Recently the use of constitutional referendums has become increasingly common in Europe and beyond. National governments tend to initiate or support referendums to settle constitutional questions. For instance, in 2013 in Croatia the electorate was asked to, and indeed did, define marriage as a union between a woman and a man. Two years later, Irish voters approved changing the Constitution to extend civil marriage rights to same-sex couples. In 2016 Italian voters refused the extensive constitutional reform proposed by the Italian government. The British people decided to leave the EU after their government put the Brexit decision to a referendum vote. The Hungarian government asked the electorate to refuse the EU refugee allocation scheme to protect the country’s “constitutional identity.” This spring a referendum was held in Turkey on constitutional changes that crucially expanded presidential powers. And we all remember the illegitimate Crimean status referendum of 2014.
These examples show that the result of a referendum can be as critical as that of a general election. Sometimes it is even more critical, when changing the country’s constitutional setting is at stake. Theoretically, the referendum is an expression of the sovereignty of the people. Ostensibly one of the main goals of holding a referendum is to ensure the participatory rights of the citizens. But in practice referendums do not always serve democratic purposes: they are very often populist, manipulative or even lead to antidemocratic results. In other words, in some cases the referendum is not about using but abusing popular sovereignty. Nevertheless, there are no internationally recognised legal standards and mechanisms concerning constitutional referendums.
Important substantive and institutional guarantees ensure the democratic quality of the general elections. The most important election principles are enshrined in international human rights documents, for instance in Article 25 of the ICCPR and Article 3 of the First Protocol to the ECHR. Even individual petitions can be submitted to the ECtHR in the case of the violation of the right to free elections. In addition, international election observations are regularly conducted by the OSCE to ensure that the basic principles are fulfilled.
In the case of a referendum these substantive and procedural guarantees are almost completely missing. Only international soft law deals with the question of the democratic quality of the referendum. In 2005 the Council of Europe’s Parliamentary Assembly issued the Recommendation “Referendums: towards good practices in Europe.” A year later the Venice Commission adopted a “Code of Good Practice on Referendums.” The Commission underlines that the internationally recognised fundamental principles of electoral law, including universal, equal, free and secret suffrage apply mutatis mutandis to referendums. Furthermore, there are soft law principles which apply in a special way to referendums. The judicial review in the case of a referendum includes the validity review of the texts submitted to referendum. And there are democratic requirements that are specific to referendums, like for instance the requirement of clarity and unambiguity of the texts submitted to the electorate. This year the Venice Commission issued a compilation on its opinions and reports concerning referendums to provide a source of reference for national law-makers. Taking these soft law requirements into account, the OSCE can open referendum observation missions following an invitation by the national authorities.
This is what happened recently in Turkey. Simply put, this spring Turkey held a referendum on transforming the country from a parliamentary system into a limitless presidential one. The OSCE held in its preliminary conclusions that holding the referendum was contrary to international good practices: the legal framework was “inadequate for the holding of a genuinely democratic referendum”, “fundamental rights were restricted by extraordinary state of emergency powers” and that the referendum “took place on an unlevel playing field and the two sides of the campaign did not have equal opportunities”. Nonetheless, the referendum has already had wide-ranging implications. It was a stepping stone for Turkey to shift into an unchecked presidential system.
Although the OSCE did not open