Instead of the Popular Initiative, let’s try the democracy of chance

François de Closets writes in l’Opinion.

The Gilets Jaunes, the coalition, the opposition, everybody seems to support the Popular Initiative (référendum d’initiative citoyenne, or RIC). This mechanism for popular participation would offer both a renovating reform of our republic and a way out of the crisis. Wouldn’t it, however, be a false solution? Wouldn’t it be embraced more because it is in the air rather than through thorough reflection? Wouldn’t it be masking a real solution? “Let the people speak”, who can object to that? No one, and it is for this reason that we must not give in to moral terrorism.

Popular sovereignty, the foundation of democracy, struggles with the question of the government. Beyond the scale of the city, even beyond that of a village, collective power is no longer operative. Representative democracy must be utilized. Every nation has arrived at this conclusion. That is, popular sovereignty does not mean governing but appointing and recognizing rulers. It also means that the citizens see themselves as being represented by those who speak and act in their name.

Despite this delegation, the people remain the ultimate source of truth, their word being superior to that of their representatives. In particular, their word must be imposed through referendum when it comes to the supreme law: the constitution. Representative democracy is therefore a compromise due to the impossibility of the ideal of direct democracy.

A crutch. In practice the system risks the formation of an enclosed political class which usurps power from the people. Conversely, direct democracy can be used by manipulators who under the pretext of “letting the people speak” impose their point of view on the majority. Real democracy, that of the general will, is therefore a historical construction which must fend off both of those perversions. The RIC should be examined from this perspective.

It is evident that the 5th Republic is experiencing a crisis of representation. Initial evidence was in the spring of 2017 with the routing of the established parties. The Gilet Jaunes are providing more evidence by occupying the roundabouts, ransacking the Champs-Elysées always providing the same explanation: “People like us, ordinary people, are no longer represented and have to wreak havoc in order to be heard”. Adding 100 euros to the minimum wage will no re-create the link of confidence between the people and the elites.

Two methods may be used to overcome this failure of the Republic: accepting the suspicion of the people toward the elites and remedying it using the crutch of direct democracy, or refashioning representation so that the Frenchpeople recognize themselves within it. The RIC, which has the appearance of a miracle solution, is in fact a crutch, giving a voice to a people who can no longer recognize itself in the elected representatives. This suspicion is therefore remedied by the right to intervene at any moment in any matter. Using the device of the referendum it will be possible to make laws, to annul existing laws, to block the politicians, even to depose rulers.

Recourse to direct democracy is always put forward by extremist parties which want to use it to impose their law onto the majority. A demonstration of that has been provided in 1793. Today they aim to put in question the results of the 2017 elections and to use the referendum in order to snare Emmanuel Macron as was done to General de Gaulle in 1969.

A representative democracy lacking legitimacy cannot regenerate itself using a crutch – which would only paralyze itself. Power, in constant fear of the people would avoid doing taking any unpopular action. It would be condemned for inaction, and would live in the shadow of demagoguery. Action is not judged by its short term effect but by its long term results. Let us therefore forget about this false solution, because there do exist real solutions which do not present these drawbacks.

“Ordinary” Frenchpeople are revolting because they feel that different forms of national representation have been taken over by a self-promoting elite and that people like them have no chance to participate in decision-making, to be heard by power. In order to fight against this distrust it would suffice to have an article in the constitution saying: “In each deliberative assembly, from the municipal council to the Parliament, one quarter of the members must be selected by lot among the voters who have taken part in the vote.

Athens. The principle is simple and all-inclusive. The application would be tricky, no doubt requiring a process of experimentation and evaluation. However, far from making a break with democracy, this goes toward rediscovering the it origin, since allotment was very widely used in the city-state of Athens. Let us not forget that chance is felt by the French as being fundamentally democratic. The proof is the fact that the lottery winner is only rich person toward whom there is sympathy.

