Decisions made by a democratic National Assembly (or Parliament or Congress) have democratic legitimacy. That’s what we want in a democracy, decisions that are valid because they are made democratically.

However, strangely enough, the democratic legitimacy of decisions correctly made can sometimes be a problem. Whatever the National Assembly decides, must be correct because it is democratic. If someone is able to get the National Assembly to make a decision in their favour, say in a matter of taxation, they have won, because the National Assembly has put the stamp of “democratic” on that decision.

One agent who has an interest in getting the National Assembly to make certain decisions is the government. Governments have agendas they want pushed through, and they want to do that with as little trouble as possible from the lawmakers. National Assemblies are therefore under pressure to produce certain decisions and to do so without resistance.

The potential problem here is that this may push the National Assembly into making badly planned decisions because the government is desperate to get those decisions made that it has promised the electorate and to get the Assembly’s stamp of “democratic” on them. Such bad decisions are a big problem: since they are democratic, it is very difficult to overturn them and the country is stuck with potentially serious consequences of mistaken decisions.

Such mistakes happen. A case in point is Brexit. Then Prime Minister Cameron was able to get Parliament to sanction a referendum in a quick and easy decision without giving him any trouble or resistance in the matter. That was clearly a mistake. The country is now tearing itself apart and is unable to extricate itself from the mistake that has fallen down upon it. (Although it is my belief that Parliament will eventually find a way of correcting this mistake, it is, as we are seeing, very difficult to overturn a decision that has the legitimacy of a referendum behind it.)

The lesson is that National Assemblies should be able to make good decision and protected from making bad ones. Their decisions should be democratic but they should also be good, productive and workable, and certainly not counterproductive.

National Assemblies need assistance to manage the difficult combination of democratic and productive decisions.

They need the assistance, first of all, of protective procedures. Procedure is a boring matter for those of us interested in politics, but terribly important. National Assemblies need to impose rules upon themselves whereby they force themselves to not making decisions without careful scrutiny of consequences. They must avoid knee-jerk decisions because such decisions are in high risk of being bad. They must give themselves time and they must take themselves through routines of scrutiny. The Brexit decision, for example, was taken by Parliament without any preliminary work on what the consequences might be, and we are now paying the price.

They need assistance, secondly, in knowing what is in the interest of the people. One might think a good way of doing that is to ask the people, for example in a referendum. But we now know, from modern psychological research (such as by Daniel Kahneman who won the Nobel Prize in Economics for his research), that this is too simple. You and I and all of us are prone to making mistakes about our own good because there are mechanisms of bias at work in our minds. Our instinctive preferences are not necessarily what we really want. It turns out that people often change their minds and correct their preferences if they are given the opportunity to reflect and work on them with some care.

From this, political scientists are concluding that the popular will is not something that just exists in the population but what emerges from what they call “deliberation.” The German theorist Claus Offe has suggested that National Assemblies should have the support of more sophisticated information about “the will of the people” than raw expressions of preferences. He suggests “deliberative panels” consisting of a random sample of the population – he calls them “citizen deliberators” – charged with working their way through appropriate procedures from raw preferences to “reflective preferences” in important political matters. He thinks the panels should be constituted by some kind of lot among all citizens, that their task would be to help both citizens and lawmakers to form considered judgements, and that their authority would be exclusively advisory for National Assemblies.

As often, original ideas on first encounter seem odd, but this suggestion is really quite common sense. The reason we have National Assemblies, is that in a big population we for practical reasons must appoint a sample of the population to make decisions for us. If preferences depend on deliberation, the same logic would apply to the pre-decision process of forming preference. We will not get proper deliberation unless we design proper procedures to do it.

As for additional panels to advise the National Assembly, that’s the rationale of for example the House of Lords in the British Parliament and of various other “upper houses” in other national legislatures. Offe’s idea is slightly different, in how the panels are made up and precisely what they would do, but the idea is much the same.