The Harbinger Illusion
On Democracies: What we mistake for the new normal
In 1942, the low point of the twentieth-century democratic retreat, a careful observer counting functioning democracies on the planet would have arrived at roughly twelve. Britain. Canada. Australia. Switzerland. Sweden. Ireland. The United States. New Zealand. Costa Rica. Uruguay. A handful of others, depending on how one defined the term and on which exiles in London still counted. Most of Europe was occupied or fascist. Most of Asia was imperial Japan or its colonies. Most of Latin America was caudillo or junta. The Soviet sphere was its own category of horror. Democracy as a form of government had been reduced to a perimeter, and a shrinking one at that, held primarily by the English-speaking peoples and a few neutrals.
By the early 2000s, the same observer would have counted roughly eighty-nine democracies, depending on the methodology. Five times as many. A clear trend. A vector. To anyone looking at that line and projecting forward, the conclusion looks obvious: democratic government is the form that history is moving toward. A harbinger. A direction. A consummation.
My friend and fellow podcast host, Sean, pushed back on me in a recent conversation. I had said that what we have here, citizens electing leaders, peaceful transfers of power, constraints on the strongest among us, is anti-historical. It is not the human pattern. He countered: why couldn’t the last 250 years and the global expansion of representative government be the new normal? Why couldn’t we read the trend forward?
I have been thinking about his question for some time now. The trend he’s pointing to, I believe, is not actually a trend. We Americans see the world from inside a bubble we did not build and cannot see the edges of. The expansion of democracy is downstream of one specific event, and the survival of any given democracy is downstream of forces that nearly didn’t operate. We are inside a survivorship lens. The view it gives us is not history. It is the photograph at the end of a war the photographer happened to win.
Karl Popper had a name for what my friend was doing. In 1945, sitting out the war in New Zealand and watching the most educated civilization in Europe devour itself, Popper wrote about a particular intellectual error he called historicism. The historicist treats current trends as laws of history rather than contingent outcomes. He sees a vector and assumes a destiny. He sees what is happening and deduces what must happen. The phrase Popper most worried about was “the right side of history,” because it converts a hope into a verdict and a verdict into a license.
The harbinger argument is the soft version of historicism. It claims only that we have enough evidence to forecast a continued expansion. The claim looks empirical. It is doing the same epistemological work Popper diagnosed: treating a contingent run as a law.
The contingent run starts in 1945. The global expansion of democracy after the Second World War happened for a specific reason. The United States, with British and Soviet help, won the largest shooting war in human history. Then, with the Soviets now its adversary, it held a security and nuclear umbrella over Europe and Japan for seventy-five years. Under that umbrella, defeated fascist states were rebuilt as democracies. Decolonizing nations were pulled toward the democratic side of a bipolar world. The 1989 collapse of the Soviet bloc was the second-order effect of the same victory.
*We have only ever seen democracy expand inside the perimeter of one country’s military and economic dominance.*
The harbinger argument also has a measurement problem. It conflates the existence of elections with the substance of constitutional self-government. The expansion of “representative government in some flavor” includes a great many regimes that are technically electoral and functionally authoritarian. Russia holds elections. Hungary holds elections. Turkey, Venezuela, Nicaragua, El Salvador all hold elections. Political scientists call this *competitive authoritarianism*. These regimes wear the costume of democracy while hollowing out its working parts: the independent judiciary, the free press, the constraints on the executive, the peaceful transfer of power. Freedom House has now registered twenty consecutive years of net global declines in liberty, beginning in 2005. V-Dem and Polity show similar patterns. The variable that matters is whether anything resists the winner. An election is easy to stage; a constraint on the winner is hard to fake.
If you look at that number, the count of regimes with functioning constraint on power, the curve has been bending the wrong direction for almost two decades. The harbinger may be pointing somewhere we did not expect.
There is a deeper problem with reading the trend forward. We are looking at the democracies that survived. The ones that didn’t survive are not in the chart.
Weimar Germany held elections. The Spanish Second Republic held elections. Chile under Allende held elections. Each of those societies, in its moment, considered itself a stable democratic order. Each had a constitution, a parliament, a free press, an electorate. Each fell. The pre-fascist Italian liberal state, the Argentine democracy of the 1920s, the Pakistani democracy of the 1950s, the Greek democracy of the 1960s, the Brazilian democracy of that same decade, the Turkish democracies of multiple eras, the Venezuelan democracy hollowed out from inside under Chávez: each of these felt permanent until the moment it was not. The base rate of failure for the experiment looks very different when you count the failed cases alongside the survivors.
Some of those failures the umbrella’s holder helped break. The hegemon that protected democracy in Western Europe also helped overthrow it in Santiago in 1973, in Brasília in 1964, in Athens in 1967. The umbrella was not always benevolent. That is part of the harbinger illusion too: assuming the host always plays the host.
