Minnesota Public Radio has a page showing examples of contested ballots, with a poll under each one, so you can see how your call on each vote compares to what MPR readers in general think. It’s surprisingly entertaining.
Nate Silver has a good discussion of the ongoing Minnesota recount process, which is still neck and neck between Franken and Coleman.
When capitalized, "Sie" is the formal way to address adults of either gender in polite German. I majored in the…
If one assumes that a person is making their money, including enough money to have no incentive to "work", by…
Ob sie wirklich Deutsch spricht. Normalweise, kommt die US Soldaten nicht ins Stadt und spricht nicht so oft mit denn…
As for being a Nazi, ich spreche fliessend deutsch und wohnte in Deutschland fuer drei Jahre mit der americanishe Luftwaffe…
And of course this "incentive" only applies to poor people: Rich people who already have enough to live off of…
Why don’t people just ask for a new ballot when they screw up? I like the “arrow” ballot best: it’s the most clearly desperate argument. I have to say that both campaigns have come up with some bizarre arguments, though.
I don’t know, I think the “underline” argument might be the most desperate. It looked to me more like a cross-out than an “underline.” The “arrow” is pretty desperate too though.
It looks like Minnesota needs a rule about how to count ballots where two ovals have marks in them, but one mark is bigger than the other. My personal feeling is that a dot or a line shouldn’t be counted if another oval is filled in, but a mark that is thicker than the width of the pen (i.e. the voter “scribbled”) should be counted the same as filling in the oval. But my personal feeling shouldn’t be used to determine the results of an election; this kind of thing should really be decided in advance.
I like Los Angeles County’s Inkavote system. It’s similar to the old punchcards, but the voter uses a tiny stamp pad instead of a punch needle. It’s very difficult to make an accidental stray mark on the ballot, and it’s very difficult to make an ambiguous mark in an oval without filling the oval totally in. After voting, the ballot is run through a scanner which returns the ballot to the voter if there are any overvotes. (It would be nice if the scanner showed the results of the vote to the voter on a screen, but that’s the only improvement I could think to make.)
I found the “underlined/crossed out” ballot to be completely inscrutable. Half was crossed out, half was underlined, and both bubbles filled completely in – I’d say you just can’t decipher voter intent from that one. Basically all of the rest of them are determinable. That one – no way.
Yeah, the underline one was rather pathetic… as was the “eraser marks mean he wanted to vote for the erased vote”.
As was the arrow.
Seriously, though, I’m surprised so many people have trouble with a bubble sheet, enough that they might turn the tide in an election.
Reminds me of a story my dad tells about how, in his younger and more politically active days, he attended a count. One ballot was completely clean, apart from an epithet written beside the name of one particular candidate. In accordance with the rules, she was shown this ballot and asked whether she wanted to accept it as a vote. She declined, on the grounds that the voter’s intent was fairly clearly not to vote for her.
Why don’t they just admit that the election was a tie and do a run off between the two top vote getters?
I agree with ADS that the underline/crossout was indecipherable: could have been intended to indicate one way, could have been intended the other.
The lizard people ballot was in no way an overvote: the voter wrote in “lizard people” but did not fill in the oval, indicating that he or she was not voting for the LP in that race.
Yep, that was my opinion as well. Even if he doesn’t win the election, Al Franken definitely defeated the lizard people.
Nope. Check out the relevant statute. Subd. 4 indicates that if the voter writes in a name in the proper place, then the ballot shall be counted for that individual, whether or not there is a mark beside it.
I suppose one could make the argument that Subd. 4 implies that a write-in should be counted even if other bubbles are filled in, because Subd. 4 says “If…then a vote shall be counted for that individual”, and nowhere else does it say that a filled in bubble indicates that “a vote shall be counted for that individual”…but that’s a pretty weak argument.
Really, they should have had a programmer write the vote-interpretation algorithm for them. It’s our job to avoid mistakes like that. ;)
Subd. 4 indicates that if the voter writes in a name in the proper place, then the ballot shall be counted for that individual, whether or not there is a mark beside it.
