What’s going on in the UK right now is already old hat: Scotland possibly planning another referendum, the young feeling disenfranchised by a ‘democratic’ vote, and MPs scratching their heads on how to eventually enact something the vast majority amongst them never wanted, all based on a majority of a few percentage points, and on a quite divided demographic.

But focussing on the UK’s current predicament ignores something vastly more important: Europe’s ‘insurgent’ parties, the ones running on a mandate, for better or worse, to disrupt the status quo, are winning not just single battles like Brexit, but a much larger war, and one that is much more dangerous than simply a challenge to any status quo can be. The reason is that, in the Quebec, Scottish, and Brexit referendums,  we all have allowed a creeping and dangerous misunderstanding of the workings of a functional direct democracy. Today, it sounds like heresy, but let me make the case nonetheless: A stable democracy cannot always give in to whatever the majority wants at any given time.

In fact, these problems were already clear to Athenians in the 5th century BC. Athens had a very direct democracy: Although many were excluded from participating, the ‘citizens’ shared very equally in their responsibilities, with terms of office being set so short, and so rarely repeatable, that almost everyone who wanted to hold office did so during their lifetime. Accountability when in office was extreme: one was supposed to coordinate and run the system, not to influence it, and penalties for doing otherwise were severe. But importantly, the systems lacked some of the constitutional checks and balances which forced the ‘demos’ (the voting public, the citizens) to stick to its own rules and laws. The ‘rule of law’ could be undone by a single democratic majority decision, as was the case in 406 BC, when the Athenians at long last had won a naval victory against the Spartans. After the battle, a storm rose and the generals in command failed to collect survivors: the Athenians sentenced six of the eight generals collectively to death. Technically, this was illegal, as the generals were not tried and sentenced separately, one by one, as Athenian law required. Socrateshappened to preside over the assembly that day and refused to cooperate, to no avail. He stood against the idea that the people could decide whatever the hell they wanted, and for the rule of law. Later, the demos is reported to have regretted what happened. They decided that those who had misled the demos should be charged and put on trial, including the author of the motion by which the generals were tried and condemned en masse. This passage is often interpreted as a confession of collective regret and guilt on the part of the demos, once their anger gave way to second thoughts.

Parallels to the UK’s Brexit referendum are not entirely coincidental.

But while the demos regretted their earlier decision, they made yet another disastrous one in 399 BC: Rather than giving Socrates credit for having foreseen the limits of a democracy without constitutional and institutional checks and bounds, seven years later the demos condemned Socrates to death for corrupting the young. Plato, using the killing of Socrates as an example, criticised heavily the construct of the Athenian direct democracy, mainly on the grounds that it gives everyone the same vote and works on the principle of a simple majority even in important and irreversible matters.

Again, parallels to the Brexit aftermath seem less than concidental.

The West had long previously moved beyond the Athenian model of democracy, into a ‘rule of law’ which even the officeholders are subject to, and into a model where institutions (partly non-elected) are tasked with ensuring that decisions aren’t made rashly by those who happen to be in power (even if it’s the ‘demos’). New laws are checked on their compliance with the constitution, by usually un-elected judges. The constitution requires super-majorities to amend, and is almost impossible (usually) to change. Even Switzerland, whose model of direct democracy is closest to that of ancient Athens, requires a double majority for constitutional changes (a majority of both the people and the cantons).

In short, even if the UK had had Switzerland’s model of direct democracy, the Brexit votes on Thursday’s count would have led to a ‘Remain’ result. When the stakes are very high, the default in a democracy should be ‘no change’.

The point is, very simply, that democracies had already learnt to implement safeguards protecting against spur-of-the-moment simple majority decisions of either momentous impact, great cost, or a high degree of irreversibility.

Europe’s 47 insurgent parties on the other hand, from Catalan separatists, to the Northern League, UKIP, SNP, etc. seem to have established a new and — strangely — a suddenly widely accepted principle: that if any majority at all can be found at any time at all, they get what they want, no matter how big or how irreversible the impact.

The Quebecois voted on the same premise, and the Trump followers also believe — rightly or wrongly — that they’ll get a  lot of unconstitutional promises delivered if they win by only as much as a narrow majority. One can only hope that America’s constitution will remain stronger than its president, and I am certain it will. The new and democratic stability-defying logic seems to have crept in for one-off referendums only, at least for now.

This is not how modern, consititutional democracies were designed to function, and even Boris Johnson knows it and is now uneasy in the position he suddenly finds himself in. The Brexit vote is but a tiny taster of the chaos that can ensue if we run decisions by simple yes/no majority votes, without consideration of the diverse spectrum and the often limited temporal persistence of political views. It’s like stating the obvious to some, and yet highly confusing to others: The Brits cannot be condemned to follow a single in-out referendum by a narrow majority. They must demand the right to change their minds. Or not to change them, as the case may be.

Given that the process of disentangling the UK from the EU will take more than two years for sure, there is still room and time to check whether this is what the population really, consistently wants over time, or whether this was a one-off result skewed by un-deliverable campaign promises and misguided tactical voters. Whether the British establishment will have the courage to test this question again some time in the next years remains to be seen.

But a lesson has to be learned. We cannot make the same mistakes again and  again. The hurdles for potentially costly, irreversible decisions have to be higher, either in percentage terms, or in terms of a “persistence of public opinion”-test, or both. Otherwise, not on only the UK, but other European countries and Europe itself threaten to become the new Yugoslavia. Even if you belong to one of those countless separatists all over the European map, you have to accept that there have to be high hurdles before you can get what you want. In referendum rules, it is only fair to ask if the costs are worth it, and sometimes they are only worth it if substantially more than 50% really, really believe in the change, and do so persistently rather than on one single referendum date.

In the same way in which virtually every country requires a super-majority for constitutional changes, countries (or rather their parliaments) have to raise the bar for momentous, irreversible decisions, even if they are made by popular vote.

We cannot have votes dangling around 50:50 about the secession of Catalonia, Nothern Ireland, or Northern Italy from their nation states, nor of Austria from the EU. Why? Because we have invested so much into the laws and institutions, and to throw this investment away has to be a well-considered step. No more, no less.

In my estimation, half of England’s lawyers will be busy for half a decadeat least before a new set of rules, for trade, for movement of people, capital, and for financial regulations are finalised in a non-EU UK. No significant progress in the UK’s legal system can be made while the legal resources are thus tied up. This is a real cost. The bar simply has to be higher, for we must be absolutely certain, when high institutional, legal, and governance switching costs are involved. We need to be 110% sure that this is what people really want. This is the lesson of last Thursday, and voting systems must be made as robust as parliamentary rules on constitutional changes already are. Otherwise Europe’s divisions may soon put the Balkan’s in the shade.