Clingendael Spectator 3 – 2017 (vol.71) – Item 4 of 7


Representative democracy and populism

Frank Ankersmit

The dangers of populism, though real enough, should not be exaggerated. Populists never demanded the abolition of representative democracy, a political system we know so well, but understand so poorly. The cause of populism is that after the death of ideology many people discovered the painful truth that representative democracy is not a democracy at all, but an elective aristocracy.

No political system we know so well, but understand so poorly as representative democracy. Misunderstandings already arise from its very name. Representative democracy is not a democracy at all, but an elective aristocracy. Rousseau (1712-1778) pointed out as much already. We have been taught to believe that in a representative democracy the people’s will is decisive. And that political representation has been invented to ensure that our rulers really do what the people wishes. This is how we understand the notion of popular sovereignty: the people itself is the sovereign legislator.

But Rousseau explained to us that, alas, things are not as easy as that. He confronted us with the plain and brutal truth that in a representative democracy it’s not the people but the people’s representatives who decide what will be the law. Citizens may be dissatisfied, or worse, by their representative’s decisions and vote for some other political party in the next elections. True, but that’s no occasion for great joy.

Who is the sovereign?

Nor is much solace to be expected from the indisputable fact that between elections policy issues will intensively be discussed in the newspapers, in the media and elsewhere, and that politicians wishing to avoid political suicide cannot afford to remain insensitive to public opinion – characterized as early as the 18th century as ‘la reine du monde’. For it would be wrong to see in this, with Urbinati,‍‍[1] a refutation of Rousseau’s argument. The politician has the freedom to follow public opinion, or to refrain from doing so. Hence, he is the person to decide whether public opinion will be enacted, or not. And this makes him into the sovereign.

In a representative democracy the people is the sovereign in the proper sense of the word on election-day only.

© James Boule / Flickr

In a representative democracy the people is the sovereign in the proper sense of the word on election-day only.

So much is clear already from whom will be regarded as responsible for the consequences of the enactment of what ought to be done according to public opinion. If following public opinion by the politician has disastrous results, it is not ‘the public’ but the politician who will be considered responsible. So there is no escape from the logic of Rousseau’s provocative claim that in a representative democracy the people is the sovereign in the proper sense of the word on election-day only.‍‍[2] On all other days the sovereign is the chamber of representatives that had been elected by the people. Thus Rousseau;‍‍[3] and the English Constitution (if – with Walter Bagehot – we may call it so), since it accepts the principle of parliamentary sovereignty.

Rousseau and the English – undoubtedly ‘bien étonnés de se trouver ensemble’ – fully agree with each other on where sovereignty must be located in our contemporary democracies. Whoever believes that the principles of representative democracy and popular sovereignty can be reconciled with each other is either naive or inconsistent. So constitutional precision requires us to exchange with Guizot and Mill (and Manin)‍‍[4] the term representative democracy for technically more correct alternatives such as ‘representative government’. Nothing is wrong with the latter term, but the former irrevocably is an oxymoron.

“In a representative democracy the people is the sovereign in the proper sense of the word on election-day only” (Rousseau)

The shortcomings of free elections

It is a consolation we no longer have the hereditary aristocracy of the Middle Ages and the Ancien Régime. We now have free elections enabling us to determine what this aristocracy elected by us will look like. This gives us, as citizens, at least some grip on the actions and decisions of our elected aristocrats. But even this has its limitations. For another painful truth is that free elections are neither a necessary nor a sufficient condition for adequate political representation.

Think of Napoleon between, say 1800 and 1804, when he crowned himself Emperor of the French. He adequately represented the French population during those four triumphant and happy years, there’s little doubt about that. But he had not been elected and came to power through the coup d’état of Brumaire 18th 1799. Hence, free elections are not a necessary condition for adequate representation. Nor are they a sufficient condition. We have free elections for the European Parliament, but the European authorities in their infinite wisdom arranged these free elections in such a way that one must be capable of a truly monumental naivety to believe that the European Parliament represents the European people.