I see your point, but the name of an individual wasn’t written in. A species was. I’m not sure you can vote for an entire species. So it’d have to be declared invalid I suppose.
Dianne: It depends on the nature of the species. If the Lizard People have a hive mind like the Borg or the Buggers from the Ender novels, then I think the whole species can be a candidate.
Like most folks, I thought the one with the “underline” was the only really ambiguous one in terms of voter intent, though if I couldn’t throw it out I’d give it to Coleman.
Also, as a teacher, I’m reconsidering my opposition to scantron tests — they might just be an important civic skill-builder.
I wondered if the voter was trying to indicate that Franken is one of the lizard people, so was worthy of a vote whereas the candidates in all the other races were not.
Also, I continue to love 538. A commenter there pointed out that if Franken wins, Coleman will be the only person in the world who has lost to both Al Franken and Jesse Ventura at the same thing, which is quite impressive when you think about it…
This discussion reminds me of my favorite bit of dialog from the animated show “Harvey Birdman, attorney at law.” Lizardman, a monster/villain character, has taken a stand at a custody trial.
I doubt the voter was aware of such an obscure statue, and their intent is clear. They are voting for Franken, but are protesting the absence of lizard people on the ballot.
It is true that writing lizard people in the write-in spot counts as a vote, even without a mark, but it is also true that if the intent of the voter can be determined, then the the vote should be counted. Given that the LP voter voted for the LP by both writing Lizard People and filling in the bubble in the presidential race, but did not fill in the bubble for Lizard People in the Senate race, but did vote for Franken, the intent of the voter seems to me to clearly be to vote for Franken.
When I looked at the MPR site last night, the overwhelming majority (90%+) of voters in the online poll agreed with my interpretation (which I would also apply to the Coleman-Bach men over vote ballot), but in both cases, the online poll result has now shifted towards a 60-30 agreement with me (the Coleman-Bach Men ballot poll result is actually 30-60 against my interpretation, although the Coleman Bach Men ballot is more ambiguous, since there is less supporting evidence (at least in that photo) of how the voter was using “Bach Men” elsewhere on the ballot).
I’ll be interested to see what the election judges decide in the end on these ballots. I hope MPR does an update with the judges answers.
Perhaps it’s meant as a commentary on Franken? If I’ve got this straight – a vote for the British Royals/Crack-merchant Zionists?
SiF, I’m not sure if you’re being silly or not. So I’m going to pretend you’re not, and you can mock me if you are.
I don’t think the voter’s knowledge of the statute is relevant; the statute is intended to guide the judge in interpreting the voter’s intent. (Subd. 1 says “In determining intent, the principles contained in this section apply.”) The statute clarifies that the only reason for someone to write in a name is if they intend to cast a vote for that person. Sure, we can find other wild interpretations (they were testing their pen, they were protesting something, they have a mental illness that causes them to write “Lizard People” on everything they see), but the statute clarifies that the only reasonable interpretation is that they were attempting to cast a vote for the person they wrote in.
The case is complicated by the filled in bubble by Franken’s name, and you interpret that the intent was to vote for Franken but to protest the absence of Lizard People. But it could also be the case that the voter accidentally cast the vote for Franken, then changed his mind and did a write-in.
How would your interpretation of the ballot change if the bubble beside Lizard People were filled in?
I’m not SiF, but I agree with SiF, and I’m not kidding (although I agree that the voter’s understanding of the statute is not the most important thing, I do think it matters somewhat, since if the voter doesn’t fully understand the rules for write-ins, the voter may not have understood that it is impossible under subd 4 to write in a candidate and then not vote for that candidate).
While a name written in, without a mark next to the name, is sufficient to be counted as a vote, I don’t think that the law has to be read to say that a name written in is necessarily counted as a vote. If other aspects of how the voter filled out their ballot give a clear intent of the voter to vote for a different candidate, that intent should be taken as their vote. In particular, if the manner of the filling out the ballot shows that the voter believed that writing in a name and voting for that name were two distinct actions, then we should not count the written-in names as causing over-votes, and thereby depriving the voter of their vote.
