There is a standard two-step to democratic backsliding, and if you squint you can see it attempted almost everywhere at the moment. Step one: a leader announces that he — or she, or the movement — personally embodies the will of the people. Step two: because the will of the people is sacred and presently blocked by judges, parliaments, broadcasters and bureaucrats, the leader must be given sweeping powers to carry it out. The two steps depend on each other. The power grab is justified by the popular mandate; the popular mandate is demonstrated by the fact that, well, who can prove otherwise? The beauty of the trick is that the people’s will is a rumour that only the leader is qualified to interpret.
Joseph de Weck, a Swiss writer and associate fellow with the German Council on Foreign Relations, has a suggestion for spoiling the trick, laid out in an essay in the Guardian this week: just ask the people. Constantly. About everything. In Switzerland they do, and the awkward result, he argues, is that the far right keeps losing. Dismantle the publicly funded broadcaster by halving its budget? Sixty-two per cent of Swiss voters said no in March, to an initiative the far right had championed. Cut corporate taxes? In 2017, 59% voted that down. Automatically deport foreign nationals convicted of crimes, no judicial review? Also 59% against. And this week, 70% — a landslide by referendum standards — rejected a far-right proposal to write neutrality so tightly into the constitution that Bern would have had to drop its sanctions on Russia over Ukraine.
The point is not that Switzerland is some social-democratic idyll. It very much is not. Leftwing parties won 29% of the vote in the most recent national elections. Women got the vote in 1971. The asylum law is among the toughest in Europe — and that, too, de Weck notes, is a product of direct democracy. The Swiss electorate is not progressive; it is, if anything, famously conservative with a small c. The point is narrower and stranger: when you actually poll the demos, forensically and repeatedly, the far right’s claim to speak for it keeps failing the arithmetic. The movement dominates social media, its positions are amplified across the press, and then the actual votes come in at 59, 62, 70% against. De Weck’s phrase is that the far right “thrives on imagined majorities,” and referendums are the machine for checking the imagination against the count.
And there is a second mechanism, aimed at step two of the confidence trick. Centralising power — the thing everyone from Donald Trump to Giorgia Meloni is accused of wanting — requires capturing a small number of capturable chokepoints: a parliament, some courts, a party apparatus. Parliamentarians can be leaned on, bought, primaried or frightened into voting away their own prerogatives; it happens. Persuading several million voters to vote away theirs is a harder logistics problem. Direct democracy, in this reading, is less a policy tool than a load-bearing wall. You can storm a building; it is difficult to storm a habit.
De Weck reaches for Hannah Arendt here, and the choice is telling. Arendt, who fled Nazi Germany in 1933, spent much of her career on the question of why democracies abolish themselves — why, specifically, Weimar failed. Part of her answer was that parliamentary systems can ossify into rule by party elites and bureaucracy, and that autocrats feed on the feeling of disenfranchisement that produces. Her remedy was participation: citizen involvement of the town-hall kind. De Weck points out that the demand is still visibly unmet — the gilets jaunes protests in France in 2018 were, at bottom, a plea to be seen and heard, which is why one of their central calls was for citizens’ referendums. In a 2017 poll, 74% of Germans, French and Spaniards said governing by public referendum sounded like a good idea. The people, in other words, would quite like the job. The people who currently have it are less enthusiastic.
Now, the objections, because they are the fun part. Aren’t referendums a high-risk way to run a country? Brexit, anyone? De Weck concedes the near-misses: in June, Swiss voters rejected — but only by 55% — a far-right proposal to cap the population at 10 million and tear up the EU–Switzerland economic agreements. The Swiss defence is that no vote is ever final. Since 2000 the country has voted on its relationship with Brussels more than a dozen times, revisiting and revising as facts change. Where Britain treated one referendum as a cliff it had to jump off because jumping was the will of the people, Switzerland treats each vote as a pencil sketch. Voters, it turns out, are much better than politicians at admitting yesterday’s answer no longer fits; there is no flip-flop charge when the demos changes its mind, because the demos cannot be primaried.
Wouldn’t the voters, handed the budget, do something ruinous — vote themselves more money and less work? The Swiss have been tested: two-thirds of them rejected a proposal to extend annual holidays by two weeks. De Weck calls this a tragic case of excessive responsibility. As every parent knows, he says, responsibility breeds responsibility. There may be something genuinely counterintuitive here: a public that is routinely asked to decide behaves, in aggregate, like an institution with something to lose.
The essay’s last move is the ambitious one, and it is worth stating plainly because it inverts the usual direction of travel. De Weck wants power transferred from nation states to citizens — and, simultaneously, from nation states to Europe. Nation states, he argues, now cosplay sovereignty more than they exercise it; the actual leverage over modern life sits in Silicon Valley, Washington and Beijing, and none of it is reachable by any single European country acting alone. But ceding national sovereignty is a miserable sell on its own — it polls like a holiday cut. Pair it with the referendum mechanism, though, and the offer flips: give up some sovereignty as a citizen of a nation, gain real decision-making power as a citizen of a continent. He points to Switzerland itself as proof the trick can scale: in a country with four linguistic communities, referendums are what manufacture the shared public sphere. Being governed together is not enough; you build the identity by arguing about the same questions at the same time. His imagined European ballot papers would run to migration policy, the 2035 ban on new petrol and diesel cars, taxes on the big digital platforms. Utopian? “Perhaps. But every democratic advance once was.”
One shouldn’t oversell it. Referendums are only as good as the franchise, the question-wording and the campaign finance around them, and Switzerland is a small, rich, cohesive country with centuries of cantonal practice at this. The machinery may not export. But the underlying observation is a clean one: the two-step that opens the essay depends on the people’s voice being a rumour. The simplest way to kill a rumour is to take attendance.