To understand the origin of superdelegates, you have to understand one thing: George McGovern and the 1972 election. But let's first step back and frame things a little bit.
The superdelegate system was instituted over the spring and summer of 1982 by the Commission on Presidential Nominations (CPN), a special committee of the DNC that was chaired by then North Carolina governor James B. Hunt. Superdelegates were the most important of several such changes approved by the CPN; others included shortening the primary season, and loosening the rules for pledged delegates (such that it became easier to be a 'faithless' delegate).
There were a number of rationales given at the time for the implementation of superdelegates, none of which are necessarily mutually exclusive. The primary purpose of this diary will be to explore those rationales, based on a survey of contemporaneous newspaper accounts from the New York Times. However, it is also important to understand the underlying context: as of 1982, the Democrats had had two absolutely disastrous results out of the last three Presidential election cycles.
In 1980, Jimmy Carter was defeated badly by Ronald Reagan, losing by nearly 10 percentage points in the popular vote, and by 489-49 in the electoral vote. Although these margins speak for themselves, this defeat was actually a bit worse than it appears on the surface. In fact, it remains the worst defeat ever of an incumbent in the electoral vote, and the third worst ever in the share of the popular vote:
Popular Vote Margin
Year Incumbent Vote % Challenger % Margin
1912 Taft (R) 23.2 Wilson (D) 41.8 -18.6
1932 Hoover (R) 39.7 Roosevelt (D) 57.4 -17.7
1980 Carter (D) 41 Reagan (R) 50.7 -9.7
1840 Van Buren (D) 46.8 Harrison (W) 52.9 -6.1
1992 Bush (R) 37.4 Clinton (D) 43 -5.6
1892 Harrison (R) 43 Cleveland (D) 46 -3.0
1976 Ford (R) 48 Carter (D) 50.1 -2.1
1888 Cleveland (D) 48.6 Harrison (R) 47.8 +0.8
Electoral Vote Margin
Year Incumbent EV Challenger EV Margin
1980 Carter (D) 49 Reagan (R) 489 -440
1912 Taft (R) 8 Wilson (D) 435 -427
1932 Hoover (R) 59 Roosevelt (D) 472 -413
1992 Bush (R) 168 Clinton (D) 370 -202
1840 Van Buren (D) 60 Harrison (W) 234 -174
1892 Harrison (R) 145 Cleveland (D) 277 -132
1888 Cleveland (D) 168 Harrison (R) 233 -65
1976 Ford (R) 240 Carter (D) 297 -57
Carter failed to hold on to his core constituents. He won just 66 percent of Democratic votes, just 47 percent of labor union households, and just 44 percent of Southerners; he also won just 30 percent of independents. Moreover, the Republicans picked up twelve seats in the Senate, in addition to 34 in the House. It was an electoral disaster of the highest order, in certain ways combining the worst elements of 1994 and 2000.
This had come on the heels of George McGovern's embarrassing performance in 1972. Although McGovern did not cost the Democrats as substantially in the House and Senate, he lost the popular vote by 23 points, carrying only Massachusetts and the District of Columbia.
The party was panicked, and both Carter and McGovern were scapegoats. Carter was so unpopular at the time that there was actually some dissent when the CPN adopted a boilerplate resolution honoring Carter for his service to the Party. And Carter and McGovern were very much tied together as "insurgent" candidates. As Clinton surrogate Lanny Davis writes at the Huffington Post:
It did not seem entirely coincidental that the nominees since the Democratic Party reforms -- Senator George McGovern in 1972 and Jimmy Carter for reelection in 1980 -- suffered landslide defeats.This passage immediately strikes one as disingenuous, it being that Carter had defeated a sitting President in 1976, a result that Davis glosses over. But (although other parts of Davis' article are disingenuous) this bit is probably a fair reflection of the prevailing sentiment among Democratic elites at the time. As the New York Times wrote, the superdelegate rules "seemed infused with a desire to deny future nominations to political reincarnations of the Jimmy Carter of 1976". (NYT, 12/27/81)
The "Democratic Party reforms" that Davis refers to were those implemented by a commission led by McGovern and Minnesota Representative Donald Fraser, itself a reaction to the disastrous convention of 1968. The most important provision of these reforms was one that required all delegate selection to be "open" -- selected by voters rather than by party leaders -- which effectively ushered in the era of party primaries.
