In 2018, after heated debates in both chambers of Parliament, the Dutch government abolished the advisory citizen-initiated referendum in the Netherlands.¹ This was the provisional conclusion of a decades-long discussion of this tool of direct, participatory democracy in Dutch politics on the national level, which had led to its introduction in 2015. Although only two non-binding citizen-initiated referendums have actually been held in the Netherlands, namely those on the eu-Ukraine Association Treaty (2016) and the enhanced Intelligence and Security Services Act (2018), the history of this political tool is protracted and intrinsically linked to political and societal debates on the ideal functioning of Dutch democracy since the late 1960s. On the one hand, these debates included an increasing call against the indirect and distanced institutional character of the Dutch democratic model and, on the other hand, the continuing refusal of most established political parties to integrate the citizen-initiated referendum into the Dutch democratic system.

This history of contention over the institutional characteristics of a democratic system fits well with the historiographical approach of investigating democracy as an ‘essentially contested concept’.² This notion, originally introduced by the British philosopher Walter Brice Gallie, refers to a concept of which it is impossible to reach agreement in terms of its interpretation or correct usage.³ In other words, democracy, or being democratic, has meant fundamentally different things to different actors over time and there has and always will be contention over its definition and channelling via institutions. Consequently, when investigating democracy, historians have spoken of competing ‘democratic conceptions’ or ‘democratic repertoires,’ which clashed not merely in the Netherlands, but also in a wider European context.⁴

In the first postwar decades, Dutch democracy had been marked by a representative government with ‘an emphasis on freedom, self-government by a broadly defined elite, and a strong civil society, rather than by participation of the population at large’.⁵ However, as several political historians, most notably Martin Conway, have indicated, several old pillars that supported the Western European democratic order of distanced democracies began to crumble from the late 1960s onward.⁶

First of all, substantial economic growth, which had until then characterised postwar Europe, was no longer a given. Cultural norms were also changing, with an increasing focus on emancipation and individuality. And finally, the traditional political institutions of the direct postwar days, which were the top-down people’s parties, began losing support.⁷ In the Netherlands, sharp distinctions between progressivism and conservativism began to appear, class and religion became less important in voting behaviour, and calls for democratic renovation arose.⁸ Illustrative of these developments was the decreasing predictability of elections, as well as the increasing use of opinion polls in Dutch media.⁹

Sentiments of democracy dying or losing its postwar foundations were common in the 1970s and, according to some, Europe entered an era of ‘post-democracy’, a concept originally coined by Colin Crouch, meaning that countries remained democratic in name, while their representative institutions increasingly lost power.¹⁰ Historically, however, it would be more accurate to speak of this new era as yet another period of democratic contention, with both new and incumbent political actors proposing institutional renovations to amend the supposed shortcomings of the postwar conceptualisation of distanced democracy.

As a response to the crisis of the distanced democracies of the direct postwar era, Dutch political elites put forward competing proposals to restock the institutional, democratic toolbox. By doing so, they aimed at closing the ‘growing gap between political (party) elites and the “people”’.¹¹ Their ideas included a potential reintroduction of district voting, more attention to the interests of peripheral regions, and the organisation of election campaign tours that covered the entire country.¹² Besides, parliament increasingly valued openness and transparency, considered new parliamentary procedures such as an extra-long ‘question hour’, whilst opposition parties gained more room to criticise the government.¹³

Additionally, new political actors like the student movement and the new political party Democraten 66 (D’66) proposed and introduced more uprooting forms of democratic contention, in line with what Wim de Jong has aptly called the ‘libertarian’ and ‘politicising’ democratic repertoires. These new repertoires, which focused on individuality and radical citizen-participation respectively, competed with the existing frames of ‘diversity’, ‘unity’, and ‘discipline’, which had characterised the postwar years of distanced democracy.¹⁴ As Ank Michels states, the so-called ‘politics of accommodation’ of the direct postwar years came to an end and extra-parliamentary means of voicing concerns became increasingly popular, for example in the form of demonstrations or one-issue action groups.¹⁵

As this article aims to show, the citizen-initiated referendum was another heavily contested, yet so far largely ignored part of the democratic renovation debate in the Netherlands that arose from the late 1960s and the early 1970s. The debate on a potential introduction of the referendum is a central part of the aforementioned battle between various democratic conceptualisations. Similar developments took place in other European states as well, most notably in Italy and the United Kingdom. However, in these countries the referendum was eventually institutionalised, while in the Netherlands the binding referendum remained a theoretical proposal.¹⁶

