POLITICS MAY 17, 2010
-
Read Later
READ LATERAvailable only to subscribers. SUBSCRIBE TODAY
-
Listen
ARTICLE AUDIO
- Font Size
Charlie Crist has fled the GOP. John McCain has sold his maverick soul. Bob Bennett just got throttled by conservatives. Michele Bachmann is occasionally taken seriously. As Tea Partiers impose purity tests on Republican pols and the party’s center is dragged ever further right, is it any wonder many people are bemoaning the death of bipartisanship and moderation in American politics?
Every election cycle, a few charismatic candidates (especially at the presidential level) exploit voters’ disillusionment by promising to change the tone, ease the polarization, stem the partisan bile, and otherwise end politics as we know it. But under the current system, that is never going to happen. Ever. Congressional districts are drawn to feed polarization, and party primaries are dominated by wingers from both ends of the spectrum, making it tough for candidates with more mainstream appeal to advance to the generals. For any real shake-up to occur, a significant structural overhaul is needed. I’m opening the floor to any and all suggestions. In the meantime, one possibility I’d like to see chewed over: replacing party-based primaries with nonpartisan jungle primaries, also known as Louisiana primaries.
Yeah. I know. It feels so very wrong to suggest that other states take any cues from the corrupt, screwed-up politics of the Bayou State. Crazier still, the Louisiana system—which pits all candidates regardless of party against one another in the initial primary, with the top two vote-getters advancing to the final showdown—was introduced by legendary Governor Edwin Edwards, a man known as much for his ethical flexibility as his political prowess. (The four-term governor is currently serving a 10-year sentence for racketeering.) A Democrat, Edwards pushed for jungle primaries as a way to stem the rise of the state GOP. (In fact, while the system boosted Edwards’s career, it more broadly ended up helping local Republicans.)
The most obvious advantage to a nonpartisan blanket primary would be to free candidates from kowtowing to the craziest elements of their bases. All candidates would be appealing to all voters, diluting the influence of the wingers—or at least any one set of wingers.
This system would also reduce the kind of tactical voting where voters are loath to support less-established candidates for fear of “wasting” their ballots. (The best send-up of this remains the Simpsons’s “Treehouse of Horror VII,” where aliens Kang and Kodos pose as Bob Dole and Bill Clinton in the 1996 presidential race; when the plot is exposed and voters threaten to support a third-party candidate, Kang taunts them, “Go ahead, throw your vote away!”) In jungle primaries, by contrast, voters could go with their top choice in the first round, secure in the knowledge that, if their candidate didn’t make the cut, they’d have another chance to influence the final outcome.
Jungle primaries do lend themselves to other types of strategic voting. For instance, a block of Democratic voters could band together to try to boost a weak Republican candidate into the final round. (This, however, is a tricky business that could wind up throwing the race to the “wrong” candidate.) They also increase the odds that a voting district dominated by one party could wind up electing a candidate from the opposing party, depending on how much vote-splitting occurred. (Say, if five Democrats ran and only two Republicans—or one Republican and one third-party candidate—fragmenting the Democratic vote to the degree that none of the party’s candidates wound up in the final round.)
It’s hardly surprising that the parties haven’t exactly rushed to embrace nonpartisan primaries, which, after all, would greatly reduce their power. The Democratic Party actually sued to disband a similar blanket system briefly attempted in California, taking the case (successfully) all the way to the Supreme Court. Plenty of states, including Texas, use jungle primaries in either special elections or low-level races, but getting them in place for high-stakes races remains a tough haul. The Louisiana system is no longer used even in its namesake’s federal elections. Grappling with constitutional questions, in 2006, then-Governor Kathleen Blanco signed a law returning the state's congressional elections to the regular primary system, effective 2008. In March 2008, however, Washington state’s nonpartisan blanket primary was deemed constitutional by the high court, providing an opening for other states interested in following suit.
Here’s hoping at least a few more states give it a test drive. Every system has its downsides, and certainly nonpartisan primaries are no guarantee against nutters. (In Louisiana’s 1991 governor’s race, David Duke, running as a Republican, placed second in the initial vote—knocking out then-Governor Buddy Roemer—before ultimately falling to Edwin Edwards, who was staging his comeback bid.) But most of the objections I’ve seen to the jungle primaries haven’t been particularly compelling. Many are premised on the need to keep parties strong as a way to prevent the electoral process from degenerating into chaos. Looking around, however, it’s not clear that the parties are getting the job done. More and more, it feels like the time has come to try a different kind of chaos.
