Opinion
No. 020821/2010.
December 8, 2010.
Jaspan Schlesinger LLP, Garden City ( Steven R. Schlesinger of counsel), for petitioner. Bee Ready Fishbein Hatter Donovan, LLP, Mineola ( Peter Bee and Kenneth A. Gray of counsel), for Jack M. Martins, respondent. John E. Ryan, Mineola, for Republican Commissioner of the Nassau County Board of Elections. Thomas J. Garry and Robert McDonald, Mineola, for Democratic Commissioner of the Nassau County Board of Elections. Leventhal Sliney, Roslyn ( Steven G. Leventhal of counsel), for Joseph Mondello, Chairman of the Nassau County Republican Committee. Robert McDonald, Mineola, for Jay Jacobs, Chairman of the Nassau County Democratic Committee.
OPINION OF THE COURT
Over a more than 30-day period, employees of the Board of Elections of Nassau County, attorneys for the Republican and Democratic Commissioners, attorneys for the political leaders of each party, Joseph Mondello and Jay Jacobs, and the attorneys for the incumbent, Senator Craig M. Johnson, and the challenger, Jack M. Martins, have participated in an audit of voting machines, a challenge to affidavit and absentee ballots (before opening the envelopes), and a challenge to the face of the ballots once they were open.
Veracity Ballots
The earliest issue brought to the court's attention was a claim by the Democrats that the Republicans had fraudulently induced absentee ballots by sending letters with applications to prospective voters. This group of ballots was labeled "Veracity Ballots." After oral argument, the court denied counsel's application to preclude these ballots from being counted and they were ordered to be opened. ( Matter of Kolb v Casella, 270 AD2d 964 [4th Dept 2000].) There was nothing in the letter that would induce a fraudulent application, and, if a voter lied to get an absentee ballot, it was not caused by the letter.
Two-Hundred Eighty-Three Unopened Ballots
On November 29 or 30, 2010, the court learned that there were approximately 283 unopened affidavit ballots that had not been brought to the court's attention; the attorneys were aware of them, just not the court. Of these, 170 were consented to not be opened. The balance of 113 had been found to be unregistered as of election day by a bipartisan research team.
The Republican Commissioner agreed to follow the recommendations without further review. The Democratic Commissioner had never looked at these envelopes, or so the court was told. He had been too busy since election day to look at these ballot envelopes. He was given until the morning of December 1, 2010 to consider what position he wished to take on these affidavit ballot envelopes.
On December 1, 2010, counsel for the Democratic Commissioner stated that he wanted 48 of these opened because he believed that their registration had been canceled in error, as he indicated to the court. Of those he wanted the court to open, 38 were registered Democrats, two Republicans, and the balance of eight registered as blanks or independent. Mr. Ryan, counsel for the Republican Commissioner, expressed "shock" at hearing these numbers.
The court ruled on December 1, 2010, pursuant to the case of Matter of Mondello v Nassau County Bd. of Elections ( 6 AD3d 18 [2d Dept 2004]), that it did not have the power after election day to reinstate/reconstitute a voter's registration who had been removed from the registered roll of voters by the Board. More specifically, pursuant to Mondello:
"In a proceeding pursuant to Election Law § 16-106 for judicial review of the canvass of votes in a general election, the Supreme Court lacks the authority to render a determination as to whether a voter was `lawfully registered and eligible to vote' ( Matter of Corrigan v Board of Elections of Suffolk County, 38 AD2d 825, 827 [1972], affd 30 NY2d 603 [1972]; see Matter of Delgado v Sunderland, 97 NY2d 420, 423 [2002])." ( Id. at 20-21.)
