I was fascinated by the Sunday New York Times story about Democratic superdelegates,tracking how many of them were supporting Barack Obama or Hillary Clinton despite their home state or district having gone for the other candidate. I wouldn't have done the story quite that way. Which doesn't make the Times' methodology wrong; it just pointed up to me how hard it's going to be to select any one fair formula for determining how superdelegates should vote.
Let's first look at the Times' methodology. The piece by Adam Nagourney and Farhana Hossain examined whether superdelegates, who make up 19 percent of the Democratic delegate pool, were supporting the same candidate voters in their state or congressional district did (assuming their state had already voted.) And they found far more Clinton-supporting superdelegates resided in Obama-supporting districts than the opposite -- Clinton led Obama 79-34 in committed superdelegates who were bucking the wishes of their home-state or district voters.
But then I started looking at the category of "party leaders" (as opposed to elected officials who are superdelegates). The so-called "party leaders" are mainly interest-group leaders, former elected officials and lobbyists, and I noticed that a quarter of the Clinton-supporting party leaders the Times singled out for bucking their supposed "constituents" live in Washington, D.C., Maryland and Virginia. That's where many party leaders happen to live, which makes sense, given where the power is. But I'm not sure it's fair to ding them for not supporting the same candidate as local voters. If you remove the DC-area super delegates, Clinton only leads Obama 56-34 in committed superdelegates who aren't supporting their constituents' candidate.
But why stop with the DC area? I then removed all "party leaders" from the superdelegate tallies, because I'm not sure it's fair to expect any of them, no matter where they live, to represent the wishes of local voters; that's not how they derive their status as party leaders. (For that reason, I'd be fine with chucking the whole "party leader" category of superdelegate, since they don't have the same political checks on the way they decide who to support as elected superdelegates, who can be thrown out by their constituents if they abuse their status. But that's a whole 'nother story). Removing "party leaders" from both candidates' tallies would narrow Clinton's edge among superdelegates who are bucking their constituents' wishes to 23-15. You can still argue that those 38 individuals shouldn't vote that way, but the imbalance looks a little different.
But the entire Times project seems built on the premise that superdelegates should vote with the voters in their state or district, and that's not clear to me at all, even when it comes to elected superdelegates who do represent those constituents. New Mexico Gov. Bill Richardson expressed the confusion well in the Times piece: His state voted for Clinton, but it's still possible, even likely, that Obama will lead among pledged delegates nationwide. Is it obvious what Richardson should do in that case? Should he vote with his constituents? Or should he ratify the choice of the nation's Democrats? Sure, if either candidate has a big lead, it doesn't matter much either way. The problem shows up if there's a tie, or close to it, in either direction.
I'm not even sure that superdelegates should have only those two choices -- to vote for either their constituents' choice, or the overall top vote-getter -- rather than the option of supporting the candidate they think is best for the country and the party. Like it or not, the system was set up to provide a brake, even an override, in a situation when the primary and caucus process produced a seriously flawed candidate. (Here is a great Boston Globe primer on the process.) Picture a scenario in which Hillary Clinton beat the odds, won the remaining primaries and emerged the top delegate winner after the last primary June 9 (and remember the nominating convention isn't until the end of August.) Imagine that in the intervening two and a half months, Clinton faced a disastrous personal or political scandal, or the nation experienced a domestic or international crisis in which she stumbled and Obama shined, or that both candidates floundered in some unlikely way. (Of course, technically even pledged delegates can ultimately vote for a different candidate, so in the case of a truly earth-shaking scandal, they could change sides.)
If the party wanted the superdelegates to merely ratify the voters' choice, the party wouldn't have superdelegates in the first place. The superdelegate system is set up to allow party leaders and elected officials to anoint the best possible candidate, which could ultimately mean Obama, even if Clinton is the leading vote-getter, or Clinton in the opposite scenario, or even somebody outside the process right now. They're unlikely, but those are scenarios the superdelegate system was established to handle.
Let me be clear: I'm not defending the superdelegate system. Democrats can, and probably should, put it up for redefinition after this election. This particular configuration has never before been tested in a close race, and the fact that there's this much contention and confusion would seem to indicate there's something wrong with it. I also think it would be a disaster if there is a clear popular vote and pledged delegate leader come convention time, but the votes of party bosses selected the other candidate. Still, this is the system the party designed and that party leaders ratified to govern this election. And it's hardly the only anti-democratic (small-D) element of our democracy.
We've got many institutions that are designed to check the power of a raw voting majority, from the U.S. Senate to the Electoral College. We've also got a crazy patchwork system of primaries and caucuses governed by totally different rules, as well as Texas's now utterly crucial primary-plus-caucus. Some of those state systems try to balance out urban power by giving extra delegate muscle to non-urban areas, as in Nevada (which gave Obama more delegates than his raw popular vote totals would have); others, like California and Texas, potentially give more delegates to dense, Democrat-rich Congressional districts. Most states' rules have never mattered much nationally because most states' primaries and caucuses didn't matter; now they do matter, in all their eccentricity, in this close race, and they should all come in for some closer attention when this is over.
But claiming that there's one clear way for superdelegates to vote is either naïve or dishonest. The Obama campaign seems to know this, even if many supporters, most notably MoveOn.org, are acting as though superdelegates must support the pledged-delegate winner. Obama's remarks have shaded towards the point of view that superdelegates should support the voters' choice, while his campaign manager, David Axelrod, has said they should vote for the person they think is best. As with most things, Obama's campaign has played the media and expectations game better than the Clinton campaign, succeeding in having it both ways, and leaving itself open to making whichever argument fits its circumstances best in August. Advantage, again, Obama. But August is a long way off.