Michelle Cottle is a senior editor of The New Republic.
For more TNR, become a fan on Facebook and follow us on Twitter.
13 comments
Many democratic countries that don't have two major parties essentially have "jungle" voting systems. I have never heard that their end result is consistently better than ours.
- drofnats1
May 15, 2010 at 12:17am
What about requiring that districts are drawn by a non partisan commission, and also expanding the number of Representatives. It would also be nice if the states went to unicameral legislatures, I never understood why so many states have both a Senate and a House; it would be less confusing to voters.
- vips73
May 15, 2010 at 12:55am
Give crackpots their due. Which is to say adopt proportional representation. If 25% of voters are crackpots, let them have (roughly) 25% of the elected representatives. It gives them a voice (and an outlet for their craziness) but without putting the rest of us at risk that the crackpots would actually take control. Now, a small minority (an optimist, I continue to believe crackpots are a minority) of voters can actually determine the outcome of an election by controlling one of the two major political parties. Only a crackpot would allow such a crazy system to continue. I know, the Republican Party would be the big loser as all their crackpots would have to be separated from the rest. But over time the Party would recover as the Party adopts sane positions, even positions that would attract non-crackpots. Unlike the Republican Party today.
- rayward
May 15, 2010 at 7:33am
First off, the epithet "winger" is short for right-winger. Nobody uses it to refer to the hard left. Second, the notion that the left "dominates" the Democratic party in any respect whatsoever - including the primary process - is simply, demonstrably false, and asserting otherwise is just transparently silly pox-on-both-houses posturing.
- santoast
May 15, 2010 at 8:27am
I like rayward's suggestion. Where do I change my party registration to the Crackpot Party. I assume it will fissure into the bipolar party.
- skahn
May 15, 2010 at 1:19pm
@rayward -- with proportional representation, it's possible for niche minority parties to end up being the "swing" votes needed to achieve a majority in the legislature and actually form a government. This happens in Israel, for instance, where the ultra-religious pro-settlement parties can sometimes hold the governing coalition hostage. Even without proportional representation, this can happen, as in Britain last week, where a small minority (of parliamentary seats) party wound up with the power to make or break the government. My favorite (hopelessly impractical) idea along these lines is that serving in the (state or federal) legislature should be like jury duty. Every citizen is eligible, you get a thing in the mail, and you have to go serve in the legislature for a month, or two months, or whatever, and your boss has to let you do it and can't fire you for it. Then your term expires and you're done. It's got obvious problems, you probably couldn't get legislators up to speed very well on complex issues with such high turnover, and whoever does the bringing-up-to-speed would soon be in effective control. But it's fun to think about.
- reid
May 15, 2010 at 3:15pm
I fear your suggestion is based on an erroneous assumption ~ that voters and activists are as up to speed on civics and care about community building as much as you. We have not educated ourselves well enough (the classic, Founding Fathers' revolutionary ideal for broad, liberal education for all citizens) to maintain the current political system... hence its steady deterioration. Jungle primaries assume, just as the current system assumes, the voters understand that they are fully vested in the process and that the outcomes demand their continued involvement. If the latter is broken, how can your replacement work any better?
- ldmark
May 15, 2010 at 3:53pm
ldmark: The main problem is turnout, and this isn't a bad solution. The current primary system favors candidates most attractive to the ideological hardliners, who are the voters most likely to show up. A jungle system would tend to favor candidates with broader appeal -- and seems to me more likely to foster a sense that voters are invested in the process, since they could now vote for the best candidate in the field and make it count. The hardliners are still going to show up and vote for the wingnuts and moonbats, but their power would be diluted.
- frippo
May 15, 2010 at 4:42pm
I am with Vips on this, end gerrymandering, it is 2010, write a computer program that just lumps districts together by no other criterion than population in a concentric manner as possible.
