January 13, 2009
POLITICS: Recount Limbo
My State Senator is still in limbo due to Democratic recounts and court challenges to his election. While we hear a lot of complaints these days about needing to have one president at a time, at least we have one; Frank Padavan's constituents don't entirely have a State Senator at all, nor do Norm Coleman's constituents have a U.S. Senator. And it's January 13.
You know, I haven't followed all the twists and turns of the battle over Padavan's seat, but one thing I have concluded from watching it, and the Al Franken, Christine Gregoire and Al Gore efforts to overturn Election Day results, is that we really do not have any way as a system to deal with these kinds of challenges in a way that gives the supporters of the losing candidate - especially a candidate who was ahead on Election Day - even the slightest bit of confidence that counting decisions made after the election, under the auspices of lawyers and partisans, are at all fair and honest. Which is, as I have been saying for 8 years now, the real point of the Supreme Court's Bush v. Gore opinion. There are, to be sure, opportunities for ballot fraud and other shenanigans before and during an election, and the system has to provide some remedy for those, at least in provably serious cases. But before 2000 there was an ethos - not always respected, by any means, but to which politicians (most famously Nixon in 1960) at least needed to pay tribute - that the loser of an election did not open a second scorched-earth front designed at refighting every ballot that could conceivably be quibbled over, and that you needed a really serious reason to try to overturn the Election Day Count. That is Al Gore's lasting legacy to our democracy, and it's a deeply malignant one.
Machines, of course, can make mistakes, and if we had confidence that the counts produced by heavily lawyered recounting processes were really a more accurate and precise count, it would be worth the cost in money, time, disruption of transitions and hard feelings about democracy to review them. But we have no such confidence; we are, as a society, simply throwing resources at a series of additional counts that give us no reason to think they are any more accurate, and many reasons to think they are much less impartial than a machine count. A machine count is done behind your basic Rawlsian veil of ignorance: both sides may know they stand a chance of getting a raw deal in a close race, but they know it's basically an even chance. When lawyers and partisan vote-counters get involved, all that goes out the window, and the race to amass a superior quantity of umbrage is on.
Or think about it this way: up until the day of an election, the forces of partisanship have limited resources, less than perfect information about which elections will be closest, and face the reality of having to spread those resources over races for different offices that may be close in different geographic locations. Thus, the sheer effort that has to go into stealing elections will naturally be disbursed. That doesn't stop election fraud from happening, but it mitigates its influence, making it less practical to use as a routine tool of nationwide partisan combat. But recounts are just the opposite: once the initial counts are in, both sides know exactly which race results can be overturned in the courts, and exactly how many vote changes they need to do it. This is unhealthy in the extreme.
Restoring public confidence in the electoral system requires work on a lot of different fronts, but one major candidate should be a serious effort in state and federal races across the country to raise the showing required to trigger a recount or lawsuit over election results, to preserve the option only for the most serious and severe cases of malfeasance. The current system is unsustainable and ultimately dangerous to democracy.
"A machine count is done behind your basic Rawlsian veil of ignorance: both sides may know they stand a chance of getting a raw deal in a close race, but they know it's basically an even chance."
I don't think that Democrats think that's correct. Democrats think that, given election technology and their likely voters, they get a rawer deal -- that is, their voters are more likely to screw up the ballot somehow so that a vote is not counted. Which is why Democrats gained votes during the recount in WA Gov, MN Sen, and in FL in 2000.
I don't know why the Democrats always seem to be the ones gaining votes - maybe they are better at the legal tactics and maybe they are right that they lose more votes to bad counting than Republicans do.
AS, considering Bush V Gore, and the 2000 election, your statement, "maybe they are better at the legal tactics and maybe they are right that they lose more votes to bad counting than Republicans do." is a bit absurd. And Crank, maybe, just maybe, the Minnesotans aren't Coleman's constitutents, but Franken's.
Daryl, that line actually wasn't supposed to be argumentative. Coleman's the incumbent, and to date the system in Minnesota has neither returned him to office nor replaced him.