But the system has its limit. We should not allot government officers and, furthermore, not all the members of an assembly. Chance can perfect elections by assuring that no category of citizens is excluded by the electoral choice. The rulers, to have their majority, would have to convince the simple citizens and not only create pacts among parties. That would help avoid errors. And wouldn’t it be better to have a dialog with the Gilets Jaunes before taking a decision rather than seeing them rise up against a policy which ignores them for the duration of an electoral term?

14 Responses

  1. Sad that he starts strong with a decent analysis and then concludes with a marginal and unworkable use of sortition (one quarter of an elected assembly selected by lot). The idea that letting a few ordinary people sit in the room with the elite power broker politicians will fix things (“Chance can perfect elections by assuring that no category of citizens is excluded by the electoral choice.”) is ridiculous. A mixed chamber is VERY unlikely to empower ordinary people (perhaps if it was 90% allotted and 10% elected… MAYBE).

    Obviously some sort of transition hybrid design is necessary because elections will not likely disappear instantly. Having a second chamber of some sort that is allotted is one possibility (though I have published a paper about why I think that also is a poor idea). A better strategy is to peel off one issue area at a time and completely deny elected politicians any more power on that topic, and vest that issue area into a sortition system. If it works well, next time there is a scandal, some more issue areas could be transferred from the elected chamber to the new sortition system until eventually the elected chamber is like the monarchies of Europe… still existing but toothless.

    Liked by 2 people

  2. Please do not continue to use STAN@COX.NET as it may not work in the future. Please change my email address to: STANLEYF@ALUMNI.USC.EDU Stan Follis 619.222.3447


  3. Totally agree with Terry Bouricius. In fact I’d go further. The analysis was really excellent. The solution is half baked and the explanation of the solution – that one throws one’s lot in with ‘chance’ – seems to me to put the proposal in its worst light. Chance is a factor in all things, and when it comes to allotment, it is mainly useful only where the chance element has been reduced to negligible levels either via the law of large numbers or rules about the preponderancy of opinion necessary to be determinative.


  4. Nick,

    Your comment shows the huge divide between Blind Breakers like Stone, Boyle and Dowlen (who argue that chance is the only justification for sortition) and the Invisible Hand camp who argue that statistical representation needs to be instituted in such a way that the sample and the target population are aligned by “stochastic determination”. The two views — one negative and one positive — are diametrically opposed. For analysis see:


  5. Chance can serve two purposes: (1) It can make citizen samples statistically representative (John Burnheim’s original proposal). This property is useful to determine correct subject matter preferences, i.e. policy decisions. (2) It can introduce an element of chance which works against the power of lobby groups to empower a specific candidate of their choice (Venetian elections). This property is useful for political appointments, i.e. personnel decisions.

    In both uses, we can have epistocratic filtering, in order to maximise decision making quality. Rousseau already argued that personal talent and integrity being perfectly acceptable differentiators. I have seen “naive” sortition at work, pure chance, and can say that it defeats the purpose of good collective decision making if we were to allow random ignorance and error to enter the process.


  6. Yes, preference and personnel selection are entirely different uses of sortition. Interesting that John Burnheim is no longer interested in statistical representation. As for the need for epistocratic filtering, I think this applies more to the isegoria function as the law of large numbers should minimise the effect of outliers on aggregate voting (isonomia). I agree that it’s better if people who open their mouth to speak know what they are talking about.


  7. Terry, Nicholas,

    I agree that the analysis is rather perceptive. It does, however, strike some false notes on important issues. de Closets’s explanations of the trouble with ‘direct democracy’ miss, or at least obfuscate, the main issue – that it suffers from essentially the same problems that elections do.

    There is, however, one very good point de Closets makes which may be getting lost: he is setting an objective, results-based criterion for what democracy is (although, admittedly he is being a bit ambiguous in his phrasing):

    [P]opular sovereignty […] means that the citizens see themselves as being represented by those who speak and act in their name.

    This presents a very useful and practical way to measure how democratic a system is. This rids us of the old convoluted, formalistic, unresolvable, fruitless discussions of whether a particular system is democratic or not. With this measure, any system can be judged as to its merits and reformed if it is found wanting.