By 1982, however, the sentiment was essentially that the cure (1972, "validated" by 1980) was worse than the disease (1968).
In 1984, superdelegates proved to be helpful in getting Walter Mondale past the threshold he needed to achieve an outright majority of delegates, thereby avoiding a brokered convention. However, in all probability the superdelegates did not alter the outcome of the election; Mondale had a clear plurality of pledged delegates at the time. There is an outside chance that Gary Hart and Jesse Jackson could have teamed together to defeat him, but it would have required near-perfect coordination, and would arguably have usurped the public will, as Mondale had a substantive lead in the national polls. Thus, initially at least (and notwithstanding Mondale's eventual defeat), superdelegates were regarded as a helpful innovation.
And that's basically the last we heard of superdelegates until this election cycle.
Now that we've examined the broad history of superdelegates, let's go back and scrutinize the "framer's intent" behind their invention. There were three or four justifications advanced for the implementation of superdelegates at the time, all of which echo the historical narrative to some extent or another:
1. To increase the sense of order and avert a crisis at the Convention.
In this formulation, superdelegates basically exist as a mechanism to "break glass in case of emergency". Thus could run the gamut from providing some experienced, stabilizing voices in the event of a procedural fight on the convention floor, to potentially picking a different nominee in the event of an Eagleton-type crisis. Moreover, it was hoped that these superdelegates could restore some sense of order to the convention, as the 1972 Convention was regarded as a PR disaster, involving procedural fights late into the night and equal time to groups like gay rights activists that at the time were considered on the political fringe:
Democratic Party leaders say new rules adopted for this year's convention have fulfilled their purpose and created a more stable and predictable nominating process that favors mainstream candidates and policies.
This judgment was reinforced by a New York Times poll of the convention delegates that found that the new rules produced a group of ''superdelegates'' who were older, more experienced, more moderate and more loyal to the party than the delegates chosen by primaries and caucuses. (NYT, 7/15/84)
The 175 House members who are delegates to the Democratic National Convention plan to convene regularly and work as a unit to mediate any crises that may arise, Representative Tony Coelho of California, chairman of the Democratic Congressional Campaign Committee, said today.
The House delegates will also seek to avert potential political disasters that may not be appreciated by those involved in crisis management in the heat of the convention, Mr. Coelho said. In addition, the House delegates, who represent all sections of the nation, plan to serve as a ''natural whip system,'' he added.
The Democratic National Convention will mark the first time since 1824 that the House Democratic caucus will play a formal role in the party's Presidential nomination. The 175 House members, 164 chosen by the caucus and the rest elected by their districts, will be among 566 ''superdelegates'' who will include governors, big-city mayors, senators and other party and elected officials.
Of the 175 House delegates, about 85 percent support the candidacy of former Vice President Walter F. Mondale, 10 percent support Senator Gary Hart of Colorado, and the rest either support the Rev. Jesse Jackson or remain uncommitted.
''If there is a crisis of some sort, we would be the natural mediators,'' Mr. Coelho said. ''We have a good sense of the ripple effects of apparently short- term decisions.'' Such lasting effects could be overlooked by those involved in a dispute if they are anxious to achieve a swift resolution, he added.
''Basically you have a group in which people know each other and are willing to work for the party if something comes up,'' Mr. Coelho said. (NYT 7/4/84)
Return a measure of decision-making discretion to the national convention. Mechanisms such as candidate right of approval, designed to insure that delegates pledged to a candidate are bona fide supporters, would remain in the rules. But the addition of an increment of unpledged delegates, and the loosening of the ''binding'' rule as it applies to all delegates, would restore to the convention flexibility and an ability to respond to changed circumstances. (CPN committee report, via NYT 3/27/82)2. To get party officials more involved with the eventual nominee.