These are sad truths about representative democracy. Many people are more aware of this than ever in the last decades. This may explain why some now longingly recall the direct democracy of the Greek polis. Neither the Greek polis nor the Rome of the early Republic had political representation; decisions were taken by the people itself. In theory at least, since actual practice, both in Greece and in Rome, often was a bit nastier. Since Antiquity direct democracy never played a role in Europe.‍‍[5] The main reason is that it cannot be put into practice in large states with often many millions of people.

A further problem is that the translation of political will into political practice may be difficult, if not impossible to realize, and that for two reasons. In the first, and more dramatic place, the electorate may unanimously want things that mutually exclude each other. For example, the people may wish military protection against foreign aggression, but, at the same time, refuse to pay for it.

Generally speaking, we are perfectly free to want things that logically exclude each other, and will even happily make use of that freedom in most of our lives. Daydreaming often comes down to just this. But when it comes to action and to the execution of our will, we shall have to make up our minds and then we must drop at least one of these mutually excluding things. Perspicacious as always, Rousseau was aware of this problem, too, and he saw in it the foundation for the distinction between the legislative (willing) and the executive power (doing).‍‍[6]

Free elections are neither a necessary nor a sufficient condition for adequate political representation

In the less dramatic case, a majority of one may be in favour of ‘a’ and a minority of one in favour of ‘non-a’. So this one person, making the difference between the majority and the minority, is then decisive for whether ‘a’ or ‘non-a’ will be adopted by all of the people. This person has – although undoubtedly unwillingly and unknowingly – become a kind of despot. And this seems to be glaringly at odds with the very idea of entrusting political decision to the people.

It looks like a return to absolute monarchy. Of course one may now object that this one person who is decisive is not someone in particular, but could be just anyone belonging to the majority. But whether we can identify that person or not is of little practical relevance. This is why we are intuitively reluctant to give the smallest possible majority carte blanche with regard to the largest possible minority. In both cases the best way to deal, or better, to avoid the problem is to rely on an assembly representing the people for deciding what had best be done under the circumstances mentioned here.

So that gets us back again to representative democracy when collective decision-making is really put to the test. And, needless to say, for a well-functioning body politic we need an organization of collective decision-making that can successfully stand all such tests. Hence, in the end representative democracy – or elective aristocracy – is our only reasonable and realistic option.

The clash between representative and direct democracy

But that does not exclude that the introduction of variants of direct democracy within the framework of representative democracy – such as the people’s initiative and the referendum –might fruitfully be considered. If only because representative democracy, like anything else in human society, is prone to decay. Representative democracies have an innate tendency to deteriorate into oligarchies, as Aristotle and Polybius already argued. Of course, this need not necessarily happen. But if it does not, this will often be because instruments have been developed to prevent such derailments of aristocratic regimes. And it may then be worthwhile to investigate whether variants of direct democracy might provide us with one or more of such instruments.

Why not? A minimum condition is, however, that representative democracy and these variants of direct democracy are carefully kept apart from each other. Each of them has its own political logic and if the two are thoughtlessly mixed up the inevitable result will be a constitutional mess. Such was the case with the Brexit when the principles of parliamentary sovereignty, as accepted in England, and popular sovereignty (the referendum on the EU) came into direct conflict with each other – thus requiring the High Court of Justice to restore the balance in favour of the former. From a constitutional point of view it must have been quite a relief that Parliament came, in the end, to the same conclusion as the British people. In this way a severe constitutional crisis has fortunately been averted. (Although anyone with an academic interest for constitutionalism and its problems would have been eager to see what would have happened if the majority of Parliament had been against the Brexit.)

What would have happened if the majority of Parliament had been against the Brexit?

© Flickr / Garon S

What would have happened if the majority of Parliament had been against the Brexit?

It will be clear from the foregoing what considerations must be taken into account in order to avoid such an outright clash between representative and direct democracy. In the first place, when opting for variants of direct democracy, one has to ascertain that no inconsistencies can arise when moving from the will of the people to its implementation. This is less easy to achieve than may seem at first sight. For often many such inconsistencies will become visible only if the adoption of a certain policy has been thought through till its very last consequences. If mistakes are made here one will be unpleasantly surprised by an asymmetry in the logic of representative and direct democracy.