- blackton
May 15, 2010 at 5:26pm
I'd like to suggest a different approach: pairwise-ranked elections. That is, whatever number of candidates run in the election, each one would run head-to-head against each other candidate with the candidate who defeats each other candidate winning the election (the pairwise part). The pairwise comparison between candidates is based on the voters ranking the candidates (the ranked component) with the candidate in any pair ranked higher than the other receiver that voter's vote. On a number of measures, it is clearly superior to non-partisan primaries. As far as strategic voting goes, it comes closest to making a strategic vote an honest vote. Simply put, ranking candidate A higher than candidate B when you prefer B over A can help B get elected when A otherwise would, but cannot help some other candidate get elected instead of B. Under pairwise-ranking, candidates can't be content with exciting a large number voters. Instead they have to look at how many voters they turn off in the process of exciting their voters. Candidates who ignore this will be ranked near the bottom by most voters who do not rank them first and thus lose every pairwise-contest. Further, pairwise-ranking would allow the election to take place in one stage. From the standpoint of moderates, extreme voters are more likely to be motivated to turnout multiple times per election cycle. Thus a one-stage election will increase the voice of moderate voters compared to extremist voters. Comparing these to non-partisan primaries, Ms. Cottle already discussed strategic voting under non-partisan primaries. In non-partisan primaries, candidates are every bit as incentivized to pursue energizing their base, damn however number of people are turned off (which is the specialty of the "nutters") as they are in the present system because being a voter's second choice is no improvement over being that voter's eighth choice. Invariably, jurisdictions with non-partisan primaries have higher turnouts for whichever round is in November, with the drop-off less likely to have strong positions on the issues and thus less likely to support the "nutters." Also, conducting the election in one stage would save money in running the elections.
- sighthnd
May 16, 2010 at 11:52pm
I agree with Michelle that the Louisiana system for primaries would be an improvement. (Then again, I despise the primary system at all. If you want a say in choosing a party's nominees for office, then get off your ass and join the party and show up to its nominating convention.) And I agree with blackton that the Iowa system for drawing districts would be an improvement. But just as important is increasing the size of the House of Representatives. Smaller districts will reduce the effectiveness of Gerrymandering and give ordinary, non-winger voters more power to effect election outcomes. Congress can set the number of seats by simple statute; 435 seats is a law passed in 1912, not a provision of the Constitution. If we got back to the number of constituents per seat of just a couple of generations ago, we'd have 885 representatives, all of whom would be much easier for a challenger to beat and thus more likely to hew to the stablest possible middle within their districts. (Also, popular-vote losers could not become president were the House expanded to keep up with population growth.)
- rhubarbs
May 17, 2010 at 8:42am
I find it disturbing that more and more journalists are promoting the idea of the “jungle” primary, more accurately referred to as a “top two” primary, while third parties and elections focused political scientists oppose the idea. This proposal was recently placed on the Oregon ballot (Ballot Measure 65) as a referendum and was luckily voted down. For a pre-election citizen report on the subject, see www.pdxcityclub.org/system/files/reports/Measures63and65_2008.pdf While I am not against test driving good but untested ideas in state laboratories, deeper study strongly indicates that a jungle primary would at best maintain the status quo, and could very likely make politics and partisanship worse. Third parties oppose this proposal because it would all but completely eliminate their chances of appearing on general election ballots. For people who are not concerned about the disappearance of third parties, a more serious implication is that where the Republican party is strong, it will only become stronger, and likewise for Democratic regions. In areas such as Michele Bachmann’s district, the general election will end up being between two Republicans almost every cycle, resulting in the complete disenchantment of Democrats in the district (as if they don’t have it bad enough already). Far from breaking the power of the two major parties, the jungle primary only serves to reinforce it. The argument that Democrats and centrists will force the election of a moderate Republican in such a district is tenuous. It assumes broad-based, engaged, and strategic voting from the electorate as a whole at the primary stage of the election, a vision that is not supported by any research. Moderates fail to vote in primaries not because they are disenchanted with partisan politics, but because they are not engaged with the political process early in the election cycle. Jungle primaries will not change that fact.
- Gimli
May 17, 2010 at 9:22pm
Instead of a bipartisan commission drawing district lines for congressional and state representative office, I suggest a computer design which has the goal of having the lowest possible total perimeter for all districts. This would produce the total opposite of gerrymandering. It also would have the happenstance of having joint city/suburban districts instead of the current situation where most suburban congressional districts wrap around the city. The current system panders to polarization and special interests. The proposed system would be strongly opposed by both the far right as well as the far left since it would greatly reduce the number of "guaranteed" Republican or Democratic districts and instead increase the number of districts where the middle (generally not that interested in the primary election) would have one or two candidates in the final election.
- Ken Streit
May 18, 2010 at 10:23am