The absence of actual citizen-initiated referendums held in the Netherlands explains the scarce historiographical attention that has been paid so far to the debates on the referendum since the late 1960s. The Dutch referendum experience is merely discussed, either as an instrument that does not fit any of the Dutch ‘democratic repertoires’, or as a small part of broader studies that investigate specific referendum results after 2005, the year in which the first and only Dutch state-initiated referendum was held on the European Constitution (Table 1).¹⁷ As a result, remarkably little is known about how and why the debate on the referendum took place in the later postwar period, but also why this tool of citizen participation was never introduced. This is unfortunate, as historical debates about direct democracy can very well serve as a lens to investigate competing ideas about the ideal functioning of democracy. The present article aims to do so and is a follow-up study of my earlier publication in Tijdschrift voor Geschiedenis, which analysed the discussions on democratic reform in the direct postwar period in France and the Netherlands.¹⁸

The main goal of this article is to find out what the increased call for and the incumbent resistance against citizen-initiated referendums can tell us about prevailing and arising democratic conceptualisations in the Netherlands from the late 1960s until 1999. The article consists of four parts and largely follows a chronological order. The first part provides a short overview of the Dutch referendum debate before 1966. After all, the referendum gained momentum in the 1960s, yet, referendum debates and practices had existed before and these experiences also informed the later discussions.¹⁹

Secondly, the article examines the increased criticism of the distanced conceptualisation of postwar Dutch democracy that arose in the late 1960s. It answers the question of what different referendum proposals were on the table as a solution for the perceived crisis of postwar democracy. This section begins in 1966, as the main movement for democratic renovation, D’66, was founded in that year and because the referendum was one of the constitutional renewals investigated by the Staatscommissie (state committee) headed by former Prime Minister Jozef Cals and judicial expert Andreas Donner.²⁰ It analyses the various referendum solutions that were proposed to fix the apparent lacks of distanced democracy. This includes both those proposed by reformist movements as those by various other state committees in the 1970s and 1980s. Sources used include the reports of these state committees, as well as a limited number of political manifestos, to analyse how political actors and political experts discussed the referendum.²¹

The third part of the article subsequently uses articles that appeared in national newspapers, as well as party programmes, to investigate the political and societal debates on referendums between 1966 and 1989. It discusses which political actors proposed and who opposed the citizen-initiated referendum and what ideas about the ideal functioning of democracy were behind their positions. The newspapers were accessed through the online media archive Delpher. As this research engages with a national debate, it uses all newspapers that had a national audience at the time and considers all references to referendums in the Netherlands. This means that articles from the following newspapers are investigated: Algemeen Dagblad, Algemeen Handelsblad, De Telegraaf, De Volkskrant, De Tijd, De Waarheid, Het Parool, Het Vrije Volk, nrc Handelsblad, Nederlands Dagblad, Trouw, and Tubantia. For these newspapers, the search key ‘referendum; Nederland’ was used, to include articles that discussed the referendum as a political tool, but exclusively in the Netherlands. This paragraph examines the arguments for and against referendums advanced by a broad scala of political actors, moving beyond the views of political elites to also include individual citizens and civil society.

The final section analyses why the citizen-initiated referendum eventually did not see the light of day in the 1990s, while other proposals for democratic renovation did make it. Here, 1999 has been selected as the end of the research period, because in that year, the Nacht van Wiegel (Wiegel’s Night) on 18 to 19 May put an end to the constitutional procedure to institutionalise the binding citizen-initiated referendum.²²

All in all, this contribution firstly argues that in the post-1970s atmosphere of competing ideas on democratic renovation, the debate on the citizen-initiated referendum became one of the central battles over the correct practices of democracy, partly following, but also partly transcending traditional party lines. Secondly, it claims that the theoretical notion of parliamentary primacy in the agenda-setting process was never questioned and was deemed inviolable in this debate. Therefore, the referendum remained one of many proposals for democratic renovation to close the supposed gap between representatives and those they represent. This dominant interpretation allowed its opponents to not integrate the referendum contrary to other, more parliamentary reforms that better fit Dutch parliamentary democracy.