If the only write-in on the ballot was “Lizard People” for Senator, then we would have to assume that the voter intended to vote for Lizard People, and if they also filled in a bubble for Franken, then they over-voted. If they had added “Lizard People” as the write-in for all offices, and never filled in the bubble for Lizard People, and voted normally for non-write-in candidates, then their ballot should be counted as having voted both for Lizard People and for other candidates, and therefore over-voted in all the races (since there is no way to judge whether they intentionally over-voted, or if they merely wanted to explicitly not vote for Lizard People in each race).
However, on this ballot, where they filled in Lizard People everywhere, and then either voted for Lizard People (as in the presidential race) or voted for another candidate (Franken) instead of filling in the bubble for Lizard People, I think it is clear that they did not intend writing in Lizard People in the Senate race to be interpreted as a vote for Lizard People. Since their intent is determinable, their vote should be counted as they intended, notwithstanding that writing in a name is normally counted as voting for that candidate.
And I think it’s more likely that they forgot to fill in the bubble next to LP. So do we go with what I think, or what you think? This is why we have the statute. And this is why the statute really should explicitly address (rather than leaving us to reason out) what I’m sure is, in a state with umpity-million voters, a fairly common occurrance.
Forgetting to fill in a bubble is equivalent to not voting, so your counter argument holds no weight to me. The question is whether the voter intended for simply writing “Lizard People” to be counted as a vote for “Lizard People”. subd. 4 requires us to assume that that is a voter’s intent when they write in a name, all else being equal, but I don’t think it requires us (well, the election judges) to ignore the clear intent of the voter if the voter’s intent was not to have that treated as a vote.
Consider the extreme case, in which a voter writes at the top of their ballot: “I have filled in the name “Lizard People” in the write-in slot in each race, because I wish to explicitly have the choice of voting for or against “Lizard People.” Unless I have filled in the bubble next to “Lizard People”, and contrary to subd 4 of 204C.22 of the 2008 Minnesota Statutes, please do not consider the mere presence of the words “Lizard People” in the write-in slot to constitute a vote for “Lizard People”. For instance, in the Presidential race, I have voted for “Lizard People”, while in the Senatorial race, I have voted for Al Franken, and not for Lizard People.”
In that case, I assume that you would agree that the subd 1 requirement that we count a vote where the clear intent of the voter can be determined overrides the subd 4 requirement that a write-in counts as a vote, even if the name is not marked in any way. If that is the case, then the only question is whether the pattern of behavior seen on the “Lizard People” ballot shows clear intent, not whether subd 4 requires us to treat the senate vote as an over vote. Do you agree?
subd 1 says that:
I take technical errors to include the subsequent subsections interpreted in a literal manner, where that literal interpretation would cause the ballot to be rejected even though the voter’s intent is determinable. The subsequent subsections are guidance for determining intent, but it is the principles expressed by the subsequent sub-sections that must be followed, not the literal meaning of the subsections.
I don’t think that it would be possible to write a law that fully handles the “Lizard Men” ballot, other than one that requires us to reject both the actual “Lizard Men” ballot and the imaginary “Lizard Men” ballot I described above. A law that rejected the legitimate vote in my example would be a worse law than a law which depends upon the majority decision of 5 thoughtful and careful judges as to what constitutes a determinable intent. Whichever way they decide, I would rather have judges consider the arguments for and against the “Lizard Men” ballot than have a law that tried to fully handle all corner cases.
What specific language would you propose that would indisputably decide both the actual “Lizard Men” ballot and my example of an alternate “Lizard Men” ballot?
@Bjartmarr
Since Lizard People is obviously a joke, I tend to think that voting for Franken is clear. Now, if the write-in candidate had been a popular write-in candidate, then it would be slightly more ambiguous.
Obviously, the law may trump any ambiguous concept of “intent”. And because my interpretation of intent is highly subjective, I can see why they might make the law override intent.
