An election is supposed to be a reality check. It promises the finality of decision, in which the back-and-forth of political argument gives way to the undeniability of a particular outcome on a particular day.

Scoff at the possibility that Donald Trump could ever be the Republican nominee (as I did, once upon a time), and a succession of primaries will reveal the hollowness of your supposed expertise. Insist on the certainty of polling or political science, or some variable (rally size, yard signs, boaters) of your own choice, and the election result will put your certainty to a decisive test. Claim that your particular obsessions are shared by the American people, and on Election Day the people in their majesty will render a verdict on your claim.

But that finality is still socially and politically constructed. And democracies can fail — a scenario on many people's minds these days — when that constructedness dissolves, when it becomes possible to deny the finality of election results outright, to continue the contest outside the system or to substitute a different form of decision for the verdict of the ballot box.

We are not there yet in America, but people are right to sense that we're in a liminal place, where a combination of factors has made our election results much less decisive than in the country's past.

One factor is the increasingly immersive power of ideological narratives and virtual realities. If you can react to an election loss by retreating immediately into a storyscape where the outcome was a cheat, carried about by means of voter fraud or Russian interference, then the decisiveness of any given outcome will inevitably diminish.

Another factor is the interaction of political polarization with America's two-party system and constitutional design. A country with two parties that are increasingly ideologically consolidated, and a narrowing band of swing voters in between, will produce fewer landslides and more nail-biters, and more swings back and forth from election to election, than a country with looser and more fluid coalitions. If that country's electoral system also allows candidates to win the nation's highest office with a minority of the popular vote, then under polarized conditions this scenario will become more commonplace, decoupling the official decision of the election from the apparent preferences of the voting majority.

The irony is that historically, America's Electoral College tended to produce more decisive-seeming outcomes, both because it magnified the scale of a geographically well-distributed victory and because the possibility of losing even with a slight popular majority created incentives to seek supermajorities — to overwhelm countermajoritarian redoubts with nationwide landslides.

You can think of the 2020 campaign as a test of whether that kind of incentive structure can be made to work again. The 2016 election, as I wrote over the weekend, was an example of a shocking but non-decisive-seeming result, in which all the weirdnesses associated with Trump's minoritarian victory gave people in both parties reasons not to learn any lessons from the experience.

But Joe Biden's campaign, with its steady "moderate blue-collar Joe" energy and disdain for Twitter fads, seemed to learn more lessons than the campaigns of rival Democrats. If Biden's push to the nomination and now the presidency was uncreative and unambitious in all kinds of ways, it still seemed to go further than any candidacy since Barack Obama's in 2008 in trying to be rhetorically unifying, in seeking a broader-than-last-time coalition.

The Trump presidency and campaign, meanwhile, appeared to write off the possibility of winning a landslide or even a simple popular-vote majority, relying on their Electoral College advantages while conspicuously resisting the Democratic Party's attempts to count as many votes as possible. The scenario in which Trump's 2016 campaign-trail populism coalesced into a majority-building agenda was foreclosed early; the scenario in which Trump rode a strong economy to an easy victory was foreclosed by the coronavirus. Which has left an indecisive outcome and yet another minoritarian government as his best 2020 hope.

From a systemic perspective, from the point of view of America's potential governability, even those of us who would oppose much about a Biden presidency could draw something hopeful from the seeming possibility — as of 48 hours ago — of a landslide Democratic win. That would mean that the system can still deliver clear decisions, even if only for one cycle, and it would create incentives for whatever follows Trump in the Republican Party to seek a similarly decisive victory in its turn.

But that possibility has evaporated as of this writing, taking what remains of the expert class's credibility with it. Instead at the moment we seem headed toward the indecision of another close and potentially litigated endgame, with razor-thin margins no matter which man wins. Which implies that even under what seemed like favorable conditions for a decisive outcome, the evolving ways we're polarized — by age, geography, sex, class and faith — are still the more powerful forces, dragging us ever back toward stalemate.

It's one more sign, as if we needed more, that our system's method of uniting power with legitimacy is slowly losing both.