The history of the Dutch referendum experiment goes back as far as the late eighteenth century. The Batavian Republic held a number of direct votes to approve constitutional changes in revolutionary times.²³ The Batavian connection to the French occupation nonetheless discredited the legitimacy of these referendums and after the Restoration, the referendum as a means of representing the unmediated popular will did not return. Instead, when a limited mode of democratic government was introduced following the famous constitutional revision of 1848, the bicameral Dutch parliament was given a so-called vrij mandaat (free mandate) indicating that it could govern the country as it deemed fit, once it was installed.²⁴

It would take until 1887, almost a century after the Batavian experiments, before parliament discussed the referendum again, albeit as a marginal part of a broader constitutional debate. When a constitutional revision was implemented in that year, one of the subjects discussed was the potential replacement of the Senate by holding referendums. However, the idea was dismissed rapidly, and so was a very similar proposal that Pieter Jelles Troelstra made in the House of Representatives in 1903, due to a complete lack of parliamentary support.²⁵

After the First World War and the introduction of universal male suffrage in 1917, the referendum became the principal subject of a governmental investigation. A state committee that dealt with the form of Dutch democracy was set up in December 1918 under the leadership of former Prime Minister Charles Ruys-De Beerenbrouck. It discussed elements such as universal female suffrage, succession to the throne, and the referendum.²⁶ However, with the exception of one dissident member who pointed to the Swiss success with referendums as a positive example, it advised against direct votes.²⁷ The committee regarded the referendum as a ‘foreign plant’, that was not fit for the Dutch tradition of representative government.²⁸ It believed that there was no room for citizen involvement after Parliament had made a decision and that the people already had the opportunity to voice their concerns every four years. According to the committee, referendums would only obstruct the dominant system of party politics and slow down an already sluggish system of government. All in all, the committee thus considered the referendum at odds with the main theoretical notion of Dutch democracy, namely that members of Parliament had a free mandate for four years to represent the interests of the people as they saw fit.²⁹

Although a few experiments with referendums were actually carried out on the local level, the postwar conceptualisation of Dutch democracy was largely dominated by the belief that rule by the people required a certain level of discipline. Therefore, it was generally deemed beneficial that an educated elite would take the major political decisions and that voting once every few years established enough of a popular mandate for parliamentarians to be legitimate representatives.³⁰ From the late 1960s onwards, however, multiple Western European states experienced severe challenges to this postwar democratic order and the Netherlands was no exception.

At the time, several signs indicated that the direct postwar conceptualisation of a managed, distanced democracy was no longer untouchable. First of all, new political movements aiming for democratic renovation entered the political arena. Some of these were very small and fairly insignificant, such as the Lobby-groep Referendum en Initiatief (Pressure group Referendum and Initiative) and the minor Partij voor Volkswil en Referendum (Party for Popular Will and the Referendum).³¹ The larger party D’66, however, was much more influential and believed the Dutch system of distanced democracy had resulted in a democratic deficit that needed to be amended (Figure 1).³²

Secondly, within the incumbent parties, a changing attitude towards the postwar conceptualisation of democracy could be felt as well. For example, new political currents questioned the status quo, most prominently within the social democrat PvdA, where a group of left-wing reformers, who called themselves Nieuw Links (New Left), aimed to push the party towards a more progressive course (Figure 2).³³ Additionally, the 1967 parliamentary elections showed a clear change in political alignment. As Michels states, the Catholic kvp lost substantially and the pillars that had so long influenced election results appeared to be disintegrating. In these elections, D’66 performed spectacularly well and obtained seven seats out of nowhere.³⁴ Four years later, D’66 obtained another four seats while the kvp lost seven. Characteristic of the moment was also that progressive members of the kvp split from the party and formed the libertarian ppr.³⁵

Finally, a wider European movement for systemic democratic change arose across Western Europe, mostly in the youth cultures of Western European cities.³⁶ Emblematic of this movement were the 1968 protests, especially in Paris, headed by students who belonged to the first true postwar generation. These protesters sought more political participation, independent from the cues of political parties or the virtue of social class or religion. They strongly criticised the indirect nature of the postwar democratic model and, according to both Matt Qvortrup and Joost van den Akker, the referendum theoretically functioned as an alternative participatory mechanism for citizens that had become dissatisfied about the perceived malfunctioning of the political system.³⁷ The call for more participatory modes of democracy, thus, was not unique to the Netherlands.