Daryl - fair point as regards FL. Nonetheless, IIRC Gore gained votes in FL over time. The Bush lawyers were just good enough to stop the recount before he gained enough to win (which could have happened if the over and undervotes were counted, under some scenarios, I think).
From what I remember of the WA recount and from Powerline's description of the MN recount, the Democrats have outlawyered the Republicans in those instances.
"and the Al Franken, Christine Gregoire and Al Gore efforts to overturn Election Day results"
Uhh, Crank? I believe Minnesota law gives an automatic recount within a margin of victory. Not Franken's choice - only his choice to participate. Both he - and Coleman - could have said that they would be totally hands off. They chose not to be.
Similar to Florida's pre-2000 law, just that it is automatic.
It doesn't change what I perceive to be the main point of your argument, that once lawyers get involved, things get worse. Once partisan lawyers get involved, even more so.
"That is Al Gore's lasting legacy to our democracy, and it's a deeply malignant one."
Both parties Crank. Republican lawyers flown in from around the country to bang on the doors is a lasting image as well.
Gore petitioned (as allowed by law), but the first lawsuit was filed by Bush, IIRC.
That you would only put it on Gore is not surprising, but to say that both parties did not contribute to both the failure of Democracy then and since is a bit silly. And, without accepting this basic principle, the efforts to restore public confidence becomes less likely.
Looks to have happened in 1962 as well. You should read up on it - looks stunningly like today, and 2000.
Democrats always win the after-election vote counting because they justify fraud as a moral response to evil (see Krauthammer -- "GOP sees Dems as wrong, Dems see GOP as evil"). The worst election fraud in the country is consistently in Dem controlled areas (Chicago, Philly, Det, Milwaukee, St Louis, NY). That's not surprising given the corruption which dominates the backbone of the Democratic party -- unions.
Everyone I know expects the Dems to succeed in stealing the MN senate seat just as they stole the governor's seat in WA and tried to steal the presidency in FL. Until we get an honest news media, they will continue to get away with stealing elections with ACORN before the count and with vote contests after the count.
AS made a surprisingly honest statement: "I don't think that Democrats think that's correct. Democrats think that, given election technology and their likely voters, they get a rawer deal -- that is, their voters are more likely to screw up the ballot somehow" That is tantamount to an admission that the Democrat message is more likely to appeal to the less intelligent. Interesting.
To Crank's larger point, I think Stan hits the nail on the head. In a very close election Republicans tend to play by more genteel rules and Dems fight to win the recount by any means necessary ethics be damned.
It is comical to watch Coleman resort to the activist courts to overturn the will of the people after the ballots have been counted ad nauseum in the presence of both campaigns and election officials and he has come out on the short end. I guess his election day advice to Franken to "just concede" is something that he won't apply to himself. Classic.
Can someone explain to me why voters can't be trusted to read and follow the directions on their ballot, and to accept that their vote doesn't count if they screw up?
The biggest problem here is that folks like me with little understanding of what's going on in these processes see that election officials are inferring voter intent. The consequences of letting these votes go uncounted seem far less significant than those of allowing elected and unelected officials "interpret" them.
Largebill writes: "In a very close election Republicans tend to play by more genteel rules and Dems fight to win the recount by any means necessary ethics be damned."
And the Preppy Mob sent by House Republicans to stop the recount in Dade County was genteel? Will all you loser GOPers please stop whining?
Crank's largest point, with which I fully agree (wait, watch for lightening bolt . . .) is that no one beleives that any recount process will produce a result in which all can have confidence. That, however, ought to be a situation that is fixable. At least two things have to happen. First, there needs to be a uniform national standard for voting machines; i.e., a relatively small number that are deemed acceptable. Second, for each voting machine there needs to be a set of rules for how recounts of votes on that machine will be conducted. It is possible that there will be future elections in which a new system error is discovered, but there will be agreement beforehand that the only remedy is to make a change applicable only in future elections.
