Croker Sack

"Democracy is the theory that the common people know what they want, and deserve to get it good and hard." — Henry Louis Mencken (1880-1956)

Monday, May 30, 2005

How does it look from the voters' point of view?

If the petitioners lose in Chelan County Superior Court, as seems probable, a successful appeal to the state supreme court would require more persuasive arguments than they have so far presented.

Under the rulings already stated by Judge Bridges in that court, Washington has no law which provides a remedy in the pending gubernatorial election contest.

The neglect, misconduct, and irregularities indicated by the evidence prove beyond cavil that no one was duly elected – but the legislature decided otherwise without determining whether the official returns were worthy of belief, and now the legislature’s choice of Gregoire cannot be proven to be the wrong choice.

The petitioners’ trial strategy was built around a belief that they need not prove that irregularities affected the election’s outcome, but – having never briefed the court on that argument – they were saddled at trial with the judge’s belief (stated on May 2) that they must prove that the outcome was affected.

They would have benefited from a ruling that the standard of proof is a preponderance of the evidence, rather than the more difficult standard of clear and convincing evidence. Having never briefed the court about which standard applies, they brought up the issue in oral argument on May 2 – and got an unfavorable ruling which required proof by clear and convincing evidence. (Perhaps the judge would limit that ruling to the proof needed to show that a particular vote was an illegal vote and would apply the less difficult standard to the ultimate question whether to set aside the election, but he has not hinted that it is so limited.)

Required to prove the outcome was affected, and to prove it by clear and convincing evidence, the petitioners hoped to prove it by using the proportional deduction remedy in a novel way – to prove how the illegal ballots were actually voted. The judge noted his concerns with the proffered evidence, read the law to the petitioners’ lawyers, and reserved ruling on the admissibility of the evidence. Unless the judge later admits the expert testimony into evidence, the petitioners will have no evidence tending to show that illegal votes affected the election’s outcome.

A Washington Supreme Court opinion issued in 1912 will likely play a major role in the outcome of the trial, therefore its meaning and applicability to the case at hand will be a crucial issue on appeal.

That case was Hill v. Howell, 70 Wash. 603, 127 Pac. 211 (1912).

Before getting into the excerpts from that opinion which are quoted below, consider the statement attributed to Justice Susan Owens of the Washington Supreme Court during consideration of an issue in the November 2004 election. According to the December 22, 2004, online edition of The Seattle Times, she stated:

At a hearing this morning before the high court, Republicans argued that a recount should be a mere retabulation, and that it was too late for counties to go back and correct errors.

Harry Korrell, a lawyer for the Republican Party, said counting those votes would cause rreparable harm, but justices questioned who would be hurt.

"You're looking at it from the point of view of the winner or the loser -- shouldn't we be looking at it from the point of view of the voter?" asked Justice Susan Owens.

Ruling within a few hours of hearing the case, the court unanimously said state law and previous court rulings specifically allow county canvassing boards to correct mistakes during a recount.

Keep in mind the idea of “looking at it from the point of view of the voter” when reading the Hill opinion, and note its focus on the candidates.

Hill dealt with a statute which was worded almost the same as RCW 29A.68.070, which establishes what needs to be proven to prevail in an election contest brought under RCW 29A.68.020(1), i.e., based on misconduct of the members of a precinct election board.

Judge Bridges stated on February 4 (page 29 of the transcript of his oral rulings): “With respect to misconduct, whether that misconduct falls in 020 or 011, I think the standard is 29A.68.070, and so I want to say that so you folks have some sense of what I think the ultimate standard of proof is and what the petitioners have to show.”

The judge has, then, decided to apply the statute to misconduct (and presumably neglect too) by “any election officer,” and not just to members of precinct election boards.

This excerpt (with emphasis added) from pages 608 – 609 of the Hill opinion shows that the statute being applied was essentially the same as RCW 29A.68.070, and shows that courts long ago construed the statute as authorizing a remedy only when there is proof that the outcome was affected (in the absence of an allegation and proof of fraud):

On the merits of the controversy, we are of the opinion that the petitioner has not shown sufficient cause to warrant us in setting aside the returns of the canvassing boards. With reference to general elections, it is provided by statute, Rem. & Bal. Code, SS 4942, that no irregularity or improper conduct in the proceeding of the board of judges, or any of them, shall be construed to amount to such malconduct as to annul or set aside any election, unless the irregularity or improper [pg. 609] conduct shall have been such as to procure the person whose right to the office may be contested to be declared duly elected when he had not received the highest number of legal votes.

Applying this provision of the statute to a case where the notice of an election published by the clerk of a school district notified the electors of the district that the polls would be open until 7 p.m. of the day of election instead of 8 p.m., as the statute required, this court said:

"Another reason for denying to the respondent any benefit of this mistake is, that his information contains no allegation that, had the polls been kept open an hour longer he would have been in any wise benefited by it. In all such cases there must appear some substantial reason why courts should interfere to overthrow an election, in the absence of any allegation of fraud, to the effect that, had there been a larger number of votes cast, the result would have been different." State ex rel. Bailey v. Smith, 4 Wash. 661, 30 Pac. 1064.

In Williams v. Shoudy, 12 Wash. 362, 41 Pac. 169, we held that a resolution calling for an election to be held between the hours of 9 a.m. and 7 p.m. did not avoid an election so held although the notice of election published specified that the election would be held between the hours of 9 a.m. and 6 p.m. In Seymour v. Tacoma, 6 Wash. 427, 33 Pac. 1059, we held that the posting of the notices of an election for twenty-six days preceding an election, whereas the law authorizing the election provided that such notices should be posted for thirty days did not vitiate the election in the absence of a showing that such irregularity in some way affected the result. Richards v. Klickitat County, 15 Wash. 509, 45 Pac. 647; State ex rel. Mullen v. Doherty, 16 Wash. 382, 47 Pac. 958, 58 Am. St. 39; Hesseltine v. Wilbur,29 Wash. 407, 69 Pac. 1094.


This excerpt (with emphasis added) from page 610 indicates the significance of the judge’s questions and Nicole Way’s answers regarding her knowledge and purpose when preparing the false Mail Ballot Report. Note also how the court focused on the absence of any indication that the candidates played any part in causing the irregularities.

It will be remembered that no charges of fraudulent conduct or willful misbehavior are made or proven against any of the officers of the election who were guilty of the irregular conduct found by the commissioner. It is conceded that they acted in the utmost good faith, and that their conduct was the result of ignorance on their part of the requirements of the election statute and not from any corrupt motive. It must be remembered, also, that the candidate Steiner was in no way responsible for the acts of these officers. On the contrary, he was as blameless in the matter as was candidate Hill; and since it is evident that many legal votes were cast for him in this precinct which will be denied him if these votes are rejected - sufficient to change the result in the particular case - common justice requires that these votes be rejected and be not counted only in case the exigencies of the matter admit of no other alternative.


The next paragraph on pages 610 – 611 contains the statement which the Democrats’ lawyers have argued in court more than once – that illegal votes don’t matter unless they can be shown to have been cast for one candidate or the other.

Turning to the immediate question, it seems to us that there is no reason to reject the returns made from Beverly precinct in their entirety. It is clear that there was but one vote that can be said to be irregular that actually got into the ballot box, namely, the vote of elector Tapley. But neither the findings of fact, nor the affidavit made by Tapley in support of the application for the writ of mandamus, although the affidavit sets forth the facts fully, show for whom the elector voted. If this was an illegal vote it was proper to show for whom the elector voted, and since the fact [pg. 611] is not shown, it must be treated between the parties as a legitimate vote. Neither of the candidates were responsible for the manner in which the vote got into the ballot box, and both being innocent of wrongdoing, it would be an injustice to charge the error to either of them.

Looking at the situation from the point of view of the candidates – and focusing on who won or lost – it may seem reasonable to treat illegal votes as irrelevant; but how does it look from the point of view of the voters?

As an aside for the benefit of all those who think voter crediting is a post-election exercise which has nothing to do with the authenticity of the election results, note that the importance of comparing the number of voters and ballots to identify the presence of illegitimate ballots is older than Hill. At page 611, the court stated (emphasis added):

But counsel argue that the vote of the precinct ought to be rejected because of the fact that the polls were closed for the noon hour, without any guard over the ballot boxes. But it is not shown that harm resulted from this fact. The door of the building in which the ballot boxes were left was locked during the interim, and there is absolutely no proof that the ballot boxes were in any manner disturbed. On the contrary, proofs or inferences are all the other way. By the statute, poll lists are required to be kept by the election officers, and the returns of the election should show a name of an elector for each ballot in the ballot box. There is no claim that there was any discrepancy in the returns in this respect, and this fact precludes the idea that the ballot boxes were tampered with during the absence of the election officers.

Although the petitioners’ lawyers seem not to know it (and have never to my knowledge cited any law to this effect in court, during a deposition, or in a brief), reconciling the numbers of absentee voters and absentee ballots is (and was) required by law – as is the reconciling of the numbers of polling place voters and ballots. Hiding King County's discrepancies from the canvassing board (and the public) and presenting false ballot reports to the board sabotaged the process established for the verification of the election returns and broke the law.