Solutions to amend the postwar interpretation of democracy that were presented by D’66 included the direct election of certain (local and national) governmental positions, such as mayors and the prime minister. Nieuw Links focused strongly on holding parliament accountable for its actions.³⁸ However, for the smaller parties mentioned above, only the citizen-initiated referendum would amend the shortcomings of the disciplined democracies. In its founding manifesto, D’66 also expressed an interest for the potential introduction of referendums on particular political issues.³⁹

The debate on the referendum did not only take place within new political organisations, but also led to a response from the political elites. Most importantly, in 1967, the Cals/Donner state committee was instated. Its goal was to investigate institutionally innovative proposals for an upcoming overall constitutional revision.⁴⁰ In its 1971 report, the committee discouraged the government from holding citizen-initiated referendums, as it considered them too expensive and argued that there was too little experience with and knowledge of referendums. Besides, the committee did not observe a ‘great desire’ for the citizen initiative in Dutch society.⁴¹ Without ‘these conditions being fulfilled’, the committee believed it would be unwise to alter the system of purely representative government, which had such deep historical roots in the Netherlands.⁴² Besides, according to the committee, by introducing the referendum, there would be a chance that politicians would no longer be willing to take unpopular long-term decisions.⁴³

Still, a significant minority of the committee was in favour of introducing a form of the abrogative referendum, which would later be introduced in Italy, as it stated that this would increase the accountability of parliamentarians.⁴⁴ The focus thus was on improving the quality of parliamentary rule and not on enlarging the opportunities for citizens to advance their grievances. To this end, the committee drew up an example of what a referendum bill could look like and what extra conditions should be implemented, such as the binding nature of a direct vote, a number of topics on which no referendums should be held, such as popular tax reductions, and the minimum thirty per cent turnout number that would be required for a binding decision.⁴⁵

In 1982, a year before the major constitutional renewal of 1983, a second committee was formed to advise on the relation between voters and the policy-making process. Former Prime Minister Barend Biesheuvel, who headed the committee, stated in his 1985 report that citizen participation had proven to be important for a strong democracy, but that participation in political decision-making was rather limited in the Netherlands. He expressed the concern that this lack of actual influence would lead to frustrations among citizens and a declining interest in further participation in the democratic process.⁴⁶ The committee therefore strongly advised to integrate referendum practices into the existing Dutch representative system of democracy, to enhance participation.⁴⁷

The committee proposed three different types of referendums and was positive on all of them, albeit to varying degrees. Firstly, the corrective referendum would be a means for citizens to retract a bill that parliament had already accepted. To prevent an explosion of referendums, it did advise the introduction of certain speed bumps, such as a relatively elevated minimum number of petition signatures compared to Italy and Switzerland, to avert an upset in the parliamentary primacy of Dutch democracy, as well as a minimum turnout of thirty per cent to validate the vote. Support for this recommendation was unanimous.⁴⁸ Secondly, a majority of the committee members supported the law-initiative referendum, which meant that citizens could write a law proposal themselves and, if all conditions were met, would be allowed to defend it in parliament. If the bill would be voted down by less than eighty per cent of parliament members or if parliament decided not to discuss the proposal at all, a referendum would be held according to the same conditions as for the binding corrective referendum.⁴⁹ Finally, half the committee endorsed a state-initiated advisory referendum, allowing the government (or at a later stage parliamentarians) to consult the public on a particular law.⁵⁰

All referendum types were to follow certain rules, for example that a range of topics could not be the subject of any direct vote. These included the royal family, as a referendum on this topic would hurt the position of the monarchy no matter the outcome; international treaties, as referendums on those could sow confusion about the Dutch commitment to international cooperation; and matters that influenced the salaries of public servants, as citizens would not be able to take into consideration all competing interests. Remarkably, the committee did override objections to referendums on other aspects of the finances of state, making these potential referendums a fairly powerful tool.⁵¹

While the two state committees investigated the citizen-initiated referendum, a vibrant debate among diverse societal groups and politicians about the desirability of the referendum equally emerged from the late 1960s onwards. The proponents of such a change originated from, but certainly were not limited to, the progressive reform movements of the late 1960s that had popularised the criticism of the distanced democracies. Here, the Dutch case shows a clear difference with Italy, the United Kingdom, and France, where the referendum was largely a proposal from traditional political actors.⁵² In the Netherlands, the supporters of the referendum proved to be fragmented.