Probably the most widespread rationale cited at the time, perhaps because it's the most benign. Under the 1972 rules, senior party leaders could still be delegates to the convention -- but they'd have to run as "ordinary" delegates in their respective states. Many elected officials, particularly Senators, did not want to take the time to do this, while others were reluctant because it would have required them to officially stand behind a given candidate. It was argued that by involving party leaders more directly in the process, they could become more effective surrogates for the nominee in the general election cycle:
Under new rules adopted last year by the Democratic Party, House Democrats will hold a caucus, probably in the first week of February, to choose 164 of their number as delegates to the party's national nominating convention, where there will be a total of 3,933 delegates. The aim is to get more of the party's top elected officials involved in the nominating process.
Democratic elected officials traditionally played a major role in the Presidential nominating process until 1972, often winning seats at conventions by virtue of the fact that they held elective office. But the party, under pressure from young reformists, adopted rules that year that had the effect of greatly diminishing the officials' participation by requiring them to compete for seats, as anyone else had to. Since then, however, some Democrats have advanced the idea that ways should be found to bring elected officials back into the process as a way of uniting the reformist and office-holding arms of the party.
As a first step, in 1980, the party expanded each state delegation by 10 percent and reserved the extra seats for governors and other elected officials. But these officials were not given any independence to bargain; they were required to reflect the results of their state's primary and caucus.
Party officials insisted that most members of Congress would not want to be delegates if they had to run for the job, and that the only way to bring them back into the process of nominating a candidate and writing a party platform was to reserve delegate seats for them. That was done in 1982 by a special commission headed by Gov. James Hunt of North Carolina. (NYT, 12/22/83)
The argument over bringing more elected officials into the convention was evenly divided. Mr. Gifford said that this was necessary to get politicians to work for the nominee. Obera Bergdahl, Oklahoma state chairman, proposed giving all Democratic governors, senators, representatives and top legislators seats as delegates. But Linda Locke, Missouri state coordinator for the National Organization for Women, said that would cancel the commitment to having equal numbers of women and men as delegates. (NYT, 9/25/81)3. To nominate a candidate who can win.
While the first two rationales are more procedural, the latter two have a somewhat more specific outcome in mind. For one thing, in light of what had happened in 1972 and 1980, there was some surprisingly frank discussion about the electability of the eventual nominee:
Gov. James B. Hunt Jr. of North Carolina is chairman of the 69-member commission reviewing party nominating rules for the fourth time since 1969. He began the first regional hearing by saying that the goal was to give ordinary Democrats ''greater faith and confidence in the nominating process.''
Victory Is the Objective
''We're about the business of winning again,'' he said, in describing the objective of the commission, which is to present recommendations for action by the national committee early next year. (NYT, 9/25/81)
Gov. James B. Hunt Jr. of North Carolina, who heads the latest Democratic rule-changing group, an unwieldy, 29-member agglomeration of the innocent and the experienced, describes its task as one of writing ''rules that will help us choose a nominee who can win and who, having won, can govern effectively.'' The rules will probably matter less than the unemployment rate to a Democratic victory in 1984. But the comments underscore a traditional motive for the task of rule-changing the Democatc National Committee will finish in March. Much of this year's deliberations have seemed infused with a desire to deny future nominations to political reincarnations of the Jimmy Carter of 1976. (NYT, 1/27/82)
The concept was spawned at a meeting of party leaders after the Republicans scored smashing victories in the 1980 elections. ''There was a strong feeling,'' said one aide who attended the meeting, ''that to compete with the Republicans we had to reinvigorate our political apparatus.''