And, finally, the Hill opinion quoted a legal encyclopedia at pages 612 – 613. Judge Bridges mentioned this part of the Hill opinion on May 2, noting that it may allow for setting aside an election based on neglect and disregard for the laws to such an extent that the official returns aren’t worthy of belief.

"Where an election appears to have been fairly and honestly conducted, it will not be invalidated by mere irregularities which are not shown to have affected the result, for in the absence of fraud the courts are disposed to give effect to elections when possible. And it has even been held that gross irregularities not amounting to fraud do not vitiate an election. Where the legislature declares a certain irregularity in election procedure to be fatal to the validity of the returns, the court will effectuate that command. And while the conduct of election officers may, although actual fraud be not apparent, amount to such gross negligence and such a disregard of their official duties as to render their return unintelligible or unworthy of credence. But the power to throw out an entire division is one which ought to be exercised with the greatest care and only under circumstances which demonstrate beyond all reasonable doubt that the disregard of the law has been fundamental or so persistent and continuous that it is impossible to distinguish what votes are lawful and [page 613] what are unlawful, or to arrive at any certain result whatever, or where the great body of the voters have been prevented by violence, intimidation, and threats from exercising their franchise." 15 Cyc. 372. (Emphasis added.)

From the point of view of the voters who desire that the person installed in the governor’s office be duly elected – rather than caring only whether their preferred candidate was put in office – this last excerpt is the only one that seems to offer relief.

King County election officers’ disregard for the law and neglect in the performance of their legal duties hid material discrepancies discovered during the canvassing process from the canvassing board – leading to the board’s issuance of an official canvass report which was then cited by the legislature as a sufficient basis to declare that Gregoire was duly elected and to issue the certificate of election.

The presence of more than 900 illegal ballots in the polling place ballot boxes was hidden from the canvassing board through the silence of Bill Huennekens, the superintendent of elections for King County, and by the presentation to the board of a false Provisional Ballot Summary Report. An accurate report would have revealed that more than 900 provisional ballots had been issued but could not be accounted for. The board almost certainly would have asked whether they had been inserted into ballot boxes via the polling place Accuvote machines – and thus would have discovered that Huennekens and his subordinates knew already that this is precisely what happened.

The presence of more than 800 absentee ballots in the vote count for which there were no voters known to have cast them was hidden from the canvassing board by silence and the presentation of a false Mail Ballot Report.

The canvassing board was kept in the dark, and the legislature (despite the discrepancies revealed in the 10 days before the certificate of election was issued) chose to rely on the board’s official canvass report.

It’s all just water under the bridge, if looked at from the point of view of those who simply care who was declared the winner.

Is that the only point of view that matters under Washington’s statutes and case law?

To prevail, the petitioners need to raise all relevant issues at trial, so they can argue them in the supreme court.

For the sake of those of us who give a damn about the way a person was put in the governor’s office, the GOP’s lawyers need to do a better job than they have done so far.

Thursday, May 26, 2005

Proportional Deduction: Evidence or Remedy?

Is proportional deduction (or reduction) a recognized method to be used in determining how illegal ballots were voted?

Or, is it a remedy to be used as an alternative to setting aside a flawed election?

I would bet it's a remedy, not a form of evidence which has been used (or can be used) to arrive at a finding as to how secret ballots were voted.

Judge Bridges noted today in his comments from the bench that the test for admissibility of expert testimony involves two aspects. Not only must the method be generally accepted, but it must also be generally accepted as it is proposed to be applied in the case at hand.

If the petitioners had referred the judge to any case law showing that proportional deduction has previously been used to prove how illegal ballots were voted, then the judge probably wouldn't have made those comments.

Note that choosing an appropriate remedy comes after finding that a remedy is needed to correct the effects of neglect, error, misconduct, or irregularities in the administration of an election.

Maybe one of the petitioners' basic problems is that they are trying to use proportional deduction to prove that a remedy is needed, rather than arguing that it is an appropriate remedy.

If they could prove that a remedy is needed, the supreme court might decide that using proportional deduction is better than setting aside the election.

Absent a supreme court ruling, there seems to be nothing in Washington law which provides for a remedy short of setting aside the election when illegal votes have created a problem that cannot otherwise be resolved.

Tuesday, May 24, 2005

For curiosity's sake

Nicole Way, the supervisor of mail ballot operations for King County (who is now on administrative leave after the discovery of a few significant errors in that operation), may testify tomorrow.

If only for curiosity’s sake, the petitioners’ lawyers should ask her a few simple questions:

  • Did you obtain a report from your computerized election management and voter registration system, called “DIMS,” that stated how many absentee ballots had been accepted as valid?
  • [If yes]: Was the number of valid ballots stated by DIMS greater or less than the number of absentee ballots tabulated in your computerized “GEMS” vote counting system?
  • [If no]: Why not?
  • Did the “GEMS” system have any ability to tell you whether the ballots sent through it had been accepted as valid during the ballot verification process?
  • Did your GEMS system have the ability to tell you whether any absentee ballots sent through it had been rejected as invalid during the ballot verification process?
  • Did your GEMS system have any ability to tell you whether any absentee ballots were accidentally sent through for vote tabulation more than once?

As noted earlier, Way’s deposition showed that the Mail Ballot Report was not just falsified – it was constructed in a way that didn’t indicate whether the ballots sent through the vote tabulation system were all valid, nor did it indicate whether any ballots had been sent through more than once.

The report should have shown the number accepted as valid – taking that information from DIMS and whatever other records Way found it necessary to maintain to have that data available.

Also, the report needed to state on a separate line the number of ballots that had been sent through GEMS to tabulate the votes on them.

Then, the reconciliation required a comparison of the number accepted as valid and the number sent through the GEMS vote tabulation system.

Only that comparison of information from two different sets of data would constitute a reconciliation.

What Way did was prepare a report which purported to show a reconciliation, but which actually reported nothing more than the number of times the GEMS system registered the act of scanning a ballot to tabulate the votes.

Sure, she also said they hand counted the stacks of rejected ballots; but there was no use of DIMS or any other data base to indicate what number of rejected ballots should have been in those stacks.

If the GOP lawyers wish to show that more ballots were sent through the vote counting system than had been accepted as valid, they ought to ask the right questions while Way is on the stand.

Update May 25: The archived audio recording of Way's testimony isn't yet available at TVW's web site, so I'll have to listen later to see if I missed it -- but it appears that the petitioners' lawyers weren't curious about the number of valid ballots recorded in the election management and voter registration system ("DIMS"). Odd: They didn't ask Huennekens about the Provisional Ballot Summary Report, either. The GOP needs lawyers who possess a little bit of curiosity.

Monday, May 23, 2005

Maguire, Foreman, and the rest of you: This is your wake-up call

The GOP’s lawyers seem to have missed the significance of the fact that the Provisional Ballot Summary Report presented to the King County canvassing board on November 17 was false.

Neither the petitioners’ trial brief nor the opening statement by Mr. Dale Foreman mentioned the false provisional ballot report.

The brief discusses provisional ballots at pages 12 – 13, and makes a good point about the silence of the King County superintendent of elections, Bill Huennekens. He knew that the canvassing crew had determined that 900+ more ballots were in the polling place ballot boxes than the number of voters who had been credited with voting regular polling place ballots, but he didn’t tell the canvassing board about this material discrepancy in the ballot reconciliation data.

The significance of the false Provisional Ballot Summary Report is similar to that of the false Mail Ballot Report. Knowing that there were 900+ too many ballots in the vote count from the polling places, and knowing that the canvassing crew determined with reasonable certainty that most of those extra ballots were unlawfully cast provisional ballots, Huennekens surely realized that revealing the absence of 900+ of the issued provisional ballots would probably cause someone to ask if they had been inserted into the Accuvote machines at polling places. So, a false provisional ballot report was presented to the canvassing board to show a perfect reconciliation of provisional ballots – thereby concealing from the board the information which would probably lead them to discover what else was being concealed from them, namely the presence of 900+ too many ballots in the polling place vote count.

See Logan’s deposition at pages 153 – 154, where he testified that he was told about the false provisional ballot report after the recounts were completed and certified. Rather than report the actual number of ballots issued, the “ballots issued” figure on the report was merely the sum of the numbers in the other parts of the report.

One part of the evidence makes this concealment through silence and the submission of a false report even more significant than the false Mail Ballot Report. There was a chance to identify and remove those 900+ unlawfully cast provisional ballots from the ballot boxes and subtract their votes from the count. For the 875 absentee ballots, there was no way to identify the ballots themselves and remove them from the count.

As Logan testified during his deposition at pages 18 – 19, the provisional ballots were supposed to be folded “in thirds” by the poll book judges before being issued to voters – so the ballots would fit into the security envelopes.

Logan testified at pages 135 – 136 of his deposition that the unlawfully cast provisional ballots could probably be identified by the creases resulting from folding them to fit the envelopes.

Had the canvassing board been told that 900+ ballots of questionable validity were in the polling place vote count and that the 900+ folded, creased provisional ballots could probably be identified and removed from the vote count, does anyone doubt that the board would have to direct that Logan’s people make the effort to identify and remove those ballots?