But here’s the thing: the obvious function of the ballot is a bubble sheet – you fill the bubble to register your vote. The bubble next to the write in spot logically demonstrates one’s intent to vote for the write-in candidate – otherwise, that bubble would be redundant. That mentality is clearly in the mind of the voter since they wrote-in and bubbled Lizard People in other spots.
Now, I could see a case for ambiguity if a listed candidate was bubbled and a popular write-in candidate was written in but not bubbled… particularly if there were other write-ins in other spots on the ballots that were similarly not-bubbled, but without any other “double vote”, demonstrating the voter’s belief that writing the candidate’s name counts as a vote. However, even if a popular write-in candidate’s name is written in but another bubble is checked, I’d be tempted to interpret that as a protest of their absence on the ballot, not a sincere attempt to vote for the write-in candidate.
In general, I’d think that the bubble takes priority over the write-in… but particularly so when the write-in is an obvious joke or delusion.
So does clear intent of the voter, if it can be determined, override all else or not? And how much leeway is allowed to the judges in determining intent? Could a judge say, “This voter filled in the circle for Franken and made no other marks on the ballot but the psychic vibrations I’m receiving from it suggest that the voter intended to vote for Coleman”? Or, in a less extreme example, “That’s obviously an underline the vote goes to Franken/That’s obviously a cross out, the vote goes to Coleman”?
Oh, I agree that you have to follow the rule of law, for just that reason.
However, the law is wrong, and the voter’s intent is obvious.
But since the law explicitly calls for privileging the intent of the voter over rigid adherence to rules, SiF, refusing to count this particular vote would be a case of failing “to follow the rule of law.”
Dianne, the process involves representatives of both major parties. I hope they have the possibility of appealing obviously ludicrous decisions, so that a “psychic vibes” argument wouldn’t succeed. But I really don’t know how the process works, at that level. Maybe Jeff will read and comment — I’m sure he’s been following this story 100x more closely than me.
But the problem we’re having is that I don’t think the voter’s intent is as clear as you make it out to be. So arguments of “but since the intent is clear…” are irrelevant, as far as I’m concerned.
Think about it. You guys are saying “the intent is obvious”. But a significant number of people (including the county auditor) believe not only that the intent is not obvious, but that your interpretation of it is likely wrong. If a significant number of reasonable people are disagreeing about something, isn’t that pretty good evidence that the conclusion is not obvious?
And I don’t agree that forgetting to fill in the bubble is equivalent to not voting. I think that forgetting to fill in the bubble is a technical error equivalent to forgetting to cross a T. SiF’s argument that the voter filled in bubbles for write-ins for other races but not the Governor’s and therefore meant to treat the Governor’s race differently seems equivalent to arguing that a race with a write-in without a crossed T is invalid because other write-ins on the same ballot have crossed T’s.
I don’t think you guys are being unreasonable by interpreting the ballot as a vote for Franken. But I don’t think I’m being unreasonable by interpreting it as an overvote, either. I do think that the intent is not clear, and absent a clear intent we have to follow the mechanical rules laid out in the law.
Finally, I think you all are writing from the viewpoint of reasonable people who take voting seriously. I don’t think any of you would scribble “Lizard People” all over your ballot, therefore I think your attempt to determine this voter’s intent from the viewpoint of a reasonable, mature person is likely to be incorrect.
The point is that he filled a different bubble, not just that he failed to fill the given bubble. To my mind, the bubble gets priority unless there is something specifically indicating an intent to negate the bubble vote. The text box, which is used for specifying a write-in candidate (not voting for a write-in candidate) only becomes relevant if there is no non-negated bubble selected. The bubble and the box has different meaning – the bubble is “who are you voting for”, and the box is “if you are voting for a write in candidate, who would that be?”.
To me, this is the literal meaning of the ballot. His (or her) ability to do this correctly in other spots on the ballot indicates that they grasp the literal meaning of the ballot. If there was any indication elsewhere that they did not, I’d be more willing to believe that the statement in this vote is ambiguous.
