National party conventions are one of the great informal institutions in American—and arguably world—politics. First adopted by the short-lived Anti-Masonic Party in Baltimore in 1831, the original convention template has survived in the same essential form into 2020. This hardly means that there have not been noteworthy adjustments within this template. Indeed, these often serve as important registers of, and modest reciprocal contributions to, major changes in the larger American politics around them. So the facts behind both assertions—a stable institution coupled with internal updates—get reliably dusted off as each new convention year arrives. And then came the COVID-19 pandemic. If it was hard to escape the standard convention chronicle, it became harder to avoid the arguments about whether an electronic alternative, a “virtual convention,” was on the horizon, and what it would mean if it was.

The original convention design was a response to pressing practical problems, involving the construction of a lasting political party in a geographically dispersed nation with little in the way of inter-state communication. So the convention device became a way, and ultimately the leading way, to pull regional and local branches of an aspiringly national party together, while allowing it to create a single presidential nominee. Copied in 1832 by the Democrats, eight years later by their main opposition, the Whigs, and then by the nascent Republicans in 1856, the device has ever since been the means of confirming major-party nominees for president. In itself, this is a remarkable story of institutional continuity, for a device that is not part of the formal framework of government but has become central to the way that framework operates.

Now-classical components of a convention followed, and these have been essentially stable as well, so stable that they create an implicit formula in the heads of the major players. Under this formula, delegates to the national party convention are created in the states, through state chapters of the national party. They assemble centrally on a given date to adopt rules of procedure, certify the credentials of other delegates, and offer a fresh interpretation—a “platform”—for an evolving party program. And with all of that accomplished, usually in a pro forma fashion but sometimes after extended disputation, the convention moves on to crown the presidential nominee and shift the party focus to a general election campaign.

Such conventions have never failed to assemble in the designated place at the designated time, though the Democrats have had some close calls. The Democratic convention of 1860, an important prelude to the Civil War, met, recessed, reconvened, and then splintered into three proto-parties, each with its own presidential nominee. Conversely, the Democratic convention of 1872 assembled for only six hours and endorsed the nominee of another proto-party, the short-lived Liberal Republicans, in that sense never achieving a separate nomination of its own. And the Democratic convention of 1924 went for 17 days and 103 ballots before finding its essential compromise. It is no accident that the Democrats were the party achieving all of these problematic milestones, since they required a two-thirds majority from their founding until 1936, while the Whigs and then the Republicans stuck with a simple majority throughout.

Otherwise, a set of more modern adjustments characterized the convention form as it returned for 2020. The biggest was the informal disappearance of actual nomination-making: The two contests of 1952 would be the last in American history to require as much as a second ballot before the nominee was confirmed. This exile of the effective nomination was then formalized after 1968, when a reformed process for delegate selection tied delegates to presidential contenders and ran the nominating bandwagon through a sequence of effective—and cascading—presidential primaries. Along the way, these reforms changed the social identity of the delegates themselves. Previously, they were party officials and public office-holders. Afterward, they were issue activists whose principle loyalties were to a candidate, a cause, or a group, and not to the political party.

Many other things changed in tandem with those shifts. When the function of the convention was no longer to nominate a president, its main function became instead public relations and publicity. The nominee was introduced and publicized of course; for many people, this constituted their first actual introduction to a specific individual. Yet many other political figures—and not just the other, losing aspirants—could hope to use podium time to advance their own causes, constituencies, or careers. So a subset of these struggles for podium time involved not just the question of who could secure it, but when such appearances would occur, that is, whether they brought any hope of being covered by the decreasing window of podium coverage from national television networks.

That was the orthodox modern world when the Republicans decided to go to Charlotte and the Democrats to Milwaukee in order to realize these ongoing possibilities in 2020. And then came COVID-19, so suddenly and with such force that it brought into question every aspect of this long institutional evolution and its concrete contemporary applications. Two fresh questions became especially insistent. First, what are the operational implications of an international pandemic for the national party conventions of 2020? And second, what will the chosen options say about the convention template and in its links to the larger society around it?

