Among Wisconsin's state and local election officials, only the county board of canvass has any responsibility to examine election results for signs of a miscount, and only the county board of canvass has any statutory authority to order correction (i.e., a hand count) of any results found to be miscounted. The singular importance of the county board of canvass for detection of any miscounts was highlighted at a recent GAB meeting.*
With observers present, the Boards of Canvass are more likely to do their job well. Without observers present, they are tempted to rubber-stamp whatever the municipalities send them and miss even obvious signs of miscount. (The infamous Medford miscount of 2004 could and should have been caught by the Taylor County Board of Canvass, but wasn't.)
However, to my knowledge, no one associated with the WGN Election Integrity Action Team has ever observed one of these public meetings, which occur sometime in the 14 days following Election Day. Nor are they officially observed by any other group I know of--League of Women Voters, ACLU, WCEP, or anyone else.
For the November 2014 elections, we're recruiting people we trust to observe the county board of canvass meetings in as many counties as we can. This webpage isn't linked from any other webpage; if you're reading this, you have been invited.
Please read the information below and let us know if you might be interested in observing the board of canvass meeting in your county, by commenting below; emailing Karen McKim at firstname.lastname@example.org; Julie Crego at email@example.com; sending Karen a Facebook message; or calling Karen at 608-212-5079.
As I write this (Saturday night), these meetings are scheduled, but we haven't found out exactly when and where yet. It will vary by county and could be as early as Thursday, Nov. 6, or as late as Monday, Nov. 17. I don't know how long the meetings take.
What is the Board of Canvass?
The Board of Canvass is the only clearly partisan actor/entity in Wisconsin's election-administration process. It consists of the county clerk (a partisan office), who pretty much runs things, and at least one Republican and one Democrat the clerk selects. The Board doesn't do much at any time other than in the week following each election. They spring into the public eye only when there's a contested recount. I've heard--and seen nothing to dispute--that the county clerk pretty much dictates to the other members of the board of canvass. I don't yet have any sense that most boards let partisanship dominate their actions. The office of county clerk is too administrative to attract ideologues; generally they are sincerly interested in fairness.
What does the Board of Canvass do?
Scott McDonell, Dane County clerk, told me that when he first took the office and reviewed Dane County's canvass procedures, he thought he'd stepped into some ancient superstitious ritual. The procedure had something to do with the members of the board of canvass reading numbers out loud to each other and then declaring the results correct. He couldn't see a useful thing about it. He told me he changed the procedures (to what, I don't know), but I strongly suspect other counties use similarly pointless rituals and call it a canvass.
The GAB has some decent instructions for the boards to use, but based on what we've learned from observing the pre-election voting machine tests, it's highly likely that not all clerks know they exist, and that even fewer actually follow them.
We've learned that having observers simply ask "How will this procedure on page 2 be handled," or "When will you be doing that step described on page 2?" will often cause election officials to follow important steps they would otherwise have skipped. Newcomers to observation are often quite astonished when they see how very many local election officials don't understand the purpose of required tasks, and how very often they mishandle even the procedures they try to perform.
One instruction in the GAB's Board of Canvass procedures is particularly significant for detecting possible voting-machine malfunction. It seems to have been added in an attempt to detect the next Medford-type miscount (and Humboldt County CA and Brooklyn NY, and almost certainly others that were not caught):
"For each reporting unit, check to see that there is not a large difference between the total number of voters and the number of votes cast for the office that would have garnered the largest number of votes at that election. For example, in a presidential election, the total number of voters should be compared to the total number of votes cast for the office of President of the United States. This is an important step in the process because a large drop off between these two numbers might signal a problem with the voting equipment. Additionally, the board of canvassers shall examine the Voter Statistics section of the Inspectors’ Statement to ensure no large discrepancies are evident. The number of votes cast should never exceed the number of voters."
Basically, what this says is that many voters might skip the less-important races down the ballot, but you should expect nearly everyone to cast a vote for governor. In Tuesday's election, if a precinct had 1,000 voters but only 950 votes for governor, it's a good sign that machine might have been malfunctioning, because it's a safe bet that one in every 20 voters didn't go to the polling place to not vote for governor.
If I had to bet money--we won't know until after we observe these meetings--I'd bet not one in ten county boards of canvass follows that instruction. GAB doesn't oversee their work in the slightest.
Here are the more detailed instructions for observing the board of canvass meetings.
Please let us know if you have any questions, and if you can help. You can comment below, or contact either Karen McKim or Julie Crego. Our contact information again: Karen McKim at firstname.lastname@example.org; Julie Crego at email@example.com; Karen is on Facebook under her real name; or call Karen at 608-212-5079.
* At the GAB meeting on Tuesday, Oct. 28, Director Kevin Kennedy was trying to convince the Board not to change the policy that voting-machine audits be delayed until after the election results have been declared final. In doing so, he repeatedly referenced the county board of canvass, and tried to convince GAB Board members that the county boards of canvass were so dedicated and thorough that they were sure to notice any electronic miscounts even without voting-machine audits.