His answer is “Franken, not a write-in candidate” and his response to the question on the line is “Lizard People”. So he’s saying, “I am voting for Al Franken, not a write-in candidate, and if I am voting for a write-in candidate, then I am voting for Lizard People”… but he is not voting for a write-in candidate. Those statements are not contradictory. The latter part of the statement is totally irrelevant, but does not contradict the first.
Damn, I think I just made myself an advocate for voting machines. On a computer, you can enforce these concepts. Too bad voting machines are a trainwreck of their own.
Now, if the ballot had no filled bubble, I’d be more inclined to believe in the vote for Lizard People. After all, if the user has chosen to use only one element of the interface then you have to take what you can from that one… or if the user had scribbled-out or otherwise obviously negated their vote for Franken, you’d have to take what you can get from the remaining interface. But, to my eyes, the bubble clearly takes priority over the write-in box in cases where the bubble is unambiguous – particularly when the write-in box IS ambiguous (Lizard People is not a serious vote).
To me, the voter is speaking clearly. They are saying something silly and wasteful, but clear. A ballot with confused mind-changes or scribbles like the “underline/strikeout” ballot is not speaking clearly – it says nothing clear. If I saw a write-in that differed from the filled cell on their ballot I’d be more inclined to treat it as an error of communication.
But the Lizard People ballot has a clear meaning. The whole ballot has a clear meaning. It is a silly meaning, but a clear one.
Look at it this way – if you heard a person mumbling a lengthy stream of unintelligible nonsense and then they said “I am voting for Al Franken, but if I were voting for a write in candidate, it would be Lizard People” then you’d consider it to be one more confused part of the mumbling.
However, if you heard a person clearly list their intentions “I am voting for Lizard People for president, Lizard People for governorship, and Al Franken for Senate, but if I were voting for a write-in candidate for the senate it would be the Lizard People” then the meaning is clear – they are voting for Al Franken for senate.
Maybe. But on the other hand, I recently filled in a bunch of write-in votes with alliterative names of superhero secret identities.
(But to be fair, I only did that on the unopposed races, so I guess you’re right.)
Do we know what the county auditor decided about this vote? It may have gotten no more thought than “put it aside to consider it later, if the race turns out to be close enough to make such considerations relevant.” And I’m not sure that an online poll can be representative of what people in general believe.
Not that I’m saying you’re being unreasonable to disagree; just that your evidence that there are a large number of reasonable people who disagree on this, seems questionable to me.
SiF, I understand your reasoning. I think your interpretation is one reasonable interpretation out of many, and I disagree with it. (You don’t have to explain it yet again. ;)
Also, I don’t think we need to go to voting machines to avoid this kind of problem; in-poll ballot scanning with scanners that are programmed to reject ballots that are in any way ambiguous would solve most of the issues on the MPR page. (In this case, it would have counted an overvote and returned the ballot to the voter for a re-do.)
Amp, the paragraph above the ballot gives the auditor’s ruling. (You’re right that he probably didn’t give it much thought. On the other hand, if the scanner just looks at bubbles then this looks like a vote for Franken, so he may not have seen it until the recount.) And while the online poll isn’t scientifically rigorous, given that it’s all we’ve got I’m actually comfortable that it rises to the standard necessary for credibility on, you know, some guy’s blog. :)
They have added a second day’s worth of ballots. I think the pencil one is now the most pathetic challenge.
Bjartmarr,
Given the MPR poll and the lack of consensus here, it is likely that the 5 election judges will not have a majority who agree that this ballot is a clear vote for Franken. If they they do have a majority that agrees that this vote is a clear vote for Franken, would you consider that a gross miscarriage of justice?
Is there any system that you can imagine that would correctly handle my example of a version of the “Lizard People” ballot that would also handle the actual “Lizard People” ballot in a manner that you consider correct, that would not involve resorting, as the current system does, to the judgment of a panel of judges? I can not think of a way to design a system that does not depend on human judgment to determine intent that would reliably determine intent where intent could be determined by human judgment, so I think the current system is the best.