In the simplest terms, the impact of coronavirus on convention planning roughly followed the trajectory of its impact on public health, though relevant concerns and calculations had to differ additionally by party. By tradition, the incumbent party goes last, and the challenging party goes first. This allows the party in office to concentrate its resources for a short but intense fall campaign, just as it allows the party out of office, lacking a leadership record and ordinarily fronting a less well-known nominee, to begin work early. Tradition was only exaggerated in 2020, when the Republicans planned for the last week in August, the final week before the traditional Labor Day kick-off to the fall campaign, while the Democrats pulled away in the other direction, back to the second week in July. As is often the case in modern American politics, this placement was reinforced by a desire not to conflict in a major way with the 2020 Olympics, with their potential to dominate both coverage and audience.

So the Democrats were fated to face the earlier decisions about what, if anything, a pandemic implied for the conduct of their convention. Generically, convention planning is a three-cornered negotiation, involving a subcommittee of the party’s national committee, a host committee from the chosen city, and the staff of the expected nominee—easy for the Republicans in 2020, harder for the Democrats. In its details, convention planning becomes more or less a two year process, first forcing interested cities to demonstrate a suitable hall for the main event, sufficient housing for delegations, guests, and the media, a diverse array of meeting rooms and additional venues, plus a focused transportation system. The winner must then go on to link these all-encompassing needs together in some master arrangement.

Yet having answered those questions and moved toward application in a pandemic-free environment, what were the options for this Democratic discussion as COVID-19 became rapidly more intrusive? Among Democratic discussants, the first response was much like the initial response of the national government: Let’s stay aware of the potential threat but hope that it will go away. Yet as the pandemic blew up with impressive speed in the larger society during March, so did anxieties, local and national, about what to do with the convention. The three corners of the discussion were careful to say little in public about the threat to an orthodox convention on the established calendar, but the options were simple and stark:

• Conduct the convention as planned, the preference of all major players on both sides of the partisan aisle but increasingly implausible for the Democrats, while remaining more open-ended for the Republicans;
• Move the Democratic date farther back into the summer, though the logistics of such a move were enormous, since they entailed giving up halls, hotels, venues, and facilities, then reclaiming them all in the same place at a later date;
• Move the locale of the convention if, as seemed likely, a comprehensive move in the same place proved impossible, though this would entail condensing what was inherently a two-year planning process into a couple of months somewhere else; or
• Figure out how to hold a “virtual convention.” Everyone could talk about it; no one was enthusiastic about it; and no one knew what it meant.

In extremely short order, the pandemic removed the first option, even as it remained the abstract preference of the major players. Though recognizing the new reality was at least equally facilitated by the sharp and condensed rise of Joe Biden as the apparently inevitable nominee. So the Biden camp could take a leadership role in ongoing three-cornered discussions, and Biden himself went on to speculate aloud about the impossibility of July in Milwaukee for a gathering of somewhere north of 30,000 participants in a one-block space. The third option, a different time in a different city, was always trumped by the second, a different time in the same city, if it could actually be accomplished. To the surprise of most observers, it could, and the Democratic convention announced a new time for 2020, still in Milwaukee but now in the week immediately before the Republican convention.

Yet the fourth option, the virtual convention, was by then receiving serious attention as well, not just by the established players but now, quite centrally, by the media that would have to cover such a thing. Examples of a virtual gathering were readily available; educational institutions and business organizations were already doing their version of it. On the other hand, the mechanics of its operation in the convention context could be known only as and when it was actually implemented: The details were every bit as seat-of-the-pants as those guiding the Anti-Masonic convention of 1831. Yet if this was truly uncharted territory, it was clear that the pandemic might well compel it. So someone, and presumably lots of “someones,” would have to think about: 1) What would be gained by sustaining the convention through a virtual format?; 2) What would be lost in the absence of the usual participatory version?; and 3) What would these answers say about the evolution of American politics more generally?

By definition, the major gains derived from salvaging something rather than nothing from the usual convention products, though the scale of these contributions was hard to estimate in the abstract, that is, in the absence of any virtual experience. At the top of the list was the fact that the convention serves as the first real introduction of two major-party contenders for president of the United States for a sizable minority of the general public. Ordinarily, there would have been an extended nominating campaign beforehand, providing this introduction for another sizable minority. Yet this year, much of that had evanesced after Super Tuesday, as primaries were effectively uncontested and campaigns were confined to the candidate’s home studio. So that role, even of a virtual institution, might be more important than usual. Seen from the other side, this is also ordinarily the first major opportunity for the nominee to introduce himself as a potential president, in the fashion that he wishes to be seen, that is, without alternative interpretations from internal party opponents or the external news media.