Thus, the court is without authority to rule on whether those individuals who cast affidavit ballots were lawfully registered and qualified to vote. Furthermore:
"On the day of [an] election, if a voter is denied the right to vote in a general election, the voter may seek a court order pursuant to Election Law § 16-108 (Election Law former § 331). This section authorizes the voter to challenge the determination of the Board of Elections that he or she is not properly registered and further provides that if the voter establishes that he or she was unlawfully denied the right to vote the court shall direct that the voter `be allowed to vote at his [or her] polling place' (Election Law § 16-108 [3]; see Election Law § 8-302 [3] [e] [i])." ( Mondello at 22.)
When a voter does not pursue this remedy and they are not a registered voter pursuant to Board of Elections research and determination, then this court lacks the power to reconstitute or restore their registration.
The court also gave the Republican Commissioner's counsel the opportunity to review the remainder of the 113 envelopes to determine if he believed any of those should be reinstated if the court had the power to do so.
The following day, December 2, 2010, counsel for the Republican Commissioner produced 64 envelopes which have been preserved as court exhibit IV which, he contended, should be opened if the court followed the position of the Democratic Commissioner. These were not opened.
Review of Referee Rulings
On December 2, 2010, the court ruled on numerous decisions made by the Court Attorney Referee. Each of these rulings is reflected in the record of December 2. The court would only reference three of these now, those with handwriting that the court directed be counted, numbers 181, 182, and 183. Traditionally, ballots that contained writing which could distinguish the ballot from others cast, and mark that ballot for identification, would be excluded. ( Matter of Scanlon v Savago, 160 AD2d 1162 [2d Dept 1990].) In examining these three ballots, the court refuses to reject logic. There is clearly handwriting on each of them, writing that voters intentionally placed there, but what was their intent? Was there an intent to bring attention to their ballot of a third party, for example, a political leader?
One voter thought he should be voting on two amendments which were left off his ballot. He was wrong, since the amendments were intended for a different political area. Nevertheless, he wrote in the information he sought to impart. Another noted in the upper right-hand corner of the ballot where the letters SD and the number 7 are found, that he did not reside in School District 7 and wrote that this was not the correct school district for him. Of course, the SD referred to the senatorial district, and had nothing to do with the voter's school district (information you place on your New York State tax return). The third voter, after reading the printed words on the ballot front that stated that there were instructions on the other side of the ballot, drew a line in pencil, and on the reverse side of the ballot, which was blank, drew a large question mark.
It was, and is, the court's opinion that the writings on these ballots were not intended to identify them for some nefarious purpose and clearly were not done in a surreptitious way We live in a community in which the populace continuously wishes to be heard and make themselves heard. They blog, they text, and they tweet. They respond to news stories which they read on the Internet and then comment on each other's responses. With that as reality and the court's evaluation of these three ballots, it allowed them to be counted. All other court rulings as to each of the ballots are on the record, and are incorporated in this decision. The court is aware it would be simpler to exclude any ballot with handwriting, but that would be unfair to these voters.
The Voting Machines
This brings us to the new element, added this year to our election, the electronic voting machine. As a matter of law, the County must conduct an audit of 3% of all County voting machines pursuant to 9 NYCRR 6210.18 (c) (1)-(3). "Any discrepancies between the corresponding audit results and initial electronic vote counts shall be duly noted, along with a description of the actions taken by the county board of elections for resolution of discrepancies" ( 9 NYCRR 6210.18 [c] [2]).
Pursuant to the requirements for a 3% audit, 32 machines were audited in Nassau County. Seven of those machines were from the Seventh Senatorial District.
There were three basic types of errors found in the audit:
1. Fewer ballots in the ballot box than reflected on the machine (one machine in the Seventh Senatorial District);
2. More ballots in the ballot box than reflected by the machine count of ballots (two machines in the Seventh Senatorial District fell into this category; machine No. 541, two additional ballots added two votes to Johnson; No. 706, providing one additional vote for Johnson);
3. Machine count and ballot box are the same, but there was an undervote on a machine that was not detected by the visual audit. Machine No. 805 produced one additional vote for Martins, which was originally noted as an overvote.