Responsibility is the clue here. In case something goes awry in a representative democracy, not the people but the people’s representatives are responsible. In contrast, in a direct democracy the people itself is responsible for the consequences of its decisions. But where everyone is responsible, no one is. Hence, political responsibility will tend to evaporate in direct democracy. As long as the outcome of the decisions made in a direct democracy is satisfactory no problems are to be expected here. But when the outcome is disappointing, one cannot be sure that the people will blame itself for this – as it should.

Recall that we assumed that variants of direct democracy will always be introduced within the framework of representative democracy. And, next, that human beings have an innate inclination to blame others for their own mistakes. It then follows that the guilt for the failures of direct democracy will preferably be put on the doorstep of representative democracy. This is, then, why it is of the greatest importance that one delegates to direct democracy only those decisions whose implementation is not likely to occasion consequences that will be both acceptable and not acceptable to the people taking these decisions. For in that case a conflict between direct and representative democracy is to be expected.

Sovereignty understood as legislative power is a modern notion that came into being only in the 16th century

And in the second place, decisions taken with the instruments of direct democracy should preferably have the support of substantial majorities in order to avoid the despotism of those very few distinguishing the majority from the minority. Thirdly, if representative democracy should be protected against invasions by direct democracy, the reverse holds as well. That is, decisions made within the framework of direct democracy have to be respected by the institutions of representative democracy. Put differently, the outcomes of direct democracy are binding for all of the body politic.

But if these three conditions have been satisfied variants of direct democracy can undoubtedly be a welcome addition to the instruments of representative democracy.

Medieval political representation

Mobilizing direct and representative democracy against each other should all the more be avoided because the latter is itself an unstable mixture already. In order to see this one need not go all the way back to Antiquity, as with direct democracy. Nevertheless much further back than to the birth of representative democracy at the beginning of the 19th century. Namely to the late Middle Ages, some six to seven centuries ago. Political power then was much different from what it is now. To put it quite briefly, the princes and kings of that time did not possess sovereign power as would be the case with the monarchs of the 17th and the 18th centuries.

Sovereignty understood as legislative power is a modern notion that came into being only in the 16th century; Jean Bodin (1530-1596) and Thomas Hobbes (1588-1679) were the first to define it, and it could not have been defined before them for the simple reason that before then it had not yet come into existence. Medieval princes and kings were judges, no legislators. The Middle Ages had a spontaneously developing customary law and knew no legislation in the proper sense of that term.

The prince vs. the Third Estate

Nevertheless, the medieval prince often had to do business with the inhabitants of his territories. For example when he was in need of money; as was almost continuously the case. Because of wars, dynastic plans or domestic troubles the prince rarely managed ‘to live of his own’, as it was put in those days (and until the 17th century). The prince then convened the Estates-General, in which the nobility, the clergy and the Third Estate (cities and/or provinces) were represented.

Now, in practice the money the prince needed had to be obtained from the Third Estate, since this is where big money could be found. The result were negotiations between the prince and the representatives of the Third Estate in which the latter normally declared themselves to support the prince financially on condition that he granted them certain privileges such as the right to levy toll taxes, to mint coins or stack right. Note, that what the Third Estate paid the prince were no taxes in the real sense of the word, but a private law arrangement between the prince and the representatives’ cities or provinces.

In fact public law did not exist in the Middle Ages; private law was all one had. But, as such, law was all pervasive in the Middle Ages. Law determined the very texture of medieval society; hence the dictum: ‘quod non est in textu iuris, non est in mundo’. It is not hard to see who lost most from the transition from the Medieval to the modern way of collecting the financial means needed for public purposes.

Now, the crucial thing is this. The representatives of the Third Estate were always tied by a binding mandate to their grassroots. Hence, they could not follow their own compass when coming to an agreement with the prince. In that sense the grass roots were themselves present, in the persons of their representatives, in the latter’s negotiations with the prince. This is the essence of medieval political representation.

The transition from medieval to modern political representation

Modern political representation is basically different. Illustrative is article 67 subsection 3 of the Dutch constitution: “all members [of the Estates-General (F.A.)] vote without instructions” (originally the article stated, with a tacit reference to the situation under the Republic of before 1795: “without instruction and consultation”). That is, when casting their votes in the Estates-General the representative’s own conviction should be decisive and not that of those who had elected them. If those who elected him opt for ‘a’, but the representative is convinced that ‘not-a’ best serves the nation’s interests, he should vote ‘not-a’.