First of all, there existed a wide array of controversial topics on which citizens believed it would be useful to hold a referendum. For example, the right-wing Dutch parliamentarian Honig van den Bossche (Boerenpartij, Agrarians) argued in 1973 that a potential parole of the infamous ‘Drie van Breda’ – a group of three German war criminals – should be judged in a referendum.⁵³ Another example is that of a female author, C. Cassa, who in 1973 suggested holding a referendum on the question of abortion: ‘the fairest way to reach a just system of [abortion] laws is to hold a referendum for all women. Its outcome should be respected’.⁵⁴ Many authors also argued in favour of holding municipal referendums as a way of democratising local politics.⁵⁵

Most importantly, from 1982 to 1984, the matter of the nuclear question, according to some, could only be solved by organising a referendum. This concerned the use of nuclear energy, but mostly the placement of American nuclear missiles on Dutch soil, which created high tensions in Dutch society. Parliamentarians, committees, and individual citizens called for a referendum (Figure 3) and there even was a television series in which fictional referendums on nuclear energy were held.⁵⁶ One opinion poll from 1984 reported that 75 per cent of the Dutch population wanted a referendum on this topic, but hostility to this proposal in The Hague prevented such a direct vote.⁵⁷ In 1987, the pacifist movement finally retracted its support for the referendum.⁵⁸

Secondly, divergent political movements supported the referendum as a tool to renovate democracy. Both well-established parties as well as new fringe movements often discussed their stances on the referendum during publicly held party congresses, which were widely covered in the newspapers. As mentioned above, the wish to install the referendum did not turn out to be an exclusively leftist or progressive position. On the contrary, following the early referendum debates in the late 1960s, the liberal vvd had been the first major national party after D’66 to seriously consider the idea of a referendum. At its national congress of 1969, a section of the party pled for obligatory referendums on constitutional amendments and facultative referendums on individual laws.⁵⁹ This point of view had previously been associated with the socialist sdap.

In the 1970s, several new, albeit minuscule, one-issue parties emerged that argued for a revision of the postwar democratic order through the introduction of the referendum into the Dutch model. The tiny Referendum Party Nederland (rpn), for example, was founded in 1975 by a dissident member of the PvdA to advocate for the introduction of a Swiss-like referendum on a national scale. The Conservatieve Volkspartij (Conservative People’s Party), founded in 1977 as a successor to the right-wing Boerenpartij, also fruitlessly advocated the introduction of the referendum in the Dutch parliamentary system.⁶⁰

Meanwhile, during and following the work of the Biesheuvel committee, many political parties that did have seats in Parliament started their own investigations into the potentiality of a citizen-initiated referendum.⁶¹ In 1984, the leadership of the PvdA ordered such an examination and two years later, the party argued in favour of abolishing the Senate, although it also partly retracted its support for the referendum.⁶² Simultaneously, the communist cpn also renounced its earlier taboo on referendums, as a result of the nuclear missile controversy, and a working group within the party demanded to examine the adoption of referendums in the Dutch political system. By doing so, the party let go of its previous argumentation that referendums would only result in the subversion of the representative model and that they could easily be controlled, a stance that had fitted well with the direct postwar democratic repertoire that citizens were generally easy to manipulate.⁶³

Finally, the rise of certain extra-parliamentary action committees proved that the discussion on the referendum did not strictly adhere to official party positions. For example, in 1981, four individual parliamentarians of cda, vvd, D’66, and the Democratisch Socialisten ’70 (Democratic Socialists; ds’70) joined the ‘Actiecomité Referendum Ja’ (Action Committee Referendum Yes), even though their parties viewed the referendum in completely different ways. The cda, in its electoral programmes, ignored the referendum, spoke of the primacy of the representative model instead, and expressed that participation ran via elected representatives.⁶⁴ vvd and D’66 supported an investigation into a consultative referendum by a state committee, although D’66 was also eager to strengthen representative democracy.⁶⁵ ds’70, finally, strongly opposed the binding referendum as it would hurt the representative character of Dutch democracy, but the party kept the door open for a consultative referendum, without in any way committing to it.⁶⁶

The Action Committee Referendum Yes, founded by the organisation Civis Mundi and presided by the Rotterdam-based professor of law Wim Couwenberg, favoured experimenting with referendums to complement the Dutch model of representative democracy.⁶⁷ According to Couwenberg, in 1986, the political parties themselves were seriously hindering the development of a satisfying model of citizenship, by holding on to a purely representative system.⁶⁸ Ultimately, the committee became mostly known for its connection to the far-right Centrumpartij (Centre Party; cp) that actively resisted the influx of immigrants into the Netherlands. Although the committee had originally not served this purpose, the cp members who had joined the Action Committee Referendum Yes were mostly interested in holding a referendum on the presence of foreign immigrants in the Netherlands.⁶⁹

