Ginsberg dings Bell, but did he dong it?

I need to briefly revisit Bell v. Gannaway, the 1975 Minnesota Supreme Court case I posted about yesterday because of some new information that came to light in the Contest trial yesterday and some comments that Coleman lawyer Ben Ginsberg made about the case after the testimony. For those who missed that post, a quick recap:

The Bell case has been a thorn in the side of Team Coleman, in part because the Supreme Court ruled that once an absentee ballot has been removed from its envelope and commingled with other ballots, it’s too late for one of the campaigns to challenge the acceptance of the ballot. (This was especially tough on the loser in the 1975 election between Bell and Gannaway, because Mr. Bell lost an election by one vote and one of the absentee ballots was shown at trial to be deeply flawed and should not have been acccepted.) This precedent, which has become a guiding Minnesota case on absentee ballot issues, is very inconvenient to Team Coleman because they want to make an issue out what they have argued are hundreds, maybe even thousands of absentee ballots that were counted but should have been rejected. End of recap. On to a couple of new points.

During the trial yesterday, Coleman lawyers asked Ramsey County Elections Chief Joe Mansky whether it would have been possible on Election Day last year for representatives of the Coleman campaign to have been present and to have raised challenges when absentee ballots were being removed from their envelopes.  I was not listening to this testimony, and MinnPost’s guy at the trial Jay Weiner was working on his memorable dead-man’s-ballot story. (If you missed that one, I highly recommend it.) But at the end of the trial day, Ginsberg told the assembled media that because of Mansky’s testimony the issues raised by the Bell case were not over. Said Ginsberg: “The notion that there is somehow no ability to go back and look at what has now been defined as illegally cast ballots on Election Day has been refuted today.”

Is that so?

I certainly agree with Ginsberg that it ain’t over till it’s over. The ThreeJudges have shown considerable interest in the Bell precedent, have cited it in at least two of their major orders so far and have challenged Team Coleman to explain why it should not be followed. But they haven’t, as I understand it, made a definitive ruling about the issue of absentee ballots that shouldn’t have been accepted but were.

But I do question whether Ginsberg made as much ground against his Bell problem as he claimed. For starters, he may have exaggerated what he got out of Mansky. I just spoke to Mansky and he said that while it was not possible for an individual campaign to challenge the acceptance of absentee ballots, it was possible a representative of each political party to do so. In other words, assuming that one of the jobs of the Republican poll watcher was to look out for the interests of the Coleman campaign, Coleman was represented.

As a practical matter, Mansky said, this is never done. The poll watchers at the precincts are mostly watchdogging the election-day registrations that are coming through. In most counties, the absentee ballots would already have been examined at a central location and ruled acceptable. At the precinct, after the polls closed, the election judges check to make sure the voter named on the envelope had not voted in person. If not, they open the envelope and run the ballot through the counting machine and, by the logic of Bell, it would be too late after that to challenge the ballot. But, technically if not practically, the Republican observers could have looked at the AB envelopes and challenged them before they were counted.

But even if the judges agree that the Coleman campaign had no Election Day opportunity to challlenge decisions to accept or reject absentee ballots, the Bell case is still a problem for Coleman.

A three-legged stool

Having read the case, I would say the ruling about post-election challenges to absentee ballots was a three-legged stool. One leg suggests that the Bell campaign had its chance to challenge the absentee ballots and missed it.  That’s the one about which we’ve just been talking.

Leg Two is the language of the Minnesota statute that describes when absentee ballots can be challenged. As cited by the court in Bell, it read:

“The voter and the ballots of any absent voter at any time before the ballots have been deposited in the ballot boxes are subject to a challenge by the judges or by any person who was not present at the time the voter procured the ballots, but not otherwise.”

The grammar could certainly have been clearer, but the Bell court construed that “but not otherwise” to mean that after the ballots are in the ballot boxes, they cannot be challenged. That statute has since been updated  and “but not otherwise” was moved within the sentence (see Subdivision 6), but it’s still in there with the same apparent meaning.

Leg Three is a serious practical problem in the Coleman-Franken case. In the Bell-Gannaway election, one woman cast a deeply flawed ballot that, depending on whether she voted for Bell or Gannaway, may have made the difference between a Gannaway win or an actual tie (which, under Minnesota law, would have led to a coin flip.) Maybe the woman, despite her right to the privacy of the vote, would have been willing to testify about whom she voted for. But in Franken-Coleman there would be hundreds, possibly thousands of such women and men.

It would be theoretically possible to go back over every absentee ballot envelope (we’re not talking about the roughly 12,000 that were rejected, but about the envelopes from the roughly 286,000 absentee ballots that were accepted and already counted) to see how many of them meet the high standards for acceptance that both the Bell case and the ThreeJudges in the Coleman contest case (who based theirs on Bell) have set. If Ginsberg is correct, hundreds or probably thousands of them would be deemed inadequate and should not have been counted. Remember, the ballots are out of the envelopes. It seems unimaginable that all of those voters will be asked for whom they voted.