If the ballots had been removed prior to November 17, we wouldn’t be wondering how the votes on them were cast. We would know exactly how they were voted.

And, we wouldn’t be wondering which votes to subtract, since all of them were unlawfully cast, and all of them ought to be removed. None of them belonged in the vote count unless it could be shown that any individual ballots among those 900+ ballots were cast by eligible voters who hadn’t cast another ballot.

The GOP’s lawyers need a wake-up call.

Torpedo in the water!

Earlier, I noted that it appeared the Democrats may rebut "proportional reduction" with the testimony of felons who voted.

The WSDCC lawyer said in his opening statement that the Democrats will present the depositions of 5 such voters. Apparently, those 5 voted in counties carried by Rossi, but in precincts carried by Gregoire -- thus their votes would be attributed to Gregoire under the statistical inferences made by the GOP's expert witnesses.

Of course, those 5 testified (according to Hamilton's opening statement) that they voted for Rossi and Bennett -- 4 for Rossi.

How many such specific examples would it take to persuade the court that "proportional reduction" is an inappropriate remedy? Is 5 enough?

Sunday, May 22, 2005

Peanuts, hot peanuts here!

For those of us in the cheap seats, here's something to ponder as the election contest unfolds.

The idea that the petitioners – to prevail under RCW 29A.68.070 or RCW 29A.68.110 – must show that Rossi would have won if not for the effect of irregularities and illegal votes is contradicted by RCW 29A.68.050.

The petitioners’ trial brief quotes one phrase from RCW 29A.68.050 on page 31 (page 39 of the 3.39 MB “pdf” file posted at Sound Politics), but misidentifies the statute as RCW 29A.68.060 – and says nothing about the apparent contradiction between RCW 29A.68.050 and the judge’s inclination to require proof that the election’s outcome was actually changed by the irregularities or illegal votes.

The judge stated on May 2: “While the petitioners’ arguments in this regard may be persuasive, Washington’s election contest statutes clearly require the contestant to show illegal votes or misconduct changed the election result based on RCW 29A.68.110 and .070.”

Here are the pertinent, final two sentences of RCW 29A.68.050, with the phrase quoted in the petitioners’ trial brief in bold type:

“After hearing the proofs and allegations of the parties, the court shall pronounce judgment in the premises, either confirming or annulling and setting aside such election, according to the law and right of the case.

“If in any such case it shall appear that another person than the one returned has the highest number of legal votes, said court shall declare such person duly elected.”


If, to prevail, the petitioners were required to prove that Rossi actually won the election, the first of those two sentences would have no meaning.

The first sentence requires the court to either confirm or annul the election.

The second sentence requires the court to declare Rossi the winner, “if” it appears from the evidence that he actually won.

The two issues are treated separately, because they are separate issues.

  • An election can be shown to be fatally flawed, so that the actual winner cannot be determined. In such a case the election would be annulled and set aside.
  • If, on the other hand, a winner can be determined, the election would be affirmed – even though the court rules that the person who had previously appeared to have fewer votes actually was elected.

In Foulkes v. Hays, the election was annulled, but no winner was declared – precisely as this statute indicates the court may do.

If the court were to “pronounce judgment...annulling and setting aside such election,” how could it then declare one of the candidates to be duly elected? Note that the election itself would be annulled and set aside, not the official vote count showing an apparent winner. The election would be annulled, yet someone won? That would be nonsense. Clearly, no one is the duly elected winner of an annulled election.

The court could “confirm” the election and yet find that “another person than the one returned has the highest number of legal votes,” and thus the court could “declare such person duly elected.”

The petitioners seek to have the election annulled and set aside, not to have the court declare Rossi to be duly elected. They have argued that such a remedy doesn't require proof that irregularities or illegal votes actually changed the outcome.

The statute authorizes precisely what they ask the court to do, but the judge doesn't seem to see it that way.

It’s too bad the petitioners don’t appear to notice the contradiction between the judge's stated belief and the words of RCW 29A.68.050. It would be interesting to hear the judge try to stay with his stated belief and still give meaning to the first sentence quoted above.

Programs -- get your programs!

Regarding the burden of proof, keep in mind the different issues as the trial of the Washington gubernatorial election contest proceeds.

Perhaps the judge will revisit his May 2 ruling on the petitioners’ burden of proof on Monday, since he invited counsel to address it, and the petitioners have.

“Burden of proof” and “standard of proof” are often used to refer to the same thing, but they’re actually not the same.

Think of “burden of proof” as a thing someone has to carry. Whoever has the burden had better carry it, or he loses.

In this case, the petitioners have the burden of proof on the ultimate issue – whether to annul the election.

In a sense, the intervenor-respondents (the Democrats) have the burden of proof when offering evidence in support of their own points – e.g., the existence of offsetting errors or illegal votes. If the Democrats’ evidence is not worthy of belief, then their points won’t carry any weight. But remember that the Democrats don’t need to prove that the election was valid. Their purpose is to show that the petitioners haven’t carried their burden of proof.

When you get to the bottom line, the GOP loses if they haven’t carried their burden of proof as to the ultimate issue, no matter what the Democrats did.

Think of “standard of proof” as you would any standard that must be met. The person who has to carry the burden has to carry it to a certain height – preponderance, clear and convincing, and beyond a reasonable doubt are the usual standards.

Also, note that the “elements” which must be proven are often laid out ahead of time. On May 2, Judge Bridges ruled on the elements which the GOP must prove to show that an illegal vote was cast. He decided that there are six elements. At page 7 of the transcript, he stated: “One, that the...voter was convicted as an adult, that the voter was convicted of a felony, that the voter had not been given a deferred sentence, that the voter had not...had their civil rights restored. Fifth, that the voter cast a ballot in the 2004 general election and finally, number six, that the voter marked the ballot to indicate a vote for a gubernatorial candidate.”

Standard of proof (preponderance vs. clear and convincing) is discussed at pages 35 – 36 of the petitioners’ trial brief.

Here is what they haven’t said:

Judge Bridges on May 2 appears to have equated (1) cancelling a person’s right to vote and (2) determining whether votes have been shown to have been cast by disqualified persons in a contested election – but they aren’t the same.

The judge’s ruling is on pages 6 – 8 of the transcript. At page 8 he stated: “The same standard [“clear and convincing evidence,” as is required to challenge successfully a voter’s registration] should apply when election results are contested under 29A.68.020. Inasmuch as voting is a constitutional right, no vote should be held illegal and discounted absent clear proof that the voter was legally disenfranchised.”

The May 2 ruling clearly sets the standard of proof required to prove that an illegal vote was cast by a felon.

If the judge also intends to apply the clear and convincing evidence standard of proof to the ultimate issue, i.e., whether to annul the election based on illegal votes, then the distinction between the two issues is even greater.

Denying a person the right to vote in any future election until a disqualification is removed ought to require clear and convincing evidence to prove the disqualifying fact, since voting is a fundamental right of citizens who aren’t disqualified from voting.

Deciding that an election is fatally flawed because of illegal votes doesn’t bar anyone from voting, nor does it bar future elections. Instead, it ensures that our elections are decided by legitimate votes.

Requiring clear and convincing evidence to justify annulling a flawed election makes it more difficult to protect the right of qualified electors to decide the outcome of elections.

Requiring preponderance of the evidence better protects the right of qualified electors and denies no qualified elector the right to vote – thus it is the standard of proof which ought to be applied in deciding a contested election.

Using preponderance of the evidence as the standard of proof doesn’t cause an unjustified or unreasonable disruption of the elections process. Compared to the effect of allowing the result of an apparently flawed election to stand, the possibility that more elections would be successfully contested when the lower standard of proof is required is not too high a price to pay.

Until the legislature enacts laws which establish a higher standard of proof for election contests, the ordinary standard of proof for civil actions ought to be applied – namely, preponderance of the evidence.

The state supreme court affirmed the trial court’s ruling in Foulkes v. Hays, knowing that the trial judge decided the case based on a preponderance of the evidence rather than the requested higher standard of clear and convincing evidence. The parties actually argued the issue, and the court chose preponderance of the evidence as the appropriate standard of proof.

Saturday, May 21, 2005

Dean Logan sticks to his story

In today’s edition of The Seattle Times, there is a repeat of the amazing claim made by King County spin-woman Bobbie Egan, and this time it reportedly comes straight from the horse’s mouth.

Egan had claimed on Thursday morning that the false mail ballot report prepared by Nicole Way and Garth Fell wasn’t actually a falsification.

Here’s the statement attributed to Logan in Keith Ervin’s article in The Seattle Times:

Logan said the Mail Ballot Report — which incorrectly indicated that all absentee ballots were accounted for — shows his office needs to improve its procedures for tracking and reporting ballot numbers.

"What I don't think that indicates, nor occurred, was a falsification of the report or a collaborative collusion on the part of employees here," Logan said. "... It is not the same thing as an employee trying to falsify reports or fix an election, or fraud."

When employees were asked about the report's misleading number, he said, they were forthright and made "no effort to conceal that."