Overall, PvdA, D66, and cpn eventually declared that they would welcome a referendum within the Dutch political model, while vvd and cda reported that they were not convinced of Biesheuvel’s advices.⁷⁰ Most prominently, Prime Minister Ruud Lubbers did not support the broad call for a more participatory model of democracy. In their electoral programmes, vvd refused the referendum, stating the Netherlands ought to remain a representative democracy.⁷¹

In this context, it is important to note that the incumbent vvd and cda, who constituted a centre-right government from 1982 to 1989, did not contest that the Dutch democratic model required an update. However, they believed the solution ought to be found not in an enlargement of citizen participation, but rather in an expansion of citizen control over Parliament. They held the view that the distorted connection between state and citizen, of which Biesheuvel had spoken in his report, had been resolved already, as a result of the introduction of the ‘Nationale Ombudsman’ and the ‘Wet Openbaarheid van Bestuur’ (Government Information Act; wob).⁷² These measures were meant to increase the transparency of the political process, while at the same time not infringing on Parliament’s lawmaking and agenda-setting monopoly.

Considering these stances, it is clear that none of the abovementioned proponents or opponents of the referendum really wanted to override the representative primacy of Dutch democracy. Instead, most serious advocates of the referendum saw it as a tool to complement parliamentary democracy by enhancing citizen participation and citizen control over Parliament. Their main arguments were in line with what the Biesheuvel committee had noted, namely that there existed a clear gap between elected representatives and ordinary citizens. The referendum would then allow citizens to both voice their concerns more effectively, but also to better check the work of their parliamentarians. The opponents of the citizen-initiated referendum wanted to reach the same goals, but believed that citizens directing the agenda-setting process would infringe upon Parliament’s role.

At the same time, it is important to note that besides these concerns about citizen participation, many of the most ardent supporters of a referendum, for example those in the pacifist movement, mostly pleaded for an incidental referendum on individual topics. The referendum discussion, therefore, was not only informed by critiques on distanced democracy, but also by very topical issues that dominated the political debate at the time.

Although Parliament voted down the referendum advice of the Biesheuvel committee, the discussion did not stop. The third Lubbers government (1989-1994), which did not entail a cooperation between cda and vvd, but between cda and PvdA, stated in its governmental programme that it did not necessarily close the door on democratic innovation.⁷³ In Parliament, D66 leader Hans van Mierlo subsequently used this reference to call for a new committee that would investigate all kinds of potential democratic innovations, without mentioning the referendum specifically. This call was honoured and resulted in the ‘Commissie Vraagpunten’ (Question Committee), presided by cda politician Wim Deetman, which had to identify problems in the Dutch democratic model and suggest further research into potential solutions.⁷⁴

Historians and political scientists have given multiple reasons for the increased attention for democratic innovation in the early 1990s. These include the rise of more extremist political parties such as the cp, declining voter turnout, particularly in 1989 and 1990, and declining memberships of traditional parties.⁷⁵ When we look more closely at the analysis presented by the Deetman committee, however, a more existential problem with Dutch democracy, already indicated by Biesheuvel, resurfaces. The committee observed serious tensions in the relationship between citizen and state. The increased governmental interference with people’s lives made citizens wonder on what legitimacy this intervention was based.⁷⁶ As a result, Deetman thought it would be beneficial to investigate ways of taking decision-making more directly to the people, for example through the referendum.⁷⁷ Arguing for this consideration, he posited a higher rate of educated citizens demanding more popular influence, the fact that political parties had changed from societal movements with a broader mandate to parties with a number of defined standpoints, and the fact that the political class in The Hague had become so specialised and introspective that citizens felt much less of a connection to politics than before.⁷⁸