What is to be done?

This brings us back to the idea of a re-do of the election. As a technical matter, Team Coleman has never asked the ThreeJudges to order a redo. The closest they have come was a letter from Coleman lawyer James Langdon that stated: “Some courts have held that when the number of illegal votes exceeds the margin between the candidates — and it cannot be determined for which candidate those illegal votes were cast, the most appropriate remedy is to set aside the election.” Ginsberg has been saying in the hallway for weeks that he isn’t telling the judges how to deal with the mess, but that they must deal with it.

I do assume that before the trial is over, Team Coleman will ask the court for specific actions.

If you would like to read an excellent summary of the four choices that the judges have for dealing with this situation, I highly commend this excellent piece by Sarah Cherry, an attorney who writes legal analyses for “Election Law @ Mortiz,” a blog maintained by the law school at Ohio State. Cherry identifies and discusses four options:

1. Apply the high standards to the shrinking pool of previously-rejected absentee ballots that the parties in the contest have brought forward with sufficient evidence to indicate that they should have been accepted. Count those ballots that should have been counted. But, guided by the Bell-Gannaway precedent, leave the invalid absentee ballots, that shouldn’t have been counted, in the count. The judges have said and written things, and asked questions, all of which implied that this was the direction in which they are leaning. But leaners count mostly in horseshoes.

2. Try to find out from whom the invalid absentee ballots were cast, and remove them from the count.

3. Use a precinct-by-precinct statistical analysis to subtract votes from both sides, based on how they did in the precinct and on the number of invalid votes in the precinct. Methods like this have used in rare cases, but I do not believe they have been used in Minnesota.

4. Invalidate the election.

You can also learn about all our free newsletter options.

Comments (3)

  1. Submitted by Bruce Johnson on 03/10/2009 - 05:19 pm.

    Eric — you write “it would be theoretically possible to go back over every absentee ballot…”. But my understanding is that it would not be possible to distinguish the absentees from other ballots reviewed in the recount except where there were duplicate ballots created and so marked.
    Much of the contention seems to be focused on the absentee ballot envelopes and issues of the proper registration of voters who voted absentee. For reasons of voter privacy, there should be no clear way to tie those envelope issues to specific ballots, and even if voters volunteered to identify whom they voted for, that testimony would not be reliable. Imagine a voter told that his vote was determined to be invalid and then asked to identify whom he voted for — a very severe test of honesty and integrity against personal interest that many might fail.

  2. Submitted by Eric Black on 03/10/2009 - 10:51 pm.

    You’re right, Bruce. I meant to say that it would be theoretically possible to go back over the ENVELOPES to reconsider whether they should have been counted. But, as you indicate, if the envelopes had problems, you couldn’t remove the ballot, which is now mixed in with all the others. I’ll fix the post as well. Thanks, eb

  3. Submitted by Alan Paulson on 03/11/2009 - 10:09 am.

    Hello Eric,

    Thank you for the pair of thoughtful articles on uncounting ballots.

    There is a practical problem I have not seen addressed, specifically: uncounting ballots is subject to fraud and could potentially make the problem worse, not better.

    A ballot, once counted, it is secret. In order to uncounted it, you need to figure out for whom it was cast.

    Sarah Cherry @ Mortiz suggests we ask the voter. However, if I voted for Coleman, and you asked, “who did you vote for?”, why would I tell you the truth? Knowing the court might subtract my vote, I could simply answer “Franken”. If the court then subtracted one from Franken, the net result would be +1 for Coleman (from my illegal vote) and -1 for Franken (the vote subtracted by the court), for a net gain for +2 for Coleman.

    Asking me who I voted for and then subtracting from that candidate actually made the problem worse. In general, voters cannot be considered reliable witness of how they voted, because they could help their candidate by not telling the truth.

    The other suggestion made by the Coleman campaign is to potentially subtract votes in proportion to the precinct results. This again could compound any fraud. Let’s say I’m a Coleman supporter, and I arrange with a sympathetic local election official to get 100 illegal votes added to Coleman’s total in a precinct that favors Franken by 60% to 40%. The illegal votes are detected, but no one knows who they were cast for. The court therefore applies proportional deduction, subtracting 60 votes from Franken and 40 from Coleman. The net result is that Coleman is up 60 (100 from my illegal votes minus the 40 subtracted b the court), while Franken is down by 60 (the 60 subtracted by the court), for a net gain of 120 votes Coleman. That’s right, after stuffing the ballot box with 100 illegal votes for Coleman, he actually ends up 120 votes ahead because the court deducted more votes from his opponent than from him. This is why proportional deduction should NEVER be applied in an election–because an unscrupulous candidate or his supporters could actually use it to their benefit.

    In summary, any of the proposed methods of subtracting votes are subject to fraud, and could make the problem worse, not better, and for that reason, the court should not attempt to uncount secret ballots.

    Best Regards,


Leave a Reply