Logan, who became elections director in September 2003, said he has concluded that employees have consistently misreported the number of returned ballots since at least 2000.

Of course, as the article points out, Logan’s two predecessors disagree with his assertion that the report has been routinely falsified since at least the year 2000.

Never mind them.

And pay no attention to Nicole Way, who testified that she and Garth Fell chose to prepare the false report in November 2004 because of shortcomings in their new “DIMS” election management and voter registration computer system. Way testified that she had been concerned about her inability to get accurate reports on the number of absentee ballots returned by voters even before the general election. The conversion to DIMS occurred in the first half of 2004.

Logan has picked his story, and he’s sticking with it.

The only apparent change since Logan’s deposition on April 18 and 19 is the omission from the article of his equally amazing assertion that he didn’t consider the false provisional ballot report and false mail ballot report to be “errors in the conduct of the election.” Perhaps The Times didn’t have space to include Logan’s entire story.

Appended below are excerpts of Logan’s deposition, showing how he claimed even under oath that the mail ballot report was always done that way – that is, it was always purported to be the ballot reconciliation required by law, but really was nothing more than a statement of the numbers of ballots sent through the vote tabulating system and of ballots which weren’t sent through.

Note especially how Logan describes the hurried efforts to get ballots into the vote count before the deadline. If, as he claims, his subordinates didn’t always remove those “challenge codes” in the voters’ DIMS files, then the figures for ballots accepted as valid and ballots rejected would have shown too many ballots had gone through the GEMS vote tabulation system.

Since, despite their best efforts, Logan's gang could not reconcile their numbers after the county's election results were certified based on the false ballot reconciliation reports, it seems clear (and even convincing) that they really did count the votes on hundreds of illegitimate ballots during their frenzied efforts to get as many ballots into the count as possible. They knew that their party's candidate would probably benefit from most ballots they could put into and keep in the vote count -- whether the ballots were legitimate or not.

Way, Fell and Huennekens, if not Logan as well, needed to falsify that mail ballot report to avoid revealing that the best available data showed they had counted the votes on hundreds of illegitimate ballots.

It was the discovery of 93 misplaced absentee ballots that disclosed the false nature of their mail ballot report’s supposedly perfect ballot reconciliation.

It was the presence of hundreds more absentee ballots than voters that prompted them to lie in the first place.

^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^
Excerpts of Logan’s deposition:
[Page 83] Q. Do you know how many absentee ballots were returned?

A. I don’t have – I don’t have that number. I can’t recall that number off my head, no.

Q. Did you ever know that number?

A. That number was reported to myself and the canvassing board through a mail ballot report that is produced for each election. And we obviously know the number of ballots that were counted by the – absentee ballots that were counted in the election, the number of absentee ballots that were rejected in the election, and the number of ballots that were issued in the election. So I certainly have examined each of [Page 84] those figures.

Q. The mail ballot report indicates the number of absentee ballots returned?

A. Yes, it does.

Q. Okay. And where does that number come from?

A. My understanding is that that – the number on that mail ballot report was derived by taking the number of ballots that were counted, which is reported out of the vote tabulation system, and the number of ballots rejected – number of absentee ballots rejected, which is produced from reports out of the election management and voter registration system and confirmed through manual counts of rejected absentee ballots. I’m sorry. I’m trying to picture the report. Without it in front of me, it’s difficult to –
###
Q. (By Mr. Maguire) Mr. Logan, I’ve just handed you a document marked by the court reporter as Exhibit No. 1. On the front, it says, King County Canvassing Board General and Special Elections, November 2nd, 2004 Meeting Agenda. Is that [Page 85] correct?

A. That’s correct.

Q. Is the second page which has a Bates No. B/KING 004870, the Mail Ballot Report?

A. Yes, it is.

Q. Does this help refresh your recollection as to how the number of absentee ballots are counted?

A. Yes. My understanding is that this first gives basis of information that the number identified in No. 1, Total number of ballots issued, comes from a report from the elections management and voter registration system. The total number accepted as valid and counted, Item No. 3,
comes from the vote tabulation system. No. 4, Total number of ballots rejected, comes from a report from the voter registration election management system as well as a manual count of the ballots that were rejected. And that Item No. 2 is derived by adding Items No. 3 and 4. That’s my understanding post election of how this report was prepared. It’s also my understanding of how this report, which is a common report in all elections, has been prepared in the past.
###
[Page 86] Q. (By Mr. Maguire) And nobody that you know of counted the number of absentee ballots that were physically returned by voters?

A. My understanding is that that is tracked through the batching process, that all return ballots are associated with a batch, and that those batch records, individual batch records, indicate the number of ballots in each of those batches. I’m not aware of a complete roll up report of those batches. And as I indicated before, some ballots will be moved from one batch to another batch based on processing. So simply adding up the total number of ballots associated with each of those batch slips would not necessarily get you to the total number of ballots returned, because some of those ballots had been moved from one batch to another.
###
[Page 99] Q. Prior to signing the certification, did you know of any discrepancies or inconsistencies in the returns?

A. No, there were not specific discrepancies or deficiencies in returns reported to us.
###
[Page 111] Q. ...The number of absentee ballots that are determined valid would be known at the time that the absentee ballots are verified; isn’t that right?

A. Are you referring to the number that’s on the mail ballot report?

Q. Not necessarily. As I understand the process, when some election worker is verifying the signature on an absentee ballot comparing it to the signature that’s on the voter registration, that at that time if the signature – if the ballot is validated and verified, it goes into the pool of ballots that can be counted; is that right?

A. That’s correct.

Q. And you would have a number of the ballots that had been [Page 112] identified in your data base as being valid prior to them being counted; isn’t that right?

A. Yes, but that number would be continuing to increase from the time we start counting absentee ballots to the certification of the election because we continue to receive valid absentee ballots. So every day we’re counting starting on election day, and we’re also receiving additional ballots. So there is no – there would not be a manner on which to – on each day of tabulation compare that to a number of ballots that are in the voter registration system as being returned.

Q. On the day of certification or the day after certification, you would be able to tell from your data base how many absentee ballots were verified?

A. Yes, we would be able to get a number of ballots indicated in the voter registration system as having been verified, yes.

Q. Is there any reason that that number would change from initial count through any of the recounts?

A. The number in the voter registration system, no, not unless – I mean, unless there was clean-up work being done in terms of, you know, errors made, human errors made in terms of if close to the certification of the election, you’re really pushing the deadline to be able to certify the election. And also the deadline to, for instance, where the signatures didn’t match or where a ballot was unsigned, we receive updates to those up to the day before the [Page 113] certification of the election. Those have to be processed. The ballots have to be counted and new reports have to be run before we actually do the certification on the day of the certification.
So there is a high risk during that time period that in the interest of getting the ballot processed and counted prior to certification, that the – the rejection code may not have been properly removed from the voter registration system just in terms of the sheer crunched time period in terms of getting that ready. So I would agree with what you stated that there is a – there is the ability to get a report of the number of ballots that have been verified from the voter registration system. There may have been clean-up work being done after certification to go back and be sure that we had adequately removed the rejection records or credited voters who had corrected the problem with their signature or something of that nature.

Q. Okay. Now, on Exhibit 1, the second page, which is the Mail Ballot Report, there is a Line No. 3, Total number accepted as valid and counted. Is that number produced from the data base after verification, or is that based on some other –

A. No, that number is produced, and I think we said that for that number on No. 3 is produced from the vote tabulation system to show that this is the number of ballots that are differentiated as absentee ballots that have in fact been [Page 114] counted in the election.
###
[Page 217] A....The errors in the reports that were produced as far as the mail ballot report and provisional ballot report, I don’t know that I would classify those as errors in the conduct of the election. I think that there were errors in the – errors or deficiencies in the reporting that was provided to me in the documentation that was put together. I [Page 218] would not consider that an error in the conduct of the election.

Fell says Huennekens knew

Pages 239 – 241 of Garth Fell’s deposition contain Fell’s statement that his immediate superior – King County’s Superintendent of Elections, Bill Huennekens – knew of the faked “mail ballot report.” (Hat tip and thanks to Stefan Sharkansky for posting the transcript at Sound Politics.)

It probably shouldn’t be surprising that Fell didn’t actually say it was “faked.” Note the part at the end of page 240 and beginning of 241 – the mail ballot report was an accurate “snapshot,” according to Fell.

At page 242, Fell agreed that the way the report was written meant that discrepancies in the ballot count wouldn’t be revealed. He was asked about the 95 valid but uncounted absentee ballots that were discovered in March 2005, but some day someone may finally ask them how many more absentee ballots were in their GEMS ballot count compared to the number of credited voters in DIMS. I believe that the absence of those 95 ballots from the count wasn’t their problem. The presence of hundreds of extra ballots in the GEMS vote tabulation was their problem.

^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^

[Page 239] Q. Contained in the exhibit is a Mail Ballot Report for the November 2004 election; is that right?