Representative democracy combined two entirely different things: medieval political representation and absolute monarchy

How has this transition from medieval to modern political representation been possible? Absolute monarchy is the explanation. The absolute monarch had sovereign power and could decide as he thought was best, regardless of the opinion of his subjects. Such was the theory, at least, for in practice the absolute monarch’s hands were tied by many restrictions on the monarch’s prerogatives established in the Middle Ages. Anyway, since the French Revolution the absolute monarch’s pretention to possess sovereign power was generally regarded sufficient argument for the abolition of absolute monarchy. Nevertheless, sovereign power remained indispensable for the modern state. It could not be abolished together with the absolute monarch; after 1800 a return to the Middle Ages was no option. But what to do with sovereign power, now that it had been taken away from the absolute monarch?

The French Revolution

The funny thing is that the question was never really asked in such a straightforward way. Instead, sovereign power was now handed over more or less as a matter of course to the representative institutions inherited from the Middle Ages. The French Revolution is the best example. Louis XVI had convened in 1789, for the first time since 1614, the Estates-General. When by the government’s singular ineptitude a meaningful discussion with the Estates-General failed to emerge in May and the first weeks of June, Emmanuel Sieyès (1748-1836) boldly declared on June 17th the Third Estate to represent all of the nation. And three days later, on June 20th, in the Oath of the Tennis Court, the Third Estate (together with some members of the nobility and the clergy) transformed itself into a legislative body, the Assemblée Nationale, setting itself the task to give the nation a new constitution.

Mirabeau's defiance in front of the marquis de Dreux-Brézé on 23 June 1789.

© Wikimedia

Mirabeau's defiance in front of the marquis de Dreux-Brézé on 23 June 1789.

So in a mere three days France made the tremendous leap from the Middle Ages to a modern representative democracy. However, without anyone carefully thinking through all aspects of this unheard constitutional revolution. Admittedly, three days is a bit short for doing so.

Anyway, the result was that representative democracy combined two entirely different things: medieval political representation and absolute monarchy. Our people’s representatives therefore have two tasks: (1) to represent the people (the medieval heritage); and (2) legislation (that of absolute monarchy). And the French revolutionaries soon added a third to these two: to control how the executive performs its tasks.‍‍[7]

Now, this is, in fact, a most bizarre mixture of tasks – and we fail to recognize this since we have become so much accustomed to it that we can hardly conceive that government could (and perhaps even should) be organized differently. The first task demands of the people’s representative to do all he can to realize the desires of those who had voted for him and to go on doing so until his forces fail him. But the second task requires him to leave his voters in the cold if he himself considers their desires unwise, impracticable or imprudent. From the voter’s perspective this must be as if in a law-suit his attorney is at the same time his judge. Who would not have the impression to have entered the world of Kafka under such circumstances? Sieyès, the constitutional Einstein of the French Revolution, already realized himself that something must have gone wrong here and found a more or less acceptable solution for it, the main idea being that representation and legislation should carefully be kept apart.‍‍[8]

Sieyès’s proposal would be the point of departure of the French constitution of 1795. That constitution was no success and ended with Napoleon’s coup d’état of 1799. But it is likely that the constitution’s failure was caused by the political instabilities of the country in those years rather than by the idea that had inspired it. Since then Sieyès’s ingenious solution was wholly forgotten. Above all since nobody really worried after him about the wisdom of assigning the task of representation and legislation to one and the same body. Politicians may, at times, have felt to be torn between these two tasks when asking themselves how to realize when in power what they had promised to do to their voters. But this must have seemed to them one of the inevitable complications of their job they had to live with – just as a GP should both cure his patients and to write them bills for his consults – rather than a matter of principle.

The emergence of political ideologies

The problem was wholly relegated to the background with the advent of political ideologies, such as liberalism, Christian democracy or social democracy. These ideologies created an indissoluble tie between the voter and his representative inviting them to remain blind to the problem. Thanks to political ideology the representative could be sure of having the reliable and faithful support of part of the electorate and the voter was convinced of having in his representative a people’s tribune doing for him in Parliament all he could.

