Slightly more on superdelegates

Yglesias says (pardon the long quotation):

I don't think I buy the argument that the Democratic Party's superdelegates have some kind of categorical ethical obligation to obey the dictates of the pledged delegate count. Indeed, one of the best things you can say about superdelegates is that it's fairly easy to imagine scenarios in which giving the nomination to the pledged delegates leader would have a perverse result. For example, suppose Candidate A cleans up in early primaries and jumps out to a big lead. But just when the pundits were ready to declare it "essentially impossible" for Candidate B to catch up, he unveils a very appealing new message and sweeps the remainder of the states. Thanks to the proportional allocation rules, though, it's not enough to catch Candidate A, who winds up with 52 percent of pledged delegates. But since many of those delegates came from states that voted months ago, and lots of former Candidate A supporters feel buyer's remorse; national polling shows convincingly that 59 percent of registered Democrats prefer Candidate B, who also has a lead in head-to-head polling matchups with the GOP nominee and a fundraising advantage.

Would it really be so absurd for the superdelegates to overrule the "will of the people" and instead give the people what they tell pollsters they want? I don't think so. The superdelegates have both an opportunity and an obligation to take seriously their obligation to do the best thing for the party and the country.

This is an interesting argument. Specifically, the whole "time-delay," "buyer's remorse" theme deserves attention. Now, the obvious but for the moment untenable solution seems to me to abolish superdelegation and have a national popular primary on a specific day, with attendant time- and opportunity-cost-saving resources for voters, like a national holiday (this should also apply to the national election).

This would force candidates to develop and test their campaign promises early on, rather than constantly try to one-up one another following the results of various early and well-funded state primary campaigns. As it stands now, serious political inequality is at work, since Iowa's caucuses and New Hampshire's primary voters enjoy major (and majorly arbitrary) privileges over other states. (In effect, their votes "count more.")

This argument is made often, but here I'm connecting it to Yglesias' thoughts. The idea of "buyer's remorse" seems strange to me: What is this, in fact, other than "If I knew then what I know now, I would vote for X"? But the changes in campaign strategy and rhetoric which precipitate such changes of preference seem to be purely cynical ("tactical" if you like) on the part of the candidates. The interdependent shifts in strategy and preference are then interminable, at least until the last state holds its primary. But again, this brings up questions of serious political inequality.

Yglesias' argument seems to be: when opinion polls change, let the elites respond to the public. But then the spiral of polling and voting potentially goes on ad infinitum: the "moment of choice" has to come some time, and I would much rather it be a matter of the rank-and-file party membership's judgment, not the superdelegates'. To push it to the extreme, according to Yglesias' logic, anything could happen even after the last state has cast its votes. Why should we expect changes of preference to cease just because the states have all voted? There might as well not be any primaries at all, just polling. To cut through this confusion, I think it would be best to do the one-day national poll as detailed above. But as things stand now, this "appealing new message" <=> "new opinion poll" symbiotic relationship is unworkable and undesirable. There is perhaps a democratic argument to be made here on the grounds of initiative and recall, in order to best reflect voter preferences. But during a primary this can easily be hijacked by either/any of the competing candidates at what they consider the "right moment."

The one serious point here is that it may turn out, for whatever reason, that a different candidate does end up with a "lead in head-to-head polls against the GOP candidate." Since American elections are so marginally close and the importance of winning in order to effect any changes is so high, this carries some (unfortunately rather undemocratic) weight. But again, I think this train of thought leads to an eclipse of the meaningfulness of state primaries. Why not give the superdelegates the power to pick the "most winnable" candidate no matter what? Whatever the drawbacks of the current setup, it's best to make the primary voters responsible for their votes and for the superdelegates to follow their general lead, if they must be involved.