It was a good night on Tuesday for Hillary Clinton. She treaded water in the New Hampshire primary, coming out even with neighbor-state favorite son Bernie Sanders after her decisive win in Iowa. As the contest shifts to more diverse, and potentially friendlier, terrain, she is more than nine times closer to the nomination than Sanders.

The above paragraph might read like something hastily written as a placeholder and accidentally published. But based on the rules of the Democratic nominating contest, it’s entirely true. Clinton extracted the same number of delegates from New Hampshire as Sanders, despite losing the popular vote by more than 21 points. She won 29 delegates in Iowa to Sanders’ 21, despite the virtual tie in the caucuses. And she leads in the delegate count 394-42, a deficit that will be difficult for Sanders to make up, since no Democratic primaries are winner-take-all, where he could rack up lots of delegates at once.

How is this possible? The answer is superdelegates, the 712 votes doled out to Democratic National Committee officers, elected officials, and other party luminaries. The superdelegates are free to vote for their preferred nominee, unbound by the will of the voters—and if a nominee they think is terrible for the party is close to securing the nomination, they can conceivably throw their weight behind an alternative.

Partisans for Clinton and President Obama both howled about superdelegates during the close 2008 primary, but they never went away. In fact, in candid moments, DNC officials would tell you that superdelegates were designed for precisely this year’s scenario: to cool the hot passions of voters—in this case Sanders voters—and tip the scales to a primary candidate perceived as more electable.

The problem is that if the superdelegates actually followed through with coalescing behind Clinton to deny Sanders, it would make the 1968 Democratic Convention look like a garden party. Just think of the damage to the Democratic Party—which has made voting rights one of its major causes—if it blocked the candidate with the most delegates in primaries and caucuses from receiving the nomination. It would reinforce every conception of the corruption of the political establishment that has been a touchstone of Sanders’s campaign. 


A little history is in order. The Democratic nomination fight is a race for 4,763 delegates to the party convention. The vast majority—4,051—are chosen in nominating contests around the country and in U.S. territories like Guam and Puerto Rico (even Democrats Abroad, an organization for expats with U.S. citizenship, gets delegates). The other 712 are the unpledged superdelegates, who are free to choose their favorite candidate.

One theory behind the existence of superdelegates is benign: They came about because party leaders—members of the DNC, ex-presidents and vice presidents, House and Senate Democrats, and governors—wanted to attend the convention and vote on the party platform without having to challenge the local activists and supporters who run for the right to be convention delegates. This way, the luminaries are guaranteed a seat at the convention, involving them in the process and, hopefully, motivating them to work for the nominee in November.

But superdelegates were also created in reaction to Democrats choosing nominees who were out of step with party leadership, like George McGovern (1972) and Jimmy Carter (1976). Party leaders would have dearly loved to stop McGovern in particular, but had no mechanism to do it. So the 1982 Hunt Commission, chaired by North Carolina Governor Jim Hunt, designed the superdelegate system to give the party poohbahs a check on the primary process. The leaders themselves would never say that publicly—they usually argue that superdelegates have never decided a nomination—but that was clearly the mission, to offer a safety valve for a primary race that goes off the rails.

It would not be a stretch to imagine that some party insiders believe that nominating a 74-year-old democratic socialist constitutes just such a scenario. You need only look at the current numbers—Clinton has commitments from 355 superdelegates to Sanders’s 14—to see that the superdelegates are in a wildly different place than voters, at least voters in the first two nominating states.

Superdelegates came close to being decisive in the very first nominating contest after their creation. In 1984, they preferred Walter Mondale to Gary Hart, but Mondale ultimately won a plurality of delegates. So while superdelegates put him over the top, he was also the narrow choice of the voters. But the experience of 2008 was supposed to make superdelegates a thing of the past. 

Clinton initially had control of the majority of superdelegates, but many gradually flipped to Obama as he won more primaries. Ironically, Clinton ended up winning enough later contests that she actually secured more raw votes than Obama (though that’s due mostly to a Michigan primary where Obama wasn’t on the ballot). But the combination of strategically significant victories in caucus states and support from superdelegates swept Obama to victory.

After the 2008 primary, everyone called for reform of the undemocratic system where random party insiders could defy voters and determine the nomination. But you don’t usually get far asking people to affirmatively vote to relinquish their own power. The DNC very quietly scrapped a reform plan in 2010 that would have forced superdelegates to vote for their states’ primary winner. Instead, they opted for a weak substitute of slightly limiting superdelegate influence by increasing overall delegates. In 2008, 20 percent of total convention votes came from superdelegates; this year, it’ll be 15 percent.

That puts us potentially in a similar situation to 2008, where in a close primary, the role of superdelegates can become critical. It seems unthinkable that superdelegates would overturn the popular will if Sanders falls just short of the delegates he needs to secure the nomination. This would demoralize Sanders supporters and throw the party into disarray, virtually ensuring a big loss in the general election. Recognizing this, the superdelegates would almost surely shift, like they did in 2008, and line up with the vote in their states to prevent the appearance of impropriety.

But to be clear, the superdelegate process was designed so the party can stop candidates it determined to be unelectable. “The superdelegates were created to lead, not to follow,” wrote the late Geraldine Ferraro, a member of the Hunt Commission, back in 2008. “They were, and are, expected to determine what is best for our party and best for the country.” There’s no question that members of the party establishment could talk themselves into believing that by throwing support to Clinton and denying a socialist control of the top of the ticket, they were saving the party rather than destroying it.

The backlash to that possibility has already begun, months before the process plays itself out. MoveOn.org has petitioned superdelegates to “commit to honoring voters” by announcing that they would support a nominee who received the most support from the primary-voting public. Paste magazine writes that the establishment is touting the superdelegate numbers in the wake of Sanders’s New Hampshire victory to “scare” Sanders voters away from participating in a broken system.

In truth, superdelegates are neither nefarious nor benign. But they are, by definition, a ripcord for party elites to attain control over the primary process. Their very existence creates unnecessary suspicion. Of course, asking the superdelegates to vote themselves out of existence is certainly a tall order. But this should have been done eight years ago. Party leaders already have enough power to affect outcomes; they don’t need another opportunity. Especially because, if they ever used it, the ripcord would in practice become a self-destruct button.