If the election actually comes down to a single ballot, then the exact interpretation of intent in the tiny number of ballots that are truly just at the border line of determinible intent will decide the election, but that would be true no matter where the law sets the border line of determinable intent.
The solution of having any ballot that doesn’t meet the machine’s criteria rejected when the user submits the ballot, forcing the user to redo their ballot, is a poor solution for users who are not able to fill out the ballot to the machine’s satisfaction. For instance, a voter who finds it extremely difficult to accurately fill in a bubble completely, with no stray marks, may still be able to make check marks in the bubble, or may be able to circle the candidate’s name.
Charles: No, I wouldn’t consider either ruling a gross miscarriage of justice.
The system I described would correctly handle your version of the scenario, by spitting the ballot back at the voter with a “Hey, you made way too many stray marks on this ballot, it is unlikely to be scanned correctly, you get a redo. Also, you over-voted on every single race. Please talk to a poll worker, who will instruct you in proper use of the voting apparatus.”
I think it’s important for the system to accomodate any voter who makes a reasonable attempt to clearly cast his vote. I think it’s extremely important than when such a voter leaves the poll, his ballot accurately reflects the people who he thinks he voted for. I don’t think it’s important for the process to accomodate other sorts of tomfoolery such as the hypothetical scenario you describe, or the actual LP scenario, other than to say, “Hey, that’s not going to work, you get a redo”. The current system fails at the part that I consider important, and succeeds…sort of…at the part that I consider unimportant.
Finally, I said that the scanners should reject any ballot that is ambiguous. Ballots such as those that you describe, with exactly one mostly-but-not-completely filled bubble, or a little mark outside the lines but mostly inside the bubble, are not ambiguous. And I find it hard to think of a disability that would allow someone to unambiguously circle a name, but not allow them to make a blot in a bubble. Let them use a thick pen, fer chrissakes, if they can’t scribble.
The point here is to reject and correct the relatively large number of ballots which are clearly ambiguous or over-votes (or maybe even under-votes…though sometimes those are legit). You’ll always be able to come up with a crazy scenario that will break any system we can come up with…including the 5-human-judges system.
(An even better system would report back to the voter, on a screen, the results of the ballot as scanned. But that would cost more money and you’d have to make sure the voter had privacy at the scanning station.)
I agree that a system that clearly informs the voter if there is a problem with their ballot, and also informs them of what votes they are casting, is preferable to one that doesn’t (without informing the voter of who they are voting for, the technology is very likely to simply treat incorrectly marked ballots as undervotes, since not voting in a race is not a problem with the ballot, or else it will treat anyone who has difficulty not leaving stray marks on the ballot as repeatedly producing invalid ballots), but that is a matter of technology, not of law. For whatever gets past the technology, I think the 5 judge panel is the best solution of last resort(and a voter who gives up in frustration should still have as many of their intended votes counted as possible, including whichever votes the machine was not able to understand if a human is still able to understand intent, so the voters still need to be able to walk away from the voting machine and have their votes count, even if the machine tells them there are problems with their ballot). The number of bizarro ballots is tiny, and only rarely ends up being important (and, as I said before, the chance that the Lizard Man ballot decides the election is still very small) but it is still a good idea to have a final human eye to check that the machines are reading each ballot correctly, and a final small group decision on any cases that are too weird to be handled by a lone person acting on a clear set of rules. The Minnesota system looks pretty good to me.
I’m oddly fascinated by the thought of the lizard people voter. This is a person who, much like me, makes a little joke like this basically for emself, although also for the benefit of a hypothetical poll worker who might see it.
And now this little joke has been seen by thousands and thousands of people.
The lizard people voter finds this situation very surreal, I suspect.
I certainly hope that the lizard people voter is aware that their ballot has been made public and become a matter of general discussion, because I suspect that they’d find that enormously entertaining.
(On a purely personal level, it is virtually impossible for me not to imagine the lizard people voter as an eerie Minnesotan double of Jake Squid.)
If it is an eerie Minnesotan double of me, I can assure you that he is aware of the publicity that “Lizard People” is getting and will cherish the memory for the rest of his days.