And the survivors are younger than we remember. Spain and Portugal both lived under Catholic, corporatist, fascist-adjacent authoritarian regimes until the mid-1970s. Franco died in 1975. Salazar’s Estado Novo collapsed the year before. Greece’s military junta fell in 1974. The picture of a stable democratic Europe in which I grew up was, in those countries, freshly painted, barely a decade old when I was born.
Spain is now warming to Beijing, deepening ties, treating the relationship as one that could become permanent. Angela Merkel made the same move with Russia: Nord Stream, energy dependency, the partnership treated as a marriage. You can do business with an authoritarian regime. You can have a working relationship. There is no marriage there.
The elementary objection is that the United States is also entangled with China. Both economies depend on the other. The objection misses an asymmetry the Athenians named at Melos. In 416 BCE, when Athens demanded that the neutral island either join its alliance or be destroyed, the Athenians made their case: “the strong do what they can and the weak suffer what they must.” Peers entangle and can exit. The weak get held. Spain tied to Beijing is Melos after the ships arrive.
The pattern is older still. The same Athens that demanded Melos submit was the original example of citizens governing themselves. It lasted roughly two centuries before its final extinction at the hands of Macedon. Inside its first century it nearly destroyed itself in the Peloponnesian War, which Thucydides described as a study in how confident, prosperous democracies dismantle their own deliberative capacity under stress. Word meanings shifted under factional pressure: “reckless audacity came to be considered the courage of a loyal ally; prudent hesitation, specious cowardice; moderation was held to be a cloak for unmanliness.” That was 427 BCE.
*The mechanism by which faction corrupts the language a republic needs to govern itself is older than Christianity.*
The Roman Republic ran longer, almost five centuries, before collapsing into Caesarism inside a single generation’s lifetime. Older Romans alive at Augustus’s accession in 27 BCE could remember the Senate functioning. By the time Plutarch wrote about the collapse a century later, those eyewitnesses were gone. Each step had looked, at the time, like a reasonable response to crisis.
*It is more comfortable to believe we live inside a trend than that we live inside a window.*
Two thousand four hundred years of evidence make a case for something other than the harbinger. They make a case for fragility. Self-government is the historical exception. Every functioning democracy has existed inside a finite window of conditions. Every one has eventually faced the test of whether it could maintain itself. Most have failed. The ones that remain are the ones that, so far, have not.
The chart of expanding democracy shows something narrower than history bending toward freedom. It shows that for a brief period after a particular war, an unusually powerful country protected the conditions under which freedom could spread. We are in danger of confusing that period with a permanent feature of the human condition. The error runs deeper than the counting. It is psychological.
It would be easier to leave the argument here and blame the conditions. Lose the umbrella, lose the democracies. But Thucydides documented something the umbrella story cannot reach. Pericles, he wrote, “by his rank, ability, and known integrity, was enabled to exercise an independent control over the multitude — in short, to lead them instead of being led by them.” When Pericles died, his successors, “more on a level with one another, and each grasping at supremacy, they ended by committing even the conduct of state affairs to the whims of the multitude.”
Athens still had its institutions. What it lost was the kind of leadership that could contradict the demos, the citizen body, rather than flatter it. That loss is what made the Sicilian Expedition possible: the 415 BCE invasion the assembly approved against the cautious advice of its own generals, which two years later destroyed most of the Athenian army and navy in the harbor of Syracuse. That loss produced the oligarchic coup of 411 BCE, the assembly’s later execution of its own commanders after Arginusae in 406, and the surrender to Sparta in 404.
We cannot only blame the absence of an external umbrella. We also have to ask whether we still possess the internal practice, the leadership willing to contradict, the citizens willing to be contradicted, that maintains the experiment from inside.
When Sean asks why we cannot just project the expansion forward, the answer has several layers. The expansion is downstream of one war. The chart is missing the failures. The trend is already bending the other way on the variable that matters. And inside any given democracy, including ours, the question is whether the people inside the institutions still possess the practice that keeps the institutions alive.
The founders knew the experiment was fragile. Alexander Hamilton opened the Federalist Papers by asking whether human societies were “really capable or not of establishing good government from reflection and choice, or whether they are forever destined to depend for their political constitutions on accident and force.” He was not asking rhetorically. David McCullough closes his account of the year the Revolution nearly ended with the line that, to those who had lived through it, the outcome “seemed little short of a miracle.” That is the right frame for the inheritance. Not a trend. A miracle.
Confidence that it will obviously continue tells us nothing about whether it will. It is the luxury good that wealth and peace produce, and the first thing to go when they stop.
Treating the rare thing as normal is how it becomes rare again.
We are still inside the photograph. We have forgotten that it is one, and that the photographer could have lost.
Sources and Inspiration
The Open Society and Its Enemies - Karl Popper
History of the Peloponnesian War - Thucydides
The Federalist Papers - Alexander Hamilton, James Madison, John Jay