The fact that his representative was also a member of the legislative was not in the least a problem for the voter in the age of political ideology. On the contrary: his legislative powers were precisely their strongest, and most needed weapon in their struggle with their political adversaries! To put it with a dash of paradox: thanks to political ideology modern political representation did not differ in the political experience of both voter and representative from what medieval political representation had been like. The fundamental difference between representation and legislation had now wholly disappeared behind the horizon.

‘We have free elections for the European Parliament, but (…) one must be capable of a truly monumental naivety to believe that the European Parliament represents the European people.’

© European People's Party / Flickr

‘We have free elections for the European Parliament, but (…) one must be capable of a truly monumental naivety to believe that the European Parliament represents the European people.’

But in our own time political ideologies are as good as dead; and baroness Thatcher’s TINA-principle firmly stands in the way of any attempt to resuscitate them again. So now we can no longer ignore the painful fact that those people’s representatives always pretending to listen so very well to us are at the same time the heirs of the absolute monarch who is free to decide about our fortunes, who may tax us as he pleases (what a difference with the medieval prince humbly pleading with his subjects for financial assistance!) or send us to a war we oppose ourselves.

Nice, good old representative democracy now turns out to be a far less innocuous elective aristocracy – which, in fact, it had always been, but without our being aware of it. The people’s representative is now unmasked as someone with two faces. One of them is directed to the voter. But the other to all kinds of people for whom the voter will have little sympathy and whom he may even suspect to be a threat to his interests and to all that is known, familiar and dear to him. Such as the ladies and gentlemen of the EU and of the financial sector.

The political representative: attorney or judge?

For the representative something similar is true. He will now have to choose whether he will be the ‘attorney’ of his people standing up for the realization of their desires whatever these may be, or a severe ‘judge’, who with the book of Political Reason in his hands, will often pronounce a harsh judgment on the people’s often so irresponsible and impracticable wishes. Wilders in the Netherlands and Trump in the USA choose for the first option – with the result that they either shun like the pest taking political responsibility (Wilders), or behave irresponsibly after having taken it (Trump). Traditional politicians and political parties choose for the second option – which is why they were often wrongly, but understandably, accused of representing not the electorate but the ‘elite’.‍‍[9]

Speaking generally, it is still insufficiently recognized what the death of ideology must mean for our kind of democracies. Ideology is for representative democracy what gasoline is for our cars: representative democracy comes to a grinding halt without political ideology. Self-evidently, this is not meant to be a recommendation for reinventing political ideology, but it is merely an observation on the nature of representative democracy.

Should we do something against populism? The answer is a simple ‘no’

The success of populism

This explains the success of populism on both sides of the Atlantic Ocean. After the death of ideology the ideologically unbound voter tended towards a radical simplification of the political spectrum and to divide politics and society into, on the one hand, ‘the people’ indignantly demanding the undiluted representation (and realization) of its will and, on the other, ‘a corrupt elite’ distorting ‘the will of the people’ by transforming it into an ugly caricature in which ‘the people’ can no longer recognize itself.‍‍[10]

That may look like political paranoia – even more so, that’s what it is. But we should not remain blind to the undeniable fact that populism sometimes puts issues on the political agenda that are of great importance. Such as: what is national identity, is there still room for a ‘national home’ in an economically and culturally globalizing world? What protection may we expect from the state against unwanted social, cultural, economic and religious foreign influences? Are there limits to the demand to treat people equally or to that of the freedom of expression? If so, where? And what should we do if such demands are in conflict with each other? These are good and sensible questions and anyone who fails to acknowledge their urgency is living on another planet than ours.‍‍[11]

Ideology is for representative democracy what gasoline is for our cars: representative democracy comes to a grinding halt without political ideology

In conclusion: do not worry about populism

What conclusions follow from all this? In the first place, the dangers of populism, though real enough, should not be exaggerated. Notice, above all, that populists never demanded the abolition of representative democracy. With populism, as it presently is in the West, we’re not on the threshold of a new 1789, 1917/1918 or 1933. Although, what is not yet, may come. Anyway, for the time being there are many excellent reasons for being deeply worried about the future, alas, but populism is not one of them.