Following this and other recommendations, yet another committee was established under the leadership of the staunch Christian Democrat Jan de Koning who strongly opposed the referendum. However, he found himself on a committee with several progressive referendum enthusiasts including Thom de Graaf (D66) and Marius Ernsting (cpn/Groenlinks). As a result, in 1993, the majority of the committee, excluding its president, argued for a pro-referendum proposal that was almost identical to Biesheuvel’s. As their main argument, they highlighted a decreasing interest in the more traditional forms of politics and a broad societal need for more direct citizen intervention. Additionally, the committee referred to experiments with local referendums, which had increased the interest in direct democracy.⁷⁹ In the early 1990s, following the disastrous voter turnout of the 1990 municipal elections, several cities had indeed experimented with more participatory forms of democracy by holding local referendums. The goal had been to rekindle the lost interest in politics and the concept of holding a referendum was clearly much more important for the city councils involved than the actual subject. This was mostly evident in Leiden and Amsterdam, where referendums were announced even before a subject had been chosen.⁸⁰

Although three committees had supported an introduction of the referendum, a parliamentarian debate in December 1993 showed that broader political support for this tool was missing. A proposal by PvdA to introduce the referendum was voted down.⁸¹ The real breakthrough came when D66 doubled its seats in the 1994 elections and became a governing party. It demanded a constitutional change that would allow for binding referendums. PvdA, which was pro-referendum, and vvd, which was anti-referendum, subsequently agreed to cooperate.⁸² The proposal was initially guided through Parliament quite smoothly. However, for Dutch constitutional renewals it is necessary that after the first voting round, a second round is held, in which a two-thirds majority is required to pass the renewal. Moreover, between these rounds, new elections are to be held. Despite a serious loss of ten seats for D66, the proposal made it through the House, but in the Senate a problem arose. A so-called ‘gang of five,’ consisting of five dissident vvd senators, could obstruct the bill.⁸³ Prime Minister Wim Kok personally came to the Senate to defend the amendment, but the primary dissident, the constitutionally conservative vvd politician Hans Wiegel, torpedoed the constitutional change. The next day, the entire government resigned and only after negotiations did the parties decide to continue their cooperation. Peace was bought by promising D66 a ‘temporary referendum law’ that allowed for advisory referendums. These did not necessarily require a constitutional reform and constituted the one option that had not been warmly recommended by any state committee.⁸⁴

Eventually, the parliamentary primacy of Dutch democracy, which had never really been questioned, even by pro-referendum enthusiasts, thus remained inviolable, despite democracy having changed from its much more disciplined postwar character. Although the debate on the referendum had been an expression of resistance against the disciplined nature of postwar democracy, its more significant proponents had never meant to override the primacy of parliament. Instead, on the one hand, individuals and committees had used this debate to effectuate change on particular issues that were important to them, while, on the other hand, a constantly varying array of political parties and committees employed it to highlight the gap they perceived between the people and the State. Because this gap was eventually supposedly closed by measures that better fitted the parliamentary primacy of Dutch democracy, the referendum did not make it.

Up until the late 1960s, the Netherlands had never had much of a debate on the referendum, due to the fact that the representative model had always been the primary expression of democratic government. Parliament held a special position in its institutional order and any direct or more participatory forms of political decision-making were deemed unnecessary and potentially dangerous. In the 1920s, the referendum was therefore called a ‘foreign plant’. This characterisation rang true after the Second World War, when the Netherlands was a typical disciplined democracy, in which an educated elite was considered the most capable to take political decisions.

Much like in other Western European states, however, a growing resistance against the distanced, indirect character of the postwar democracies led to serious political and societal debates about whether the referendum could bridge the supposed gap between parliamentary representatives and the needs of those they represented. Consequently, multiple formal investigations into the referendum were set up, as well as several serious proposals from political elites to introduce it as an addition to the parliamentary model.

Additionally, individual citizens, movements, and committees argued in favour of the referendum, but often only on very specific issues. Here, party lines were frequently crossed, as members of different political ideologies could find each other on topical subjects, such as immigration, or the placement of American nuclear missiles. Certain political parties, including the larger parties of the left and smaller parties on the right, also more broadly argued in favour of the referendum. Still, they never questioned that Parliament was and ought to remain the dominant agenda-setting actor in Dutch parliamentary democracy. There was no political will to undermine the supremacy of Parliament, but only to improve the relationship between Parliament and citizens and to hold it accountable in a more representative way.

Finally, the referendum did not make it, while other reforms that better fitted the parliamentary nature of Dutch democracy, such as the ‘Nationale Ombudsman’ and the wob were introduced. Even when the government almost introduced the referendum in the 1990s, one vvd senator blocked it, expressing that it would be an infraction of the parliamentary model. Different from many other Western European states, in the Netherlands, the referendum was thus not formally institutionalised in the twentieth century. The referendum remained a foreign plant.