A. That is correct.

Q. Do you know who created the Mail Ballot Report?

A. Yes. This report was created by the Mail Ballot Operations Satellite supervisor.

Q. Nicole Way.

A. Nicole Way. In conjunction with myself and additional workings with the superintendent of Elections, Bill Huennekens. The actual document was created by [Page 240] Nicole Way, I believe, and the understanding of what went into the document was shared to my knowledge, between the three of us.

Q. What was your role in creating the Mail Ballot Report?

A. Simply as an advisory role on what the numbers were and what the numbers represent and how the numbers were derived.

Q. Prior to the Mail Ballot Report being given to the canvassing board, did you understand that the total number of ballots returned on the Mail Ballot Report was derived simply by adding the total number of ballots accepted as ballots, and counted, and total number of ballots rejected?

A. Yes, I did.

Q. Did you tell Bill Huennekens that is how the total number of ballots returned was derived?

A. Yes, I did.

Q. Do you know whether anyone told the canvassing board that that is how the total number of ballots returned was derived?

A. I do not.

Q. Is the Mail Ballot Report accurate?

A. The Mail Ballot Report was accurate at the time of its creation based on hand counts and tabulation [Page 241] counts.

Q. Is it no longer accurate?

A. The data associated with some of these categories has changed, and so it is a snapshot at that particular time, at the time of certification. So I imagine that we have received more ballots back from the United States Postal Service. That’s the certification of election. There’s a potential for that. So I don’t know if that number is correct.

###

[Page 242] Q. If the total number of ballots returned was not derived simply by adding the total number of ballot hand counted and the total number rejected but was in fact an accounting by the actual number of ballots that were returned, wouldn’t the 95 have been accounted for in the total number of ballots returned?

A. If the total number of ballots returned was derived based on an independent count of all ballots returned, you would likely see a discrepancy of some sort, certainly.

Friday, May 20, 2005

What's DIMS got to do with it?

Nicole Way claimed during her deposition that the “DIMS” computer system didn’t accurately state the number of ballots returned, but it seems that she simply didn’t understand the DIMS system.

Appended below is an excerpt from pages 29 – 31 of the deposition transcript, which indicates that the DIMS system was doing exactly what it was designed to do.

DIMS is supposed to provide an alert to the employees who are processing returned absentee ballots when there is a reason to set aside a ballot for further investigation before tabulating the votes on it.

For example, if a voter requested and was issued a second absentee ballot, the first ballot issued to that voter (which can be identified by the bar code on the return envelope) must be set aside to ensure that only one of the two ballots is accepted as valid.

The second example Way gave involved a situation in which one would hope the computer system would place barriers in the way of any elections office employee who wanted to accept the ballot as valid: It involves the receipt of a second ballot from the same voter.

It seems apparent that the computer system was functioning just fine, but Way and the others seem not to have figured out how to maintain a cumulative total of all ballots returned by voters – even though they had that information in their database as a result of scanning the bar codes on the return envelopes. (Mistakenly skipping an envelope’s bar code would not result in rejection of the ballot by the system, since the system wouldn’t even “know” the ballot was in the possession of the elections office.)

Nicole Way seems to be making excuses for what she did when she submitted a false mail ballot report to hide the fact that more ballots had been counted by the GEMS vote tabulating system than the number of voters credited in the DIMS database as having voted.

It was a material discrepancy that needed to be resolved before anyone could know whether the abstract of votes was a full, true and correct representation of the votes cast.

Way and her immediate superior, Garth Fell, couldn’t resolve it, so they had the duty to bring it to the attention of the canvassing board.

They chose to lie rather than bring the matter to the board.

Their choice constituted fraud, since their false mail ballot report was presented to the board as an authentic statement of ballot reconciliation upon which the board was expected to rely in certifying the official canvass report.

That fraud makes the (almost but not quite) certification of King County’s election results by the canvassing board a meaningless act unworthy of belief.

The certificate of election issued to Gregoire by the Speaker of the House and President of the Senate was issued in reliance on that meaningless King County certification, and thus was issued in error.

^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^ - ^

Pages 29 – 31 of Nicole Way’s deposition:

[Page 29]
Q. What kinds of problems would occur in uploading the information that would make the data in DIMS with respect to the number of absentee ballots returned unreliable?

A. DIMS would automatically reject some voter ID’s.

Q. Were there particular voter ID’s that would be rejected?

A. No. There was a number of reasons why it would happen.

Q. What are the reasons?

A. If the ballot was reissued, the first ID number was suspended, and the voter was given a new ID number, so that suspended ID would be rejected. If the voter had already returned a ballot. If the ballot was from a wrong election. If the voter had been inactivated for some reason after we mailed the ballot. I think there was others, but I don’t remember.

Q. Were those categories referred to as Wanda rejects?

A. Yes.

Q. What happened to Wanda rejects after DIMS automatically rejected them?

A. We tried to sort them to their different reject reasons. Then they were put aside until the end.
[Page 30]
Q. And what happens to them at the end?

A. They’re looked up to see if that ballot should be counted.

Q. And by the end, you mean, before certification of the election?

A. Yes.

Q. Wanda rejects that aren’t later counted don’t appear anywhere in the DIMS system?

A. Right.

Q. If it’s later determined they should be counted, will they appear in the DIMS system?

A. Yes.

Q. How about for the Wanda rejects that are never brought back into the – to be counted pile? Are they accounted for in DIMS?

A. No.

Q. Are they accounted for anywhere?

A. They are counted as a challenged uncounted ballot, but there [Page 31] is no record of that voter.

Q. Does somebody do a hand-count of those ballots?

A. Yes.

Q. When is the hand-count done?

A. Shortly before certification.

Q. Is it only done once?

A. I’m not sure.

Way's Way to Fake a Mail Ballot Report

Nicole Way, the supervisor of mail ballot operations for King County (who is now on administrative leave after the discovery of a few significant errors in that operation), seems to be a bureaucrat who did things simply because she thinks they’ve always been done that way. Little in her deposition indicates that she grasps, or even cares about, the purpose or objective involved in the work and the steps taken to accomplish the work.

The transcript of Way’s deposition has been posted at Sound Politics by Stefan Sharkansky.

Note in this excerpt how the false mail ballot report prepared by Way omitted all information which would have shown a discrepancy involving more ballots in the vote tabulation than had been accepted as valid ballots.

Note also that she seems not to have been concerned about the effect of falsifying that report. The idea of reporting the matter to the canvassing board never seems to have crossed her mind, even though anyone with common sense plus a well developed sense of right and wrong would have thought of that as the only thing to do.

In the falsified report, the number for absentee ballots returned was nothing more than the ballot count from the “GEMS” vote tabulating system plus the number of ballots that weren’t sent through the vote count. Nothing about that total would enable anyone to determine whether all returned ballots had been processed in preparation for counting the votes on valid ballots.

The number for “valid and counted” was simply the ballot count from “GEMS.” Nothing about that number would indicate whether the number of ballots sent through the vote tabulation machines was equal to, less than, or greater than the number of valid ballots that belonged in the vote count.

The mail ballot report simply said “we counted the votes on this many, and we excluded this many from the vote count.”

That was perhaps an accurate depiction of what was physically done with the ballots that were set aside and those that were sent through the vote counting machines, but it provided no information from which the canvassing board could verify that the correct number of ballots had gone through the vote count or been set aside as invalid.

Pages 47 – 50, regarding the mail ballot report of November 17, 2004:

Q. (By Mr. Maguire) The second line is, Total number of ballots returned, 568,333; is that right?

A. Yes.

Q. Do you know where the data for this entry came from?

A. Yes.

[Page 48]

Q. Where did it come from?

A. We took the total amount of ballots counted from GEMS and added up the hand-count of the ballots that were not counted.

Q. So that number is not a report from DIMS?

A. No.

Q. That’s different than the information contained on your spreadsheet?

A. Correct.

Q. Did you decide to use that different number there than what was on your spreadsheet, or a different method of calculating total number of ballots returned?

A. The number on my spreadsheet was total number of ballots loaded into DIMS. It didn’t include the various things we’ve already talked about. So we couldn’t use that number.

Q. Did you have a discussion with anyone when you were creating the mail ballot report as to what information should be reflected in the total number of ballots returned?

A. Yes.

Q. With whom did you have the discussion?

A. Garth Fell.

Q. And what was the substance of the discussion?

A. We discussed how to fill out this report, because we didn’t have an accurate number of ballots returned.

Q. Did Mr. Fell decide that the number reflected on the mail ballot report of ballots returned should be the sum of the

[Page 49]

total counted by GEMS and those rejected?

A. I believe so.

Q. Was anyone else involved in the conversation between you and Mr. Fell on this subject?

A. I only discussed this with Garth Fell.

Q. You’re not sure whether it was his idea or your idea?

A. We discussed how to fill out this report, and we both agreed that the only thing we could do is take the GEMS count and add in what we hadn’t counted.

Q. Did you discuss any other alternatives?

A. Not that I recall.

Q. The third line on the mail ballot report is the, Total number accepted valid and counted is 546,222. Is that right?

Mr. Porter: Were you asking her if that’s the right figure?

Mr. Maguire: No, is that what the document says.

The Witness: 564,222 is what the document says.