And the impact is always a public relations bounce for the nominee, with only the rarest of exceptions. The size of this particular gain can vary substantially: The fall campaign would become the ultimate answer to the question of whether the bounce can be sustained, though this would now be tied directly to how much of a virtual convention was available to the public, and how much that public watched. At the same time, this is effectively the only opportunity for each of the major parties to present themselves to the general public in a collective physical portrait. Beyond that, the convention is an opportunity to hammer the themes that ostensibly explain why you should be supporting not just this particular nominee but this larger party, and why you should not be supporting the other.

Nominees and party leaders might even hope for some fresh advantages from the virtual format. If no one assembles as a body at the convention site, for example, and if the convention thus survives essentially as a central feed for the mass media—who themselves no longer need to be present on anywhere near the scale of the past—then much of the residual conflict that continued to inhabit actual conventions might disappear as well. Imagine the Clinton nomination of 2016 in the physical absence of any Sanders delegates. Their votes would still get tallied electronically, from home, but we would already know that outcome. And how would anyone even generate rules, credentials, or platform challenges, now that supporters could not roam from delegation to delegation seeking support? All of that might appeal to the party and its nominee.

Which leads to the other side of the same coin: What (or what else) would be lost in the process of going virtual? Because this involves items from an orthodox convention which recur regularly, it is possible to be less speculative here. Thus national party conventions are also major sites for state party politics, being conducted for a week in the belly of the national party. A few states actually have gubernatorial races the following off-year; many have Senate races; all have House contests. So gubernatorial, Senate, and House candidates can (and do) meet, greet, and build links not only to local politicos across their state, but to interested persons from outside state boundaries, often in the form of donors, often in the form of issue activists. To that end, state parties normally begin their day with a delegation breakfast, introducing not just state-based contenders, actual and potential, but visiting party leaders from around the country, supporting the candidate and, once again, themselves. During the day, these state parties can go so far as to assemble on an otherwise empty convention floor, to film state-based candidates speaking to an enthusiastic (albeit very narrowly photographed) audience.

And there is also always a welter of introductions, presentations, and rallies by interest groups of all sorts in the national convention environs: public land users, Hispanic women, tax reformers, marijuana partisans, school lunch providers, and on and on. A full calendar of such events reliably dwarfs the official calendar of podium presentations, even when the latter includes all of those that do not occur during prime coverage hours. Indeed, those of us who regularly attend these quadrennial gatherings make a kind of parlor game out of seeing whether we can spot rising stars or rising causes from within this confused conglomerate of efforts at political attention. And on top of all this, there is always a great deal of (at least attempted) money-raising, along with an even greater deal of contact-making and coalition-forming.

Finally, then, what would all of this contribute to—and reveal about—a larger American politics? On its own terms, the virtual convention may be the next logical evolutionary step for presidential nominations. Conventions, able to construct delegate majorities rather than just confirming them, made these nominations from 1832 through 1952. After that, the nomination exited the convention and moved into the sequence of presidential primaries plus the occasional state caucus that sprang up in the aftermath of sweeping reform of the delegate selection process. What remained in the hall was a struggle over utilizing press attention, either to launch the general election campaign or to support various alternative candidates, constituencies, and causes. Political parties much preferred the former over the latter.

Yet as the story got closer to the current moment, parties became more and more successful at containing even these public relations struggles, while the mass media responded by decreasing the scale of their coverage of what were viewed as decreasingly suspenseful contests, focused largely on harvesting the very media attention that was being withdrawn. So there were already, within both parties and over the last several conventions, serious private conversations on whether the convention should be notably shortened, based on the implicit premise that there are just not four-plus days of “news” left. Such a shortening would allow (and force) the parties to focus actively on the central themes that their nominee wants to present, along with their own description of what it means to be a Democrat or a Republican.

In some sense, then, the virtual convention may just be forcing the issue and driving toward those moves. Though in the end, if this is a potential turning-point for a longstanding but informal institution, the devil may still be in the details. A virtual convention—or two—that appear to have worked reasonably well can be very different from one (or two) that are largely a catalog of unintended consequences. The former, a benevolent virtual (non)gathering, will at least bring those private conversations about convention fortunes into the explicit debate. The latter, unforeseen consequences of producing diverse but widespread disappointments should more or less automatically produce nostalgia for a long-lived but not necessarily well-loved institution. And the next three months may well tell that story.