  8. Nicholas,

    > it is mainly useful only where the chance element has been reduced to negligible levels either via the law of large numbers or rules about the preponderancy of opinion necessary to be determinative.

    If by “rules about preponderancy” you mean requiring supermajorities to pass new legislation or make some other decisions, then it is worth noting that this does not reduce “the chance element” – it simply moves the area of uncertainty from a 50-50 split to whatever split the supermajority rule demands.


  9. I liked these expressions “popular sovereignty does not mean governing but appointing and recognizing rulers. It also means that the citizens see themselves as being represented by those who speak and act in their name”. and particularly this “refashioning representation so that the Frenchpeople recognize themselves within it”


  10. I think you have to work fairly hard to misunderstand what I said about chance Yoram, but you may have done it.

    The higher a supermajority requirement is, the more confident you can be that a decision supported by it represents the true view of the community and is not a mere artefact of chance – of the randomness by which participants are selected in the representative body.

    You can be more confident that a jury voting 12-0 represents the community’s view of a case than if the vote were 7-5.


  11. [P]opular sovereignty […] means that the citizens see themselves as being represented by those who speak and act in their name.

    This might well have been abstracted from Hobbes’s Leviathan (or any other anti-democratic text), and it would reflect the perception of the Nazi party in the eyes of the majority German citizens during the late 1930s. As the definition is entirely subjective (how citizens see themselves), presumably it would be operationalised by a public opinion survey, and would be wide open to manipulation by government propaganda. The standard definition of democracy — government of, by and for the people — is objective, and would be measured by invariance of decision outcomes (at any given point of time) as it should make no difference which individuals speak and act in the name of the people.


  12. Nicholas,

    > You can be more confident that a jury voting 12-0 represents the community’s view of a case than if the vote were 7-5.

    That is true, of course. But it has nothing to do with the bar you set up for adopting a proposal. The ambiguity at 7-5 remains whether the adoption bar is set at 7-5 or, say, 10-2.

    The notion that in case of a close split the status quo should continue may or may not be defensible, but a preference toward the status quo does not reduce the uncertainty of statistical sampling.


  13. This proposal is very bad advice to the Gilet Jaunes, as I believe we all agree.

    The underlying idea of RIC is correct, given that that idea is that the people need the power to veto the laws of the politicians, and to put forward laws of their own which the people can decide.

    RIC falls far short of the mark because of the standard and well known problems with referenda. Unsuitability for making informed decisions, voter fatigue, rational ignorance, reliance on media and people’s media choices to get a fair hearing, and unsuitability for ensuring that proposals are considered on the basis of a fair hearing on a level playing field.

    Juries are the only way to solve these problems. They can put the deciding of laws on the basis of the informed consent of the highly representative portions of the public that juries are, with juries deciding after a fair hearing on a level playing field.

    Good advice to the Gilet Jaunes is that all government-passed laws should only go into effect if they obtain the informed consent of the people through a jury, after a fair hearing on a level playing field. More specifically, they can go first to a jury of say 100 randomly sampled and decently paid citizens. If a sufficiently large super-majority of them vote one way or the other on the law after a fair hearing, that settles the matter. If not, the law goes to a larger jury of perhaps 800 to 1,000 citizens that decides it by majority vote after a fair hearing.

    In addition, the politicians cannot be allowed to continue to have a monopoly on proposing laws, as the Gilet Jaunes well understand, as we can see from the RIC proposal. Instead, arrangements need to be in place to ensure that excellent legislative proposals are made, and are brought forward for juries to decide, regardless of whether politicians support those proposals or not.

    Only informed views are a good and reasonable basis for public policy. Democracy therefore needs be about informed rule by the people, or informed rule by highly representative portions of the people. Poorly informed rule by the people is not democracy, or if one wants to call it democracy, it is an inferior and undesirable version.

    Liked by 1 person

  14. The Athenians, unlike this proposal, had the good sense not to have decision-making bodies partly chosen by lot and partly by popular election. They also, unlike this proposal, had the good sense to exclude popularly elected politicians from the power to decide laws.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: