Our Position on Fraud in the 2004 Presidential Election

In the aftermath of the Nov. 2 election, many people have promoted theories and statistical evidence of major fraud or errors. Our position on this is being distorted and misinterpreted, so we would like to take this opportunity to clarify it.

We do not believe that there is any more reason to look for problems in this election than in previous elections, but auditing of all elections should be routine. Citizens should be able to do some of these audits.

The way to encourage trust in a system is to have independent checks on a system. The best way to encourage trust is to make that system so open that anyone can check it.

We advocate the publication of detailed election statistics on the Internet, and the analysis of those statistics by members of the public. In the long run, this will result in a much healthier election system.

The current high level of scrutiny of the election is a very healthy thing.

We'll emerge from this process with more of our questions answered. It will greatly increase public understanding of elections, and highlight the problems we have -- problems which could very well have resulted in an irresolvable tie, had voters in just a few key states voted a little differently.

Unproven charges of fraud are unwise and damaging.

Unproven charges of fraud damage the country. They undermine the legitimacy of elected officials, upon which our government is based, and undermine confidence of the voters in the election system. In our opinion, anyone who claims election fraud has a moral obligation to present solid proof. If you think you see that proof, and it's not a signed confession by someone who did the deed, you need at the very least to check it out very carefully before making a public charge of fraud. For example, if you see something strange about the election statistics an a particular county, at least talk to the local election officials in that county, and talk to political scientists who are experts in election behavior to see if your analysis makes sense and whether there are other explanations.

If you note something that seems odd, it is perfectly reasonable to post the analysis with a comment that it seems odd, and watch the discussion begin. You will be doing a public service. If nothing else, you'll learn more about how elections work, and it may turn out that you've discovered something important.

Even if you don't care about acting in the best interests of the country, it is unwise to make unfounded charges because it will erode your credibility very rapidly. Remember the parable about the boy who cried "wolf".

Neither VerifiedVoting.org nor the Verified Voting Foundation favor any particular candidate.

We are not motivated by preferring any particular candidate. We value the support for publicly verifiable elections that we have seen in both major parties. Our position is that the person holding office should be the person who won the election. We hope that that is not controversial!

We are not saying the the election was fraud-free, or free of major error.

It's not clear that we could ever say that an election is fraud-free -- fraud would be well-hidden, after all.

In particular, data from this election is still being digested, and we don't even have a clear idea of what happened. Our minds are open.

So far, we have not seen convincing evidence of either fraud nor of a major error in the Presidential election.

As of this writing (11/14/04), we have seen a lot of supposed evidence of fraud or errors. In some cases, these stories can be refuted by simple fact checking. In others, experts can point to other probable explanations, In others, we haven't seen enough of the debate to know what to think. There are probably others that haven't come out yet. ss

It is important to distinguish between the statements "We have not yet seen convincing evidence of X " and " We believe X is not true."

If you can't do this, we probably aren't going to be able to have a rational discussion.

We like manual recounts.

Manual recounts will reveal problems with the election, in which case we will need to deal with them as a nation. More likely, they will confirm the election results where they are conducted. We'll have a much clearer idea of whether the election results are accurate, which is good. In some cases, a manual recount may give us other useful information, such as the types of errors that voters made, which may help improve elections in the future.

Recounts should be regarded as audits, not criminal investigations or efforts to overturn an election. Like financial audits, they increase confidence in the system.

It is unlikely that there will be an election problem large enough to overturn the Presidential election.

According to the numbers posted on Nov 3, President Bush won by about 3% (about 3.5 Million votes) in the popular vote, 2% (about 135,000 votes) in Ohio, and by 5% (almost 400,000 votes) in Florida. Some very large discrepancies would need to be found to reverse the election.

So given what we know at this time it seems unlikely to happen. If, tomorrow, someone finds 100,000 Kerry votes were miscounted as Bush votes in Ohio, we'll revise this opinion very quickly. But that hasn't happened. --