching fraud are well known to most DUers with interest in election fraud, but for the sake of completeness I’ll list them here anyhow:

The best way of preventing this kind of election fraud is to get rid of DRE machines. Hand counted paper ballots and optical scan machines have both been shown to be much less susceptible to breakdown, as well as much less susceptible to fraud.

But if DRE machines are to be used, every effort must be made to guard against fraud. Three general principles stand out as being crucially important:

1) First, the idea that “proprietary” voting machines which count our votes using secret (i.e., not accessible to the public) software have any place in a democracy must be vigorously fought.

2) Secondly, any DRE machine that counts our votes must be accompanied by a reliable paper trail that will facilitate a recount if needed;

3) and thirdly, laws must be instituted that will ensure that recounts will be available whenever controversy exists over the results of an election, and that those recounts will be conducted in a fair and transparent manner, with the requesting party given the choice of which precincts to select for the initial recount. Recounts were used in the Ohio 2004 election. However, election officials chose what precincts were to be recounted in many precincts (contrary to Ohio election policy), and voting machine company technicians were called in in numerous instances to tamper with the tabulating machines during the recount, thus raising the suspicion that they were fixing the results so that the vote counts would match, and thereby avoiding the requirement for a full hand recount of all but one Ohio county. In one case, election officials were even given a cheat sheet to ensure that the counts matched. And furthermore, even when the vote counts did not match, the required hand recount was not performed, except in one case (See discussion starting on page 36 of this report).

Finally, see Section XII of this report, sponsored by the DNC, on how to prevent election fraud with the use of DRE machines.


Voter registration purging fraud

Description
Obviously, this is a type of fraud whereby voters are illegally erased from the voter roles. To be effective, of course, it must be targeted against one of the two major parties. What a lot of people don’t understand about this is that voter registration is becoming more and more computerized. Also, Diebold was responsible for much of the voter registration in Ohio in 2004.

Evidence of voter registration fraud in the Ohio 2004 Presidential election
Here is an article I wrote which describes a great deal of evidence for the strong likelihood that voter registration fraud played a major role in the 2004 presidential election, and in fact very well could have thrown the election to Bush without the use of any other kind of fraud. Here is my summary from that article of what the evidence shows:

1) A discrepancy of more than a hundred thousand between New York Times (and other newspapers) reports of a massive increase in new voter registration and official Secretary of State figures in Cuyahoga County alone.

2) Partial confirmation of the above from the Greater Cleveland Voter Registration Coalition, which shows a similar (though lesser) discrepancy

3) An explanation for the above discrepancies from the identification of the apparently illegal and targeted purging of 165,000 Cuyahoga County voters.

4) On the ground confirmation of voter purging of an unknown but probably huge number of voters, from Mark Crispin Miller’s new book, “Fooled Again”.

5) Also from professor Miller’s new book, a probable explanation for how Democratic voters were targeted for the voter purging (via the theft of computers containing Democratic voter registration information).

6) From point number 1 above I calculated a net loss to Kerry of about 46,000 votes. But that calculation is based on the discrepancy between official figures and the newspaper reports of 111,000, not the 165,000 purged Cuyahoga County voters identified by Dr. Lovegren. AND, it doesn’t assume the ability to specifically target Democratic voters. With specific targeting of Democratic voters, that number could be much larger. AND, that’s just for Cuyahoga County.

7) The discrepancy between the official figures and the newspaper reports involves much more than Cuyahoga County. And the evidence in Professor Miller’s book also involves counties other than Cuyahoga. When other voter registration fraud from other counties (for which we don’t have specific numbers) is added to that from Cuyahoga County, who can tell how many votes John Kerry lost in Ohio?

8) Just about the only thing missing at this point is for someone from Diebold (who handled much of the voter registration in Ohio, including Cuyahoga Co.) to tell us how this was done.

And since I wrote this post, Bob Fitrakis has obtained additional verification of illegal voter purging in Ohio.

Prevention of voter registration fraud
One important step towards preventing this kind of fraud is to ascertain precisely how it was committed. Then, we must ensure that there is enough oversight of the process so that it isn’t likely to happen again.

Beyond that, it seems to me that the Democratic Party and its allies need to be much more vigilant in ensuring that its registered voters STAY registered. In the 2004 election great amounts of resources were put into getting out the vote on Election Day. Couldn’t we put a similar effort into examining the voter rolls to make sure that our registered voters remain registered by Election Day?