Q. (By Mr. Maguire) Do you know how that number was derived?

A. That’s how many ballots the GEMS data base showed as being tabulated.

Q. And the mail ballot report indicates 4,111, Total number of ballots rejected. That’s what the report says?

A. Yes.

Q. Do you know how that number was calculated?

A. We hand-counted all of the ballots that were rejected, and [Page 50] that’s the total of the various categories.

Q. Would the GEMS report used for the total number accepted as valid and counted reflect ballots cast by individuals in the Address Confidentiality Program?

A. Yes.

Q. And individuals who cast federal write-in ballots?

A. Yes.

Thursday, May 19, 2005

"We've always done it this way," says spin-woman Bobbie Egan

What an enormous relief! There was no intent to mislead by presenting a false “mail ballot report” to the King County canvassing board. In fact, it wasn’t actually a falsification at all, says King County’s spin-woman, Bobbie Egan.

As reported by King 5 News today:

SEATTLE - King County's absentee-ballot supervisor has testified that she collaborated with her boss when she filled out a report that falsely showed all ballots were accounted for in the November election, The Seattle Times reported Thursday.
###
By law, counties must reconcile the number of absentee ballots returned by voters with the number of ballots accepted or rejected. Way's report showed perfect reconciliation because it simply added the number accepted and rejected to calculate ballots returned.

On Thursday morning, however, Elections spokeswoman Bobbie Egan said that ballot reports have always been created that way and that there was no falsification.

Egan said that any discrepancy between the numbers shown on the report and the actual number of absentee ballots was an oversight, not an intentional effort to mislead.

Is there a reliable way to distinguish between sheer stupidity and a dumb lie?

To determine whether the vote tabulation has included all legitimate absentee ballots, and whether the legitimate ballots have each been sent through the tabulating machines only once, the canvassing board must know how many ballots were available to be sent through.

To know how many there were, the board must know the answers to these questions:
How many absentee ballots were returned by voters?
How many returned ballots were accepted as valid?
How many returned ballots were rejected as invalid?
How many absentee ballots were sent through the vote tabulation machines?

The sum of the numbers of ballots rejected and ballots accepted should equal ballots returned. If the numbers balance, then one can be reasonably sure that all returned ballots have been processed in preparation for the vote tabulation.

The number of ballots accepted should equal the number of ballots sent through the vote tabulating machines. If the numbers balance, then one can be reasonably sure that all accepted ballots have been counted, that they have been counted only once, and that no illegitimate ballots have somehow been sent through the tabulating machines.

It appears that the only number King County’s elections office was pretty sure of was the number of absentee ballots that were sent through the vote tabulating machines, but they had no way of knowing whether that number represented all valid ballots or whether some ballots had been sent through more than once – or whether some ballots that were illegitimate had been sent through.

The false mail ballot report presented to the canvassing board lumped two essential numbers together. Rather than report the number of ballots accepted as valid and the number of ballots sent through the vote tabulating machines, it stated the number of “valid and counted” ballots.

The only way one could say that the false report was done with anything other than an intent to deceive would be to say that even a 10-year-old child shouldn’t have been fooled by it, and the people who prepared and presented it didn't expect the canvassing board to be misled.

Unfortunately, the King County canvassing board didn’t function at or above the level of a 10-year-old child, so they were apparently fooled into believing that the abstract of votes, at least with regard to absentee ballots, was a “full, true, and correct representation of the votes cast.”

The abstract obviously wasn’t correct.

And, it is reasonably certain that Nicole Way and Garth Fell realized that the only number they could state in good faith was the number of ballots sent through the vote tabulating machines. That number was available to them in the report generated by their “GEMS” vote tabulating system.

They didn’t know how many had been accepted. They didn’t know how many had been rejected. They didn’t know how many had been returned. They didn’t know how many were omitted from the vote tabulation. They didn’t know how many were sent through the tabulation more than once. They didn’t know how many rejected ballots were nevertheless sent through the vote tabulating machines.

So they made it up and presented a report to the canvassing board which indicated that all was well – indeed, all was perfect.

Wednesday, May 18, 2005

Voting more than once

At the Secretary of State's web site is a really large "pdf" file that contains, among other things, an excerpt of the deposition of Colleen Kwan and a copy of the spreadsheet King County disclosed after the conclusion of Dean Logan's deposition – the one that showed 437 provisional ballots were inserted into Accuvote machines in addition to the 348 "PBAVs" disclosed in January.

The file is titled Petitioners' Brief in Response to WSDCC's Motion on Dual Voters. (The crux of the motion and reply is that the Democrats want only one of two ballots cast by any "dual voter" to be treated as an illegal vote, and the Republicans want both to be treated as illegal votes.)

Kwan's deposition and the accompanying exhibits are on pages 31-76 of the pdf file. The spreadsheet showing 437 more PBAVs is the work of Kwan and people assigned to assist her.

Kwan completed this particular assignment on March 23, 2005; but Logan claimed during his April 18-19 deposition that he believed there were no more than 660 PBAVs. Interesting: Kwan's work showed the presence of 437 in addition to the initially disclosed 348 – for a total of 785. Did Logan not know what Kwan's work revealed? Was it his lack of knowledge that caused King County to wait until the conclusion of his deposition on April 19 to disclose Kwan's spreadsheet to the petitioners? Who's in charge over there?

The spreadsheet showing 437 more PBAVs indicates that 97 of the people who apparently inserted those ballots into the ballot boxes were credited with casting another ballot in addition to the one unlawfully inserted into the Accuvote machines.

Previously, the spreadsheet showing a different set of unlawfully cast provisional ballots totaling 348 had shown that 40 were cast by people who had been credited with casting another ballot.

That appears to be a total of 137 provisional ballots out of the 785 so far identified by King County which were used by people who cast more than one ballot in the election.

I say "appears to be," because the petitioners in their brief state that they understand the total to be 129. Perhaps I'm misunderstanding – or they have the number 129 imprinted so firmly in their minds that it just pops out once in a while. (Update May 19, 6:08 PM: At Sound Politics, Stefan Sharkansky has posted a copy of the petitioners' subpoena and list of presumed double voters. The list contains 129 names. So, apparently there is a reason why they believe the total to be 129 and not 97+40=137.)

Did some people in King County take advantage of the amazingly sloppy administration of the election to vote more than once? Apparently, at least 137 did.

The percentage of ballots cast by poor, confused, innocent, registered voters who just didn't understand the terribly complicated instructions ("put this ballot in this envelope after you mark your votes and bring it to me") among this latest group of 437 appears to have been 59%.

Of the 437 ballots, 355 were cast by registered voters. Of those 355, there were 97 cast by people who didn't limit themselves to casting only one ballot during the election. (They are apparently not poor, innocent people who simply failed to follow instructions.)

Subtracting the scofflaws from the 355 leaves 258 provisional ballots that were unlawfully inserted into the ballot boxes by registered voters who apparently didn't vote more than once. (Of course, King County's records are so sloppy that all of them may have cast more than one ballot, but apparently only 97 of this group of 437 did so.)

So, how do we subtract the scofflaws' votes from the count?

Possible torpedo sighted off the port bow

There’s an interesting part on page 4 of “WSDCC’s Opposition to Petitioners’ Motion to Strike WSDCC’s Rebuttal Witness List” dated May 17.

It appears that the Democrats may call witnesses who will say for whom they cast their votes:

Next, Petitioners take issue with two classes of rebuttal witnesses who are Washington voters who may testify regarding, among other issues, “for whom they cast their ballot.” Motion at 2-3. WSDCC understands that Petitioners will offer expert evidence that voters in certain counties and precincts were more likely to have cast ballots for Dino Rossi than for Christine Gregoire in the 2004 Gubernatorial Election. (sic – Did they mean to say “more likely to have cast ballots for Christine Gregoire”?) These rebuttal witnesses may be offered to rebut Petitioners’ theory that precinct-based proportionate reduction is sufficient proof of for whom individual voters voted in the gubernatorial election. By testifying regarding their voting precinct and for whom they voted, these witnesses may directly rebut Petitioners’ theory of proportional reduction.

Have the Democrats located a few voters who were disqualified from voting and who will say for which candidate they claim to have voted?

That’s an interesting way to attack “proportional reduction.” Pick some precincts with few such voters and see if you can find the ones who will say they voted for Rossi.

The petitioners are trying to avoid calling hundreds of such witnesses in an effort to show how the disqualified voters actually cast their votes. The Democrats may get a lot of mileage out of a handful of them.

Tuesday, May 17, 2005

Ron Sims' task force gets underway

As the beginning of the election contest trial nears, the task force appointed by King County executive Ron Sims has begun its effort to examine the mess that was made of the general election by Logan’s gang.

The group is headed by Cheryl Scott, who proposes to search for “root causes” according to The Seattle Times:

Cheryl Scott, chairwoman of Sims' Independent Task Force, said her panel will hunt for "root causes" of election problems, then make recommendations to restore public trust and turn King County into a national model for efficient elections. Scott is a former CEO of Group Health Cooperative.