Joel: The problem with your claim is that ignores the fact that machines have error margins where they improperly reject a certain percentage of valid ballots in every election. So in an extremely close election you gotta go back and hand count them all to get a valid result, which is what was done here. And for the "interpretive" part of it; the Minnesota officials and members of both campaigns posted examples of disputed ballots online where even a single stray pen mark elsewhere on the ballot could invalidate a clear Coleman or Franken vote. in the end those sorts of corrections tend to balance out. Why should we have more confidence in the opening night count than the recount is the question that you should be asking.
Seth, you are trying to have your cake and eat it too here - either the rate of expected error on Election Day is random, or not. The only reason for it to be non-random is if there's something wrong with the way one party's voters mark their ballots. Whereas there are a whole lot of ways that a recount can be non-random in making errors.
Thus, you have the question backwards. Unless the recount materially increases the accuracy of the count, you are just taking one imperfect but unbiased system and replacing it with another imperfect but potentially biased system, at great additional expense and delay and cost to voter confidence in the system.
No, as in my comment to Joel what I'm saying is that the machine error rate present in every election is higher than the marginal change caused by "human error" where stray pen marks cause a vote to be discounted. Thus when the vote margin in the initial count is extraordinarily small, a hand recount that corrects the inevitable machine error and allows for "interpretive" changes is more accurate. The loss of voter confidence is overstated....if Bush/Gore didnt stop people from voting inthe next election or cause them to riot in the streets, a manual recount in Minnesota won't either. The polling data backs that up, and reveals that the only people are dissatisfied are , you guessed it, Republicans who wanted coleman to win anyway.
Nate Silver actually predicted with a scary amount of accuracy prior to the recount that Franken would pick up a few hundred votes and win the election based on (1) historical machine error rates in counties with particular Republican and Democratic biases (2) the not surprising fact that machine error is NOT random, ie the lousier machines are in the poorer , Democratic areas, and (3) pen mark errors are NOT random either, more likely to be made by older, infirm voters trending Democratic. That a recount could be so accurately predicted actually does what Crank says it cant...vastly improve the accuracy of the election result. 538.com archives if you dont believe me. The great damage to voter confidence in 2000 wasnt caused by the recount, but by the fact that a politically divided Supreme Court STOPPed the recount and shed doubt on the election result itself. Admittedly, a doubt split heavily along party lines.
True, machine error isnt randomly distributed.
Magrooder, I don't know that national standards are needed, but certainly we should have uniformity within each state, and of necessity that will lead in the direction of national best practices limiting the market to a handful of designs.
I find it interesting that everyone complains about New Yorks antiquated cam based machines, yet there seems to be few problems with the ones that work in getting an accurate tally.
The American public isnt as simple minded as Crank is supposing. If an election is extremely close the majority support spending money to make sure the result is correct. After all, we're talking about choosing a very powerful person here to affect the country's direction. And nobody believes that the provisional counts announced on election night are error-free. This is why the Minnesota law forcing an automatic recount if the margin is close is an excellent one mirrored around the country, because machine and human error happens, sometimes frequently so.
AS: when the newspapers went through and did their own recount in Florida, they found that the looser the standard they applied (i.e., the closer to what Gore wanted), the bigger the Bush margin; the stricter, the closer it got, until the strictest possible standard actually gave Gore the state. The result was reported, then promptly dropped from the media's collective memory.
Seth: the court itself said that the election contest would be necessary; it's part of the formal process in Minnesota, not Coleman's idea.
And as a general comment, the more I see, the more I think everyone should follow Georgia's lead: if it's disputed, have a runoff with only the top two candidates.
It's pretty rich watching you Republicans claim some righteous mantle over recounts and corruption scandals, and claim that recounts weaken voters' confidence in the legitimacy of democratic processes. All through this election season we saw you throw all kind of nonsense at the wall to plant seeds that Obama was stealing the election through ACORN, with Obama's 10 miilion vote margin being accounted for by Mickey Mouses and the Dallas Cowboy offensive lineman. Then there is all the bs about Franken "stealing" the election, when Coleman's people havent found one ounce of fraud in the recount, they just want more ballots from conservative-only districts to be counted. Kudos to Nate Silver for systematically demolishing the Coleman camp's claims about the recount. All you're left with as Republicans is the smug assurance that you're "better people".