Representative Tony Coelho of California, chairman of the Democratic Congressional Campaign Committee, said: ''Elected officials don't control the convention, but they have a tremendous influence. They provide a balance; they put oil between the wheels.'' (NYT, 7/15/84)
Supporters of this plan and others argued that it would increase the connection of the rest of the party establishment to Presidential politics. Another argument for the proposal was that this element in a national convention, or just before one, could provide a potential nominee, deemed worthy and electable, with a confirming boost, or could check the progress of a potential nominee leading by a narrow margin but not esteemed. (NYT, 1/15/82)4. To check against a plurality, factional candidate who does not reflect the prevailing sentiment of the electorate.
At this point, we need to bring up another critical point of context regarding George McGovern. McGoven was really not the consensus choice of the electorate. Although he had earned 57% of the delegates, he had only 25% of the popular vote, in what was essentially a three-way tie with Hubert Humphrey and George Wallace (Humphrey, in fact, had a bare plurality of the popular vote):
How could this have happened? Recall that the primary system was something relatively new in 1972, and the rules were highly irregular from state to state. Some states, like California, were winner-take-all (and McGovern won California by a few percentage points), and some allocated delegates proportionally. Some states had primaries, and others had caucuses (as they still do). Some states received a lot of attention, and others received but a little. McGovern, by virute of his support in the grassroots, could compete everywhere. Undoubtedly, he also had some advantages because he understood the delegate selection rules, as he had helped to draft them.
Moreover, 1972 was a year with a remarkably weak field of candidates, and McCarthy was able to succeed because the rest of the votes were divided amongst his competitors. In a Gallup poll taken in June, 1972, McGovern drew the support of just 30% of the electorate, barely more than Humphrey (27%) or Wallace (25%). And McGovern was the sort of guy who either tended to be your first choice or your last choice. If there were any sort of instant-runoff voting, it is unlikely he would have won. (The same would also be true of George Wallace, so the nomination would likely have gone to Humphrey or Scoop Jackson).
Essentially, McGovern won the roughly 25% of the Democratic electorate that represented anti-war progressives, and Wallace won the 25% of Southern/segregationalist votes. The remaining 50% vote of mainline Democrats was split between several weak candidates, and so one of the factional candidates was able to win. It would be analogous to Dennis Kucinich prevailing over a weak field that included Evan Bayh, Tom Vilsack, Al Sharpton and Zell Miller.
So, rather than to override the will of the electorate, superdelegates were created in some sense to enforce it in elections whose results were skewed by odd delegate allocations or weak multi-way fields that would allow a fringe candidate to win a plurality. Both of these conditions prevailed in 1972; neither of them do in 2008. Let me cite a couple newspaper passages to this effect and then some concluding comments.
Our commission believes the future is potentially bright for our party system, but we are not inclined to take its durability and its health for granted. Accordingly, strengthening the party as a cohesive force in government and within the electorate has been a primary concern as we have recast the rules governing our nomination process.
We are convinced that the changes the party has made in its nomination procedures since 1968 have, for the most part, had a constructive effect. Secret caucuses, unpublicized procedures, closed slate-making, racial exclusion - such abuses, it is to be hoped, are behind us forever.
We propose to reduce the party's fragmentation and to increase the legitimacy of the nomination process by shortening the season and reducing the disproportionate impact of single early states.
We propose to make our convention more representative, both by maintaining a strong affirmative action program and by giving a greater role to those elected and party officials who speak for broad constituencies within the party.
Finally, the commission's recommendations would place a premium on coalition-building within the party prior to nomination and would promote a stronger party tie among our elected officials. (CPN committee report, via NYT, 3/27/82)
The concept of the uncommitted delegates, who will include up to three-fifths of all Democrats in both the House and the Senate, drew hostility last year. Initially, it stirred fears of smoke-filled rooms and boss-run conventions, like the 1968 Chicago convention that spurred the whole movement for change. That movement has brought forth a new set of rules after every Presidential election since, mostly aimed at increasing the influence of ordinary Democrats and decreasing that of party officials. Concept Gains AcceptanceNote that these two passages come directly from the Committee Report and its Chairperson, respectively.