Of these seven machines, excess ballots are found in two of them. The Democratic Commissioner found that any machine with excess ballots was an unresolvable discrepancy, while the Republican Commissioner found four reasons why this could occur:
• Election inspector error — the ballot box is accessible to the inspectors at the end of the night. The inspectors are asked to remove the ballot bin and put any remaining ballots inside the bin, replace into scanner, and reseal and lock. The inspectors may have put the additional ballots into the bin. The accounting of the ballot stubs can reconcile this.
• Voter error — ballot jams can be caused by voters improperly using scanners in any number of ways. Ballot jams can be released by lifting and lowering the metal bar in the back of the scanner, and inspectors are instructed in this process. If this action releases the jam, the ballot can drop into the ballot bin without being scanned and counted.
• Voter error — if a voter puts a ballot into the scanner before waiting for the previous ballot to finish scanning, the voter will get an error message. If the voter nevertheless presses "OK" on the screen, the ballot may drop into the box without being scanned.
• If a ballot was jammed, then another ballot was put in, the second ballot could release the jam; however, the first ballot would not have been scanned and counted.
"The presence of more ballots in the ballot box does not demonstrate that the scanner has `failed,' merely that the machine operated as it was designed to do — but with the result that some number (in this case, two) ballots were not scanned. The vote discrepancies are consistent with the presence of two additional ballots." (Republican Commissioner's comments for machine No. 541.)
It also should be noted, as the Republicans did, that the "stub" books, the pad from which the ballots were detached, have numbers that match up or would match up with all ballots found in the ballot box. The Democratic Commissioner made no comment as to this situation at any time on December 4, 2010.
Counsel for the Democratic Commissioner, in sum and substance, did not necessarily deny that the reasons stated were not correct or possible, but that he would not "resolve" the discrepancies. He found unresolvable discrepancies in three machines out of the seven audited in the Seventh Senatorial District (two had additional ballots). He further found unreconcilable discrepancies in eight out of an additional 25 machines that made up the balance of the 3% countywide audit. (Total of 32 machines equals 3% of all County machines.) And, of course, the Republicans believe all the discrepancies are reconcilable.
The additional eight are in machine (M) 95 — audit (A) 8 (two extra ballots); M 153 — A 9 (one extra ballot); M 104 — A 14 (one ballot); M 259 — A 16 (one less ballot); M 940 — A19 (two extra, one of which was for a wrong E.D.); M 463 — A 21 (two extra); M 24 — A 22 (one over); M 1011 — A 31 (two excess ballots). These ballots were apparently blank, but the Democratic Commissioner wrote that since the machine did not accurately count the ballots, it failed.
It is interesting to note that in the Seventh Senatorial District the Democratic Commissioner failed machine No. 805 (audit 5) where the ballot count matched, but the allocation of votes on the machine differed from the hand count. In addition, in a number of races, the hand count did not find overvotes that had been recorded by the machine.
"There is no way to `reconcile' these tabulation discrepancies." "Because the audit revealed that the DS 2000 did NOT accurately tabulate the ballots, this machine fails the audit." (Remarks from audit report by Democratic Commissioner.)
In the additionally audited machines, more specifically in M 112 — A 6, M 987 — A 10, M 485 — A 11, M 138 — A 13 and M 637 — A 26, in what appear to be similar discrepancies, the Democratic Commissioner passed these machines concluding that the discrepancies were due to how the machine registered certain marks that did not complete the ovals.
He wrote: "Because these discrepancies appear to result from the precise position and density of the marks [or] the amount of ink in and touching the oval."
It appears to the court that the finding on M 805 — A 5 was inconsistent with similar discrepancies on the above machines which were found resolvable, yet M 805 — A 5 was not.
This is the first time that this court, or any court in this state, has been faced with the scanner machines and the impact of the audit process on a contest. But for the new process, there would be nothing more to say and Mr. Martins would be declared the winner. However, because of the mandated audit process, the court must take into consideration numerous factors before reaching such a conclusion.