Central tabulator mediated fraud

Description
Each county (as far as I know) has a central tabulator, which adds up all the votes that it receives from each of the precincts in the county. It then reports out the official county-wide vote count, along with the vote count from each of the county’s precincts. These vote counts are referred to as “post-tabulator” or “official” vote counts.

The vote count that each precinct sends in to the central tabulator is referred to as the “pre-tabulator” vote count. Obviously, the pre-tabulator vote count and the post-tabulator vote count for every precinct should be the same. If not, then either central tabulator fraud or an innocent mistake occurred, since there is no legitimate reason why a vote count should change after a precinct sends to the central tabulator its supposedly final count.

Central tabulator fraud may involve vote-switching, but it may NOT involve vote-switching. If the central tabulator simply adds votes in the same proportion as the real votes to a heavily Bush voting precinct, that will help Bush even though his percentage of votes in that precinct will not change. Or, the same effect will occur if votes are subtracted from a heavy Kerry voting precinct. When this is done the fraud escapes detection by the kind of statistical analyses that were performed by Professor Mebane in his DNC report or by the Election Science Institute.

Evidence for central tabulator fraud in the 2004 Ohio Presidential election
Some relatively minor evidence for central tabulator fraud was provided in the 2004 Ohio election when Bush received 4,258 votes from one precinct in Gahanna, which had only 638 registered voters, and when an additional 19,000 votes were reported from Miami County (in exactly the same proportion as the previously reported votes) after 100% of that county’s precincts had already reported, giving Bush an additional net advantage in Miami County of 6,000 votes. I call this “relatively minor evidence” because it could have been accidental. I just don’t know.

Much more suspicious IMO was the infamous Warren County “lockdown”, which allowed election officials to tally the Warren County vote in private. This was rationalized by a bogus “national security emergency”, which election officials used as an excuse to tally the Warren County votes in private. They claimed that they learned of this “national security emergency” from the FBI – a claim that was soon denied by the FBI. Yet the Warren County results continue to stand, and without any serious investigation. It also may be significant that this event occurred towards the end of the evening, when it still looked very much as if Kerry would win Ohio, and by the time the Warren County votes had been “counted”, victory had all but slipped away from the Kerry/Edwards ticket.

After much studying of the vote in Cleveland I came to suspect that many thousands of votes were deleted from that heavily Democratic city, as discussed in this thread (See first section of this post). My initial suspicions were aroused because of an anomalous relationship between voter turnout in Cleveland and the number of machines per voter, as discussed in Section IV, page 3, of the DNC report on the Ohio election, and because of very low turnout in many of Cleveland’s precincts, as reported by Richard Hayes Phillips. My suspicion was further aroused when I realized that the very low voter turnout in Cleveland was reported despite the fact that voting lines were quite long throughout much of the city, as described in this thread (See section on “Why so many long voting lines but such a low turnout in Cleveland?”). And the observation by a Green Party observer to the Ohio recount of several anomalies didn’t serve to allay my suspicions.

Some people would say to me, in response to my voicing of my suspicions of central tabulator fraud in Ohio, that that kind of fraud was unlikely because it could be so easily proven by simply comparing the pre-tabulator to the post-tabulator vote count, to see if they matched. But when I tried to ascertain pre-tabulator vote counts for Cleveland I couldn’t find anyone who knew what they were. I contacted Michael Vu, the Director of the Cuyahoga County Board of Elections, to request those counts from him, and he promised to look for them, but he repeatedly failed to get back with me on this. And I also spoke with Ellen Theisen of Voters Unite! about this, and she told me that persons interested in investigating the 2004 election were having a hell of a time trying to get pre-tabulator vote counts from anywhere in the country.

So finally, after discussing this issue with fellow DUer adagiopop, he undertook an effort to obtain pre-calculator vote counts from Cuyahoga County. His initial efforts at this task identified several probable anomalies, as described in this thread, and that investigation is still continuing.

Prevention of central tabulator fraud
It seems so simple. All we need to do is have one volunteer in every precinct in the country (in states where elections are expected to be competitive), to obtain the pre-calculator vote count at the time that the polls close. Then, if the results of an election seem suspicious, all we need to do is compare the pre-calculator counts to the post-calculator counts, and if we identify large mis-matches, then we can feel confident that that’s where the problem is. Then, a full recount of those precincts where mis-matches are identified should be demanded.


Summary

It seems that many thousands of people are working very hard to prevent vote-switching fraud in future elections. That’s wonderful. But please let’s also try to prevent voter registration fraud and central tabulator fraud, which appear to have played a bigger role than vote switching fraud in determining the results of the 2004 Presidential election.