According to today's edition of the Seattle Post-Intelligencer, a staff member named Brian Malarky (no relation to Morton Brilliant, I suppose) hopes to find the “big picture”:

The task force, which includes prominent representatives of academia and the law as well as elections officials, will look at the "big picture," staff member Brian Malarky said at the group's first meeting yesterday at Seattle University.

If Sims’ task force intends to find the “big picture” and identify “root causes,” they need to see if they can agree on this policy: Our elections ought to be decided by the legitimate votes of eligible voters, not by simply counting the votes on every ballot someone manages to insert into the vote tabulation process.

If they think some other policy ought to be implemented in conducting elections, they should let us know at the outset – so we can avoid wasting time paying any attention to them and their eventual report.

As for “root causes,” they need to begin at the top and work their way down to the bottom rung of the elections division. Start with Dean Logan, who stated in his deposition that the false ballot reconciliation reports presented to the canvassing board aren’t part of conducting an election. Move on to Bill Huennekens, who stated in his deposition that he didn’t know what was meant by the phrase, validity of each vote, when he was asked if the canvassing process is essential to ensure the validity of each vote and to protect the integrity of our elections process.

Then ask Linda Sanchez who it was that decided to hide the fact of 900+ ballots inserted into the ballot boxes in excess of the number of regular ballots cast by eligible voters at polling places. She, Logan and Huennekens were aware of the substantial discrepancy, and someone – or all three of them – decided to conceal the problem from the canvassing board.

And don’t forget to ask why King County failed to keep track of the total numbers of regular polling place ballots, provisional ballots and absentee ballots that they received from voters. Their records, such as they are, show that they counted hundreds more ballots in each category than had been cast by eligible voters. If they had desired to ensure that they counted only legitimate votes, counted them all, and counted them only once, they needed to know how many were available to be counted – but they have not yet claimed to know that most basic information.

That ought to get the task force started. Perhaps they will see the “big picture” (a desperate effort to include every ballot, legitimate or illegitimate, in the vote tabulation in order to erase Rossi’s slim lead) and the “root causes” (incompetence magnified by a desire to achieve a certain outcome).

Monday, May 16, 2005

Adolph & Handcock for the Democrats

Are the Democrats’ experts playing a shell game in their reports on the question whether illegal votes changed the outcome of the gubernatorial election?

Two reports written by expert witnesses for the Democrats have been posted at the Secretary of State’s web site.

While there are other points made in those reports which may be sound, one criticism of the Republicans’ expert witnesses’ reports is common to both and seems to miss the point.

Assuming the judge is correct that prevailing under RCW 29A.68.110 requires proof that “illegal votes” changed the outcome of the election, then the question is whether votes which can be shown to have been illegal votes caused Gregoire to be declared the winner of the election.

The question is not whether all illegal votes – known and unknown, proven and unproven – changed the outcome, but is instead whether the ones that are known and can be proven changed the outcome of the election.

Note how the two expert witnesses for the Democrats base their criticisms on the inability of the Republicans’ experts to know whether they have included all illegal votes in their analyses.

Note especially how Mr. Christopher Adolph shifts from talking about the need to include “all invalid votes” to “these alleged invalid ballots” which have been identified by the petitioners. The absence of a complete “census” of all invalid ballots is not only a flaw: It is possibly a bias designed into the data to favor the petitioners.

In a perfect world, we would have a set of data that included all illegal votes cast in Washington’s gubernatorial election.

This isn’t a perfect world, as evidenced by the careless way in which the election was administered in King County.

So, the question in this imperfect world is: What was the effect of the illegal votes which we do know about?

Here are relevant excerpts from the reports by the Democrats’ expert witnesses. Are they dealing with the actual question in this case or trying to shift to a different question?

Page 2 of the report written by Christopher Adolph:

The three independent problems in Gill and Katz are:
1. Use of a non-random, non-representative, and incomplete sample of invalid votes, which is useless for answering questions about the net effect of all invalid votes on the state-wide election outcome. (Emphasis added.)

Page 3 of Adolph’s report:

Are the petitioners’ data a census, a random sample, or a representative sample of invalid ballots? Neither Gill nor Katz claim they are. This omission is unusual and conspicuous. The first thing another scholar would ask about these reports is “Where did the data come from? Are the data a valid sample.” Unless the petitioners can make and support the claim that these data are a representative sample of invalid ballots across the whole state, it will be impossible to make even minimally valid scientific claims regarding the likely effect of these alleged invalid ballots on the election outcome. In particular, if petitioners over-sampled precincts that voted overall for Gregoire, then their method will tend to produce biased results suggesting, perhaps incorrectly, that Gregoire benefited from the inclusion of invalid votes.

Page 4 of the report written by Mark S. Handcock:

2.1 Assumption I: Their data set is a known fraction of the invalid voters

This assumption states that the list of invalid ballots in their data set is either: (1) a complete enumeration of the (sic) all invalid ballots cast in the election, or (2) a sample of the invalid ballots, where the probability of each ballot entering their sample is known.

The first approach would claim that the set of invalid ballots in their dataset is a true census of the invalid votes cast. This means that they have every invalid voter and have not included any valid voters as invalid voters. If this assumption was correct, then there would be no question of “design based” statistical inference from our sample to the population of invalid voters: this is the population of invalid voters. Assuming for a moment that we knew how these invalid voters had actually voted, then we would be able to say with certainty whether these votes had changed the election outcome. The findings obtained from our data would be the findings of interest, there would be no need to evaluate whether an inference can be made from these data to the set of all invalid votes (and by extension, to the election outcome). (Emphasis added.)

Saturday, May 14, 2005

Sam Reed answers plainly

Secretary of State Sam Reed apparently understands the role of county canvassing boards regarding the certification of a county’s election results.

Excerpts of the deposition of Secretary of State Sam Reed have been posted on the SecState web site as part of a document filed by the petitioners.

At page 141, the question and answer present the crux of the matter regarding the concealment of discrepancies by Logan’s gang in King County:

Q. So when you receive certified returns from counties, you are relying on the county canvassing boards investigating and identifying any problems and resolving them to the extent they can prior to certifying the returns for their county?

A. That is correct. It is the role of the canvassing board to – if there are any problems, to work to resolve those before they certify the election.

The question and answer which immediately follow are the sort of thing the news media and many partisans love to quote, but they aren’t as important as the clear statement of the canvassing board’s role.

Q. Do you know whether the certified returns of the manual recount were accurate within 129 votes?

A. I do not know that.

No one can know that for certain, when there are almost three million ballots being counted, and the apparent margin of victory is so slight.

But when there are significant discrepancies affecting the validity of the results reported by King County, the issue is not whether the statewide total was accurate to within 129 votes. Instead, the issue is whether the reported election results are at all credible.

Canvassing boards are presumed to perform their duties in a careful fashion, so their certified election results are presumed to be correct; but when the substantial and plentiful discrepancies in King County were concealed from the canvassing board, no such presumptions can be given effect. Indeed, it ought to be presumed that the deliberate concealment of discrepancies made the county’s official canvass report unworthy of belief.

The excerpts from Reed’s deposition contain other interesting statements.

In two places, Reed was asked about his opinion of Dean Logan’s performance as King County’s chief elections officer and as a member of the canvassing board.

At pages 35-36, Reed described observing a King County canvassing board meeting at which Logan advocated further research into the circumstances connected with ballots which had been misplaced and then found in late November. Reed’s opinion was that Logan should have been recommending against including such ballots in the vote count, since they had not been in secure storage.

And, at pages 96-97, Reed stated:

Q. And Ms. Durkan asked you about the sentence in which you wrote, “I disagree with some of his decisions and actions this year, but I know he is now reporting to Ron Sims, not Ralph or me.” Is that correct?

A. That is correct.

Q. And the person you were talking about was Dean Logan?

A. Correct.

Q. What did you mean by the reference to Mr. Logan “reporting to Ron Sims, not Ralph or me”?

A. The example I gave to Ms. Durkan was how he handled the canvassing board that I was at in terms of whether to let the prosecutor’s office have more time to look at this – they voted no by a two-to-one vote – whether to just say, “Since these ballots have been sitting out at schools and churches unsecured, we just shouldn’t consider them.” Instead they did the opposite. And, you know, again, if he were reporting to me, I wouldn’t have accepted that at all. But obviously, he doesn’t report to me. He reports to Ron Sims. And I’m kind of assuming that that must have been, you know, the way he was supposed to be doing his job there.

Perhaps Reed has noticed that Logan and Huennekens appeared to focus their efforts on getting as many ballots into the vote count as possible once the close gubernatorial election became apparent. For example, signatures on provisional ballot envelopes were examined multiple times in an obvious effort to find someone who would be willing to say that the signatures matched those on file. And, the 785+ unlawfully cast provisional ballots were concealed from the canvassing board rather than being removed and referred to the board for a decision whether to reject them.

At pages 75-76, Reed described the role of the canvassing board with regard to remedying errors before certification – a role that was sabotaged by the concealment of the problem involving 785+ unlawfully cast provisional ballots:

Q. And is part of the process that mistakes that are made are corrected prior to certification?

A. The county canvassing board – it’s the role of the county canvassing board, if a mistake is reported to them, to see that it is corrected before they certify, right.