From the very first day of this process, counsel for Johnson has requested a manual count of the entire Seventh Senatorial District. He has never wavered from that position and now argues that the audit results support a manual canvas, not a 5% audit.
9 NYCRR 6210.18 (c) (1) provides for the methodology and the content of the reconciliation report; subdivision (c) (2) provides that any discrepancy between machine count and manual count be noted along with a description of the actions taken by the Board for resolution of the discrepancy. The Democratic Commissioner's report does not appear to reflect action taken by the Board for resolution of discrepancies. Rather, the report claims that the discrepancy is not resolvable. The Republican Commissioner's reports attempt to provide reasons for the excess ballots.
Counsel for the Republican Commissioner spoke of the aforementioned stubs and the matching of numbers on the stubs to stub numbers on the ballots themselves. This reconciliation, or "resolvable" factor, was never addressed by the Democratic Commissioner. It would appear that it should have been considered; perhaps it was considered, but simply not mentioned.
9 NYCRR 6210.18 (e) (1) (i) and (ii) provide what statistical discrepancies would require the expansion of the audit to an additional 5% of the County machines. As a result of the clearly partisan positions taken by counsel for each side of the Board of Elections on whether discrepancies are reconcilable and what an unresolved discrepancy actually means, the court is left to its own devices to interpret statutory requirements. It is the court's position that the term "reconcilable" means that there is a clear and logical explanation/reason why the "discrepancy" occurred. As previously noted, the Republicans have propounded four possible reasons in each case where additional ballots were found inside a machine. The Democrats automatically assumed that it was an irreconcilable discrepancy.
The court concludes that the discrepancies noted by the Democratic Commissioner that fall within the Seventh Senatorial District are resolvable.
The Standard for Accuracy
During the course of the court session on December 4, 2010, counsel for Craig M. Johnson argued that the New York State regulation, which mandates a margin of error of less than .1% was exceeded to the extent that the rate of error for the seven audited machines was .12%. While he took the position that this required a hand count of the paper ballots, in reality, an error rate in excess of .1% would lead to an additional 5% audit. There was no specific mathematical calculation by which the error rate was derived, but one can, by use of extrapolation, duplicate the results which were claimed.
The total number of ballots in the Seventh Senatorial District was 85,433. There were a total of 249 voting machines, producing an average number of votes per machine of 343. Multiplying this number by the seven audited machines produces a total vote count of 2,402. Dividing that number into the three allegedly unreconciled errors produces an error rate of .12%, the number propounded by counsel.
This is merely the product of speculation. There has been no evidence produced as to the precise number of ballots in each of the seven machines. The only audit reports which mention the number of ballots are those in which a discrepancy was found, and two of these are substantially greater than the propounded average of 343. Simply stated, there has been no objective evidence that the margin of error in the seven machines considered in the 3% audit was equal to or greater than .1%.
Analysis Pursuant to 9 NYCRR 6210.18 (h )
Putting aside the court's conclusion that there are reconcilable discrepancies, the court believes it is compelled to examine the facts of this case pursuant to 9 NYCRR 6210.18 (h), which has also been requested by counsel for Johnson:
"(h) The standards set forth in Subdivisions (a)-(g) above are not intended to describe the only circumstances for a partial or full manual count of the voter verifiable paper audit record, but instead are designed to set a uniform statewide standard under which such hand counts must be performed. The county boards of elections, as well as the courts, retain the authority to order manual counts of those records in whole or in part under such other and additional circumstances as they deem warranted. In doing so, they should take into consideration: 1) whether the discrepancies were exclusively or predominantly found on one type of voting machine or system; 2) the size of the discrepancies; 3) the number of discrepancies; 4) the percentage of machines or systems with discrepancies; 5) the number and distribution of unusable voter-verified paper audit trail records as described in Section J below; 6) the number of cancellations recorded on the voter-verified paper audit trail records reported pursuant to Subdivision (c)(1) herein; and 7) whether, when projected to a full audit, the discrepancies detected (no matter how small) might alter the outcome of the contest, question or proposal result."