Next, the cause of populism is that after the death of ideology many people discovered the painful truth that representative democracy is not a democracy at all, but an elective aristocracy. Hence all this raging against the ‘elite’; which is, by the way, an exact copy of the abuse of the ‘establishment’ by the generation of 1968 when it came to a similar discovery about our political systems.‍‍[12]

In the third place, should we do something against populism? The answer is a simple ‘no’. Populism was born from the discovery that representative democracy is, in fact, an elective aristocracy. It is understandable that this was the cause of disappointment, frustration and anger. But the insight is perfectly correct. And it is better that people face reality than that they continue to believe in an illusion.


N. Urbinati, Representative democracy. Principles and genealogy, Chicago; Chicago University Press, 2008.
Rousseau, Du Contrat Social, Paris, 1962, pp. 301, 302. For the sometimes amazing vagaries of Rousseau’s notion of political representation, see R. Fralin, Rousseau and representation. A study on the development of his concept of political institutions, New York, 1978.
B. Manin, Principles of representative government, Cambridge, 1997. See also H. Hoffmann, Repräsentation Studien zur Wort- und Begriffsgeschichte von den Antiken bis zum 19. Jahrhundert, Berlin, 1974.
Apart from in some medieval city-states in Italy and the Netherlands, and down to the present day in the Swiss cantons Glarus and Appenzell-Innerrhoden.
“[Q]uand je marche vers un object, il faut premièrement que j’y veuille aller: en second lieu, que mes pieds m’y portent. Qu’un paralytique veuille courir, qu’un homme agile ne veuille pas, tous deus resteront en place. Le corps politique a les mêmes mobiles: on y distingue de même la force et la volonté: celle-ci sous le nom de puissance législative, l’autre sous le nom de puissance exécutive. Rien ne se fait ou ne doit s’y faire sans leur concours.” (Rousseau, Du Contrat Social, Paris, 1962, p. 272).
In 1793 Condorcet and Hérault de Séchelles observed already that the members of the legislative body (at that time the Convention) found legislation far more sexy than controlling the government and therefore suggested proposals for how to compel the Convention to perform its task to control the government. Since then, what Condorcet and Hérault put on the agenda has lost nothing of its urgency, because in most Western democracies control of the government still is an issue deserving far more attention than is given to it.
The construction was, basically, that the people should elect a Tribunat, having no other function than to represent the people, and, next, the Executive. These two institutions should debate what best serves the nation’s interests. Lastly, the people elects a Legislative Body deciding like a kind of neutral judge about the arguments adduced in the discussion between the Tribunat and the Executive. See E. Sieyès, ‘Opinion de Sieyès, sur plusieurs articles des titres IV et V du projet de constitution, prononcé à la Convention le 9 thermidor de l’an troisième de la République’, in ibid., Oeuvres de Sieyès, Vol. 3, nr 40 (23 pp.); s.l., s.a.
Hence the perfectly justified sneer by Mr Rutte, the Dutch Prime Minister, at the populist leader Mr Wilders, in a debate between the two gentlemen on March 13th 2017, that Mr Wilders “was content with twittering from his couch”, whereas he – Rutte – takes seriously the politician’s real responsibilities.
‘Populism is defined here as a thin-centred ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ and ‘the corrupt elite’, and which argues that politics should be an expression of the volonté générale of the people. C. Mudde & C. Rovira Kaltwasser, ‘Populism and (liberal) democracy’, in: ibid. (eds.), Populism in Europe and the Americas. Threat or Corrective for Democracy?, Cambridge, 2012, p. 20.
“What is the meaning of national identity, patriotism of a feeling at home in an individualist culture open to all of the world? What protection can the nation-state offer against unwanted social, cultural and economic influences from abroad? What kind of community are we; who do and do not belong to it? Which cultural differences are valuable and which are not? What are the limits of toleration, of freedom of speech and freedom of religion?” (D. Pels, Het volk bestaat niet. Leiderschap en populisme in de mediademocratie, Amsterdam, 2011, p. 31.)
It is an intriguing question why one now prefers the term ‘elite’ to that of ‘the establishment’. Whereas the latter is suggestive of the possession of political power the connotations of ‘elite’ are social and cultural rather than political. This would be in agreement with the fact that the generation of 1968 glorified Marx and welcomed revolutionary change, whereas contemporary populism rarely goes further than to reclaim what was lost and gambled away by an allegedly narcissist ’elite’.