At page 81, Reed stated the proper remedy for a situation involving the presence of unlawfully cast provisional ballots in the ballot boxes – take them out:

Q. And by having individual pieces of equipment at each poll site that allows ballots to be tabulated at a poll site, does that increase the possibility that provisional ballots would be counted prior to being verified to determine whether they were cast by a lawful registered voter?

A. Not necessarily. Obviously, it did in this case. But with proper preparation, that shouldn’t happen.

Q. What kind of proper preparation could prevent that from happening?

A. By either bar coding those ballots so they don’t – so they will be rejected by the precinct counts. In Thurston County, what we do is we have different colors of ballots. So if somebody, in this case because it’s a punch card, drops the card in the ballot box rather than giving it back to the election board worker, we can see that it was a provisional ballot and not supposed to be in there, and separate it out. So there are ways of dealing with the issue.

Starting at page 97 and going on for many pages are questions and answers regarding the essential elements of ballot accountability and reconciliation of ballots cast and counted. It’s worth reading it all.

At pages 99-100, Reed was asked about the “audit trail” needed for polling place ballots, and his answer clearly and plainly stated the required voter-to-ballot reconciliation of total numbers at polling places:

Q. And why should a County keep an audit trail?

A. Again, it’s your basic elections practice to not allow for the stuffing of the ballot box by knowing how many, you know, voters you had that cast a ballot compared to how many you actually have within your ballot box.

At page 124, Reed stated what is plainly true about voter crediting with regard to absentee ballots – but has been obscured by the smokescreen spewed out of the “talking points” coordinated among the auditors of Kitsap, Pierce, Snohomish, and Clark counties, the state elections director, Nick Handy, and Dean Logan of King County:

Q. Ms. Durkan was asking you some questions about crediting. Mr. Secretary, do you know whether the crediting of absentee ballots takes place during the canvass?

A. The crediting of absentee ballots takes place at the time that they receive them. And the reason they do that is so they’ll know if they have more than one ballot that has arrived in from a person or in case there’s a provisional ballot cast by the same person.

Q. The crediting process with respect to absentee helps identify whether the person has submitted multiple ballots?

A. That is correct, right.

Q. It’s important that that is done prior to certification?

A. That’s correct. That’s a control feature.

Dean Logan has claimed that county canvassing boards have no choice but to certify the election results as of the deadline – no matter what.

Note this simple, straightforward answer from Reed at pages 140-141 about certification of county election results:

Q. Mr. Secretary, should a County certify results that do not accurately reflect the number of lawful votes cast?

A. No.

Don’t you just love it when a question which can be truthfully answered with a “yes” or a “no” is actually answered that way?

Reading these excerpts from Reed’s deposition was like watching miners drilling holes for the placement of explosive charges.

Let’s hope we soon get to the part where someone yells, “Fire in the hole!”

Thursday, May 12, 2005

A closer look at King County

The petitioners in the Washington gubernatorial election seem to be cutting it awfully close with the timing of their depositions.

On the Secretary of State's web site is a subpoena dated May 11 which is directed to King County Division of Records, Elections and Licensing Services. The subject of the deposition seems to include, among other things, the so-called ballot reconciliation process done by King County. The deposition would occur on May 19 according to the subpoena -- only four days before the start of trial.

Here's hoping that the petitioners already have a pretty good idea what the answers to their questions are.

Tuesday, May 10, 2005

Repeat after me: State law requires....

While it remains to be seen whether the Republicans’ lawyers can repair the damage done so far, there is a faint glimmer of hope in today’s press release.

The party chairman, Chris Vance, actually said some of the magic words about the essential elements of canvassing the votes and certifying a county’s election results:

In addition, Vance reiterated that the Republican case is not just about illegal votes and proportional analysis. State law requires counties to reconcile the results of the election before the certification, and recent depositions of King County election officials have added to the mountain of evidence regarding how flawed King County’s reconciliation was. “In some areas there are still hundreds more ballots than voters, in others there are more voters than ballots, and now we learn that King County had no real security over blank ballots and no valid absentee audit trail. Those facts will be part of our case,” said Vance.

It’s a small thing, I suppose, but it would be nice to hear the Republicans’ lawyers say in their opening statement at least 10 or 11 times: “State law requires....”

State law requires “voter crediting” before certification of the election results, therefore it is not a “post-election administrative exercise that does not bear upon the authenticity of the election results” – no matter how many times Nick Handy and the auditors who coordinated their “talking points” with him say differently.

State law requires the crediting of voters with having voted at the polls by a notation made in the poll books at the time the ballot is issued – that is, before the ballot is voted, much less cast. It is beyond cavil that voter crediting at the polls is accomplished before the polls close, therefore it is impossible for voter crediting to be honestly described as a “post-election administrative exercise.”

State law requires an examination of the poll books to ensure that voters don’t successfully cast more than one ballot by casting a regular ballot at the precinct in which they are registered and a provisional ballot at another polling place – or an absentee ballot in addition to one or both of the other types of ballots. If processing more than 30,000 provisional ballots makes it necessary to transfer the voter crediting notations to the computerized voter registration database – to avoid having to go to the poll books before deciding whether to accept each one of those ballots – then the transfer of data is done before certification of the election, not after, just as it was done in King County. If the transfer of data was done in a careless, sloppy and mistake-ridden fashion, that raises more doubt about the accuracy of their election results, not less.

State law requires counting the signatures of voters who signed the poll books before being issued regular ballots and comparing that number to the number of ballots in the box – and remedying the situation created by ballots in excess of the number of regular ballots that ought to be in the boxes. No matter how many times Nick Handy, Dean Logan, and the others say differently, it is clear that reconciling the number of voters credited with voting at the polls and the number of ballots in the ballot boxes is required by state law.

State law requires the crediting of voters with having voted by absentee ballot by notations made in the computerized voter registration records at the time their ballots are accepted as valid. Thus, absentee voters may be credited with voting even before election day – hardly something which could truthfully be called a “post-election administrative exercise.”

State law requires an “audit trail” that shows how many absentee ballots were returned by voters and what happened to each ballot – thus requiring a reconciliation of the numbers of voters credited with voting and the number of absentee ballots included in the vote tabulation.

State law requires the county’s chief elections officer to swear to the authenticity of the information presented to the canvassing board – not just that he hasn’t altered or erased the ballots submitted by voters (a silly oath by Dean Logan anyway, in light of the tens of thousands of enhanced ballots).

State law requires the chief elections officer and his subordinates to refer ballots of questionable validity to the canvassing board for a decision as to their disposition.

State law requires the chief elections officer and his personnel to bring discrepancies in the counts of voters and ballots to the attention of the canvassing board when those discrepancies cannot be resolved by showing that the counts actually reconcile with one another. It is the canvassing board’s duty to determine what remedy is appropriate when there are invalid or questionable ballots in the boxes.

State law requires a truthful presentation of the facts to the canvassing board when substantial discrepancies are recognized, for example, the discrepancies known to Dean Logan and his subordinates when King County included roughly 780 more absentee ballots, 785 more polling place ballots, and 780 more provisional ballots in the vote tabulation than could legitimately be in the count according to their own records.

State law requires the canvassing board to verify the accuracy of the auditor’s abstract of votes, not just certify that it is a true copy of itself – as King County’s canvassing board so ignorantly did.

State law requires the removal of ballots not lawfully cast and referral to the canvassing board for a decision whether to reject them – including both absentee ballots and provisional ballots that are unlawfully inserted into the ballot boxes via Accuvote machines at polling places.

State law requires the canvassing board to exclude from the vote count all illegitimate ballots which can be identified and removed. When the canvassing personnel discover that a certain number of invalid ballots is in a polling place's ballot box, they have a duty under state law to inspect the ballots and remove the absentee ballots and folded, creased provisional ballots that are clearly identifiable as being something other than valid, regular polling place ballots.

State law requires the inclusion of all legitimate ballots in the vote tabulation, which is why state law requires an audit trail for absentee ballots, a separate system for tracking the provisional ballots, and yet a third system for reconciliation of ballots by precinct which were cast at polling places.

State law requires that the number of ballots included in the vote tabulation equal the number of ballots accepted as valid, which is why state law requires the chief election officer to swear to the authenticity of information presented to the canvassing board – rather than claiming ignorance regarding false and misleading absentee and provisional ballot reports and about the deliberate concealment from the canvassing board of discrepancies involving hundreds of provisional ballots inserted into the ballot boxes.

State law requires a good faith effort by the canvassing board to determine whether the abstract of votes is a full, true and correct representation of the votes cast – because others will be relying on that certification from the canvassing board rather than canvassing the ballots all over again to check the county board’s work.

How many times is that? Sixteen? More? Okay, then say "state law requires" at least 16 times in the first three minutes.

Somehow, the petitioners have to educate Judge Bridges and overcome the misleading statements from the Secretary of State and the Democratic Party which he has, so far, swallowed hook, line and sinker.

Assuming the petitioners can marshal the evidence and present it in court, it would help if the judge started off with a rough idea of what is actually required by state law.