The court had concluded that the discrepancies noted by the Democratic Commissioner that fall within the Seventh Senatorial District are resolvable. However, reviewing these factors and accepting the Democratic Commissioner's finding for this purpose:
• (h) (1) — Not applicable. We only have one type of machine.
• (h) (2) — Size of discrepancies. Each discrepancy is very small or nonexistent within each machine audited.
• (h) (3) — Number of discrepancies. This is also a very small number within the Seventh Senatorial District.
• (h) (4) — Percentage of machines with discrepancies. The Democratic Commissioner states that 34% of the County's audit of 32 machines had unresolved discrepancies.
• (h) (5) — Number and distribution of unusable voter-verified paper audit trail records as described in 9 NYCRR 6210.18 (j). None related to the court.
• (h) (6) — Number of cancellations recorded on voter-verified paper audit trail records reported pursuant to 9 NYCRR 6210.18 (c) (1). None related to the court.
• (h) (7) — Whether, when projected to a full audit, the discrepancies detected (no matter how small) might alter the outcome of the contest.
• Only subdivision (h) (7) now becomes important for this process.
Using the seven machines audited by the Board and a maximum error of two additional Johnson votes in two sevenths of the machines (two on one machine; one on another, but rounded up to two each) in the Seventh Senatorial District would result in 142 more Johnson votes.
249 machines x 2/7 = 71 machines x 2 votes each.
If we add two more machines that were audited pursuant to an earlier ruling of the court we have nine machines. The court finds one of those machines would have had no deficiency. The other would have had a clearly resolvable "deficiency." (Machine stopped working — took voter's ballot but did not tally it.)
We then have two ninths of the machines in the Seventh Senatorial District adding votes to the Johnson columns. Two ninths of the 249 machines is 55 machines. At two votes per machine, Mr. Johnson would gain 110 votes, leaving Mr. Martins with a 341-vote lead. This would not include any additional votes for Mr. Martins which could be extrapolated from the audit.
If we considered machine No. 805, where the ballot totals were the same but the vote tabulation differed and where Martins received an additional vote, one could extrapolate this out to add 35 votes to Martins.
If the court were to take the results of the 3% countywide audit, which concluded, pursuant to the Democratic Commissioner's numbers, that 11 machines had irreconcilable discrepancies out of 32 audited, which would be equivalent to 84 machines in the Seventh Senatorial District, and we hypothetically assigned two votes to each of those machines for Johnson, then Martins' lead would be reduced by 168 votes (not counting any additional votes for Martins pursuant to the audit), leaving Martins with a 283-vote lead.
The court, therefore, finds there is an insufficient basis to order a manual audit of the voter-verified paper audit trail records in the Seventh Senatorial District.
The court hereby orders the Board of Elections to certify Jack M. Martins as the winner of the race for State Senate in the Seventh Senatorial District by 451 votes, and that he is declared duly elected to that public office; and it is further ordered that the Nassau County Board of Elections is hereby directed to forthwith take such other and further ministerial actions as would be necessary and appropriate in furtherance of this order, including the execution of such documents as will attest to the certification of Jack M. Martins as the winner of the election for the Seventh Senatorial District in the November 2, 2010 general election; and it is further ordered that all other requests for relief contained within the instant petition ( Johnson v Martins and Nassau County Bd. of Elections, index No. 20821/2010) are denied in all respects and dismissed; and it is further ordered that a copy of this order shall be served on attorneys for all parties to these consolidated actions, as well as upon the counsels for the Democratic and Republican Commissioners of the Nassau County Board of Elections; it is further ordered that this order is stayed, until December 14, 2010, pending Johnson's appeal to the Appellate Division, Second Department. Any further application for a stay shall be made to the Appellate Division.