Update: 1/11/2008 I have scanned in the actual report of the New Canaan recount sent to the Secretary of the State’s Office. Read the full report <here> the critical information provided by the Head Moderator is:
Totals were the same as the original count or close with the exception of Paul Foley (-31 votes) [MINUS 31]. An assistant registrar, a registrar, and I independently recounted (again) all tally sheets and totals for the recanvass and the tapes’ data from the tabulator print-outs on the moderator’s report for the 11/06/07 returns finding no error. The cause of the variance of Foley versus the other six candidates is unknown.
In other words the machine counted 31 votes more than the paper indicated. Perhaps we can say a lot of things but we cannot say that “the machines counted perfectly or accurately”. We could say “the machines counted every vote (and then some)”.
Original Post 1/4/2008:
After the November election there were some forty recounts of close races held across Connecticut. Two press reports indicated significant discrepancies in the counts: one in Greenwich and the other in New Canaan. At CTVotersCount we pointed to these as examples of good reasons to have recounts in close races and to manually count ballots in recounts and audits:
For those who question the value of manual counting in audits and recounts, here is an example from Greenwich, where despite pre-election testing an error apparently slipped through.
[New Canaan] Recount Reaffirm’s Wisdom of Recounts in Close Races
On the other hand there were more questions than answers in the news reports. We suggested to the Secretary of the State’s Office that investigations might be in order. We subsequently asked for and received copies of the reports of all the recounts sent to the Secretary of the State’s Office. Lesley Mara, Deputy Secretary of the State and the entire Office of the Secretary of the State have been very responsive to our requests for information. In December we received an explanation of the situation in Greenwich and on Wednesday a response from the Deputy Mara concerning New Canaan.
There is some good news and, in our opinion, some not so good news:
The good news is that in Greenwich it was apparently a matter of two instances of misunderstanding/miscommunication, one on election night between election officials, and another between election officials, a candidate, and the newspaper. The newspaper report that the cause of the change in the totals was the scanner counting write-in ballots for a candidate was incorrect. The write-in ballots were counted for a candidate, but it was a human error of adding the write-in ballots into the count for a particular candidate that caused the error and the subsequent correction. Good news for the integrity of the optical scanner and also a confirmation of the wisdom of recounts in close races. This particular case the confirms the wisdom of empowering moderators’ to call for recounts in the case of concerns brought to their attention.
The not so good news is that the New Canaan situation apparently will never be resolved. The recount report showed the discrepancy between the original vote and the recount to be 31 votes. That is a significant difference. Unfortunately the law for recanvassing does not call for an audit and in the words of the Deputy Secretary of the State “they had a recount, not an audit so there is no statutory or regulatory framework that would call for the kind of investigation you suggest”.
The bottom line is that we have a significant discrepancy which will never be investigated, and can never be investigated, in a race counted presumably in a very careful manner with two interested candidates involved. The memory card, optical scanner, and ballots are now beyond the chain of custody as well. Sadly, we have learned that the Secretary of the State’s Office believes that they and UConn are powerless to investigate.
We have heard the often repeated statement this fall “if there is ever a question the paper can be counted”. Perhaps that is enough for a candidate when there is a question about a particular race, but clearly more is needed when a recount raises a question about the integrity of a voting machine.
The law needs to change. An independent audit is necessary — an audit by an independent entity that is motivated, empowered, and required to investigate whenever and wherever obvious questions and concerns exist. Additionally this clearly demonstrates the folly in exempting districts with recounts from audits, when an audit would have at least provided the Secretary of the State with the power to investigate and determine the cause of a discrepancy.
We are left with with legitimate questions unanswered about integrity of the optical scanner and memory card in New Canaan. We are left with significant concerns for existing laws to provide the follow-up necessary to generate confidence in our election system. The lack of an investigation also provides a safe haven for others to claim that no problem has ever been found or proven.