Mr. Manatt, noting that no one spoke against the concept in the meeting, termed the proposal ''an idea that got more and more acceptance'' as Democrats decided they needed the officeholders' participation and that the high-ranking officials needed the freedom of not being tied to a candidate in advance.
Gov. James B. Hunt Jr. of North Carolina, chairman of the commission formed last July to draw up the proposals that were adopted today, insisted that despite the changes the Democrats would remain ''the only major political party in the nation that is truly open.'' He said the changes were necessary if the Democrats were to be controlled by the interests of their party as a whole and not its factions. (NYT, 3/27/82)
One last point that needs to be addressed. To our contemporary eyes, there might seem to be some tension between #3 (selecting the electable candidate), and #4 (selecting a candidate that represents some broad consensus of the voters). What if the Democratic voters had a slight preference for a candidate who was liable to perform slightly worse in the general election (in a pool that included independents and Republicans?). In 2008, with its climate of polarization and relatively unpopular party institutions, this tension seems quite palpable: the establishment candidate may not be the most electable one.
However, in 1972 or 1982 terms, points #3 and #4 are really one in the same thing. It was assumed, because of the example of McGovern, that an insurgent candidate supported by some bare plurality of activists would necessarily be unelectable. So to the extent that there are any threads about supporting the establishment candidate in the rationale for superdelegates, those threads are couched in an argument about electability -- and that argument no longer applies because the threads between establishment support and electability have been broken, at least in the presence of candidates like Barack Obama and John McCain.
Finally, it's worth reflecting on what's absent from the contemporaneous rationales behind superdelegates. There is not any sense that superdelegates have a mandate to express what is essentially their own personal preference for President; rather their duty is to look out for the best interests of the Party. Of course, the phrase "best interests of the Party" can mean different things to different people. But clearly, rejecting a candidate who was perceived to have received the majority of support from the voters in a two-person contest would be harmful to the Party, all else being equal. And so the superdelegates would have to have an extremely strong rationale to do something like that. The only plausible such rationale that seems to be present from the "founders' documents" would be to nominate the more electable candidate. However, because a candidate who was nominated in contradiction to the will of the pledged delegates would probably be rendered less electable by that process, there would need to be a very strong difference in electability for that to come into play.
As Chris Bowers writes, it is disingenuous to argue that the debates about superdelegates are debates about rules. There really aren't any rules about superdelegates, and to the extent there is any guidance on how superdelegates should behave, it is that they should tend to reflect the popular will of the voters.
Nobody is really arguing that superdelegates should be removed from the process, at least in the 2008 cycle. (Having studied the origin of superdelegates, my feeling is that they do have some purpose, particularly in multi-way nomination fights, but that their number should probably be reduced in future election cycles).
Rather -- let me just clip from Mr. Bowers, because he makes the case as succinctly as it can be made:
During the numerous discussions that have taken place on the subject of super delegates, the notion that super delegates can vote for whoever they wish is continually raised. Let me make the best counter-point to this argument as simple as possible:
DNC rules do not obligate super delegates to thwart the popular will of Democratic primary voters and caucus goers. Just as DNC rules allow for super delegates to thwart the popular will of Democratic primary voters and caucus goers, those same rules allow for super delegates to ratify the popular will of Democratic primary voters and caucus goers. Both are well within the rules. The decision is up to the super delegates.
The difference is that if super delegates decide to ratify the popular will of Democratic primary voters and caucus goers, then super delegates are upholding both the rules of the DNC and the principle of democracy. In other words, voting to thwart the popular will upholds our rules, but not our values, while voting to ratify the popular will upholds both our rules and our values.
Super delegates should uphold both our rules and our values by ratifying the popular will. That is as simple as I can make it.