Don Moynihan, a Georgetown public policy professor and public commentator, recently referred in The Bulwark to political scientist Adam Przeworski’s dictum that “democracy is a system in which parties lose elections.” This sentence – and the argument behind it – has shaped how many political scientists think about the problems of U.S. politics, especially after the events of January 6, 2021. Despite its simplicity, it’s arguably one of the most important ideas that political science has produced. But few people outside political science would recognize Przeworski’s name, or understand why these few words matter. Here’s what he said – and why it’s important.
Przeworski’s big idea didn’t appear, as many ideas do, in an academic article. Instead, it was the first chapter of an otherwise very good but now outdated book on a more specialized topic – the transition of countries from authoritarian state socialism to democracy. Democracy and the Market was highly relevant when it was published in the 1990s – but few people who are not specialists have much reason to read beyond the beginning. It is that chapter that lays out a much bigger and much more general argument that has shaped how comparative political scientists think about democracy ever since.
And this argument has currency today. Przeworski was asking questions about the likelihood that new democracies would become more stable, but his answers turned out to have surprising – and unfortunate applications to older, seemingly well-established democracies like the U.S. that have gotten into trouble.
What Przeworski argues is that democracy is more fragile to being damaged by certain words and actions than we think. Until quite recently most people didn’t see well-established democracy as fragile at all. What Przeworski did, way back in the 1990s, was to show how all democracies were in principle profoundly fragile. Their survival depends on the willingness of powerful groups to continue to behave in somewhat surprising ways.
The fundamental puzzle of democracy, according to Przeworski, is why the losers accept that they have lost. That is why the “parties lose elections” bit matters – it highlights an important puzzle. Why do losing parties accept the verdict of the voters, delivered through some imperfect set of institutions? Until recently, in the U.S., we’ve usually taken this for granted. But if you think about it objectively, it’s quite surprising that parties and politicians behave in this way. Politicians – more even than most people – like having power. Elections give one group of people power, while relegating another to the sidelines, likely for years, forcing them to accept the government’s choices, like it or not.
So why are losers willing to put up with this? This is particularly perplexing when it is the reigning government that accepts being booted out of office – after all, they have apparent control over the engines of government, the military, and so on.
Przeworski’s answer (he is a game theorist, and impatient with arguments based on norms, how politicians admire the sublime awesomeness of democracy, and similar notions) is that losers accept the loss of power if and when they have good, self-interested reasons to do so. They decide that they are better off living with the verdict of the system, however painful it may be, than challenging it.
One of their key motivations is that democracy, when it is working well, combines institutional certainty with uncertain outcomes. The rules are predictable. Everyone knows that there will be another election in a few years. But the outcomes of future elections are impossible to predict. Perhaps today’s losers will win tomorrow.
That gives the losers some incentive to accept their loss this time around, in the hope that things will go differently in the future. And that, in turn, helps underpin democratic stability. As Przeworski puts it, “Democracy is consolidated when … no one can imagine acting outside the democratic institutions, when all the losers want to do is to try again within the same conditions under which they lost.”
Przeworski’s work inspired a lot of political scientists to use game theory to determine the conditions under which democracy was “self enforcing”: that is, how everyone’s beliefs and actions might line up to make democracy a self-fulfilling prophecy. People assumed that consolidated democracies would stay consolidated – that the losers would stay inside the rules. But his argument powerfully suggests a theory of democratic fragility, too. What happens when the losers of elections – or other powerful and organized political forces – decide not to comply with the system but to challenge it?
Przeworski identifies two forms of non-compliance. First, some powerful organized force, such as a political party, may look to overturn democratic outcomes. Second, such a force may look to “drastically reduce the confidence of other actors in democratic institutions.” Although he doesn’t dwell on this, these two possibilities imply that under certain circumstances, democracy can become self-unraveling rather than self-enforcing. If you (as say the leader of the Republican party) look to overturn an election result through encouraging your supporters to invade the U.S. Capitol, and claim that the election was a con, then I (as a Democratic party leader) am plausibly going to guess that my chances of ever getting elected again will shrivel into nonexistence if you gain political power again and are able to rig the system. That may lead me to be less willing to play by the rules, leading to further collapse of confidence on your part and so on, in a downward spiral.
That – of course – is the bind that the U.S. is in today. It’s a somewhat less pressing worry in September 2023 than it was in September 2022. There is circumstantial evidence that the threat to democracy caused a counter-reaction in places like Arizona and Michigan. But the emphasis is on the word “somewhat.” Donald Trump is still the unofficial leader of the Republican party, and no realistic candidate for the Republican nomination is willing to disown his lies about the election.
That is why Przeworski’s apparently abstract mathematical theorizing has consequences for understanding the situation of U.S. politics today. If you take Przeworski’s arguments seriously, then you are likely to believe that democracy is fundamentally threatened when major political forces claim that an (apparently perfectly fair) election was a fraud, and look to press their argument through violence. Furthermore, organizing against democracy on the notional claim of fraud may have profoundly different consequences from complaining (rightly or wrongly) about fraud while accepting the fundamentals of the democratic system that you are stuck with, looking to work within it and transform it, rather than to overturn it.
Przeworski is not the only political scientist writing about democratic stability. In political science there is a very rough divide between comparativist political scientists, who study other countries and tend to worry deeply about the future of U.S. democracy, and Americanist political scientists who tend to be less concerned. If you think that democracy is likely to be stable in long-established democracies like the U.S., then you have little reason to be concerned. If, alternatively, you believe that democracy can unravel in the ways that Przeworski implies, then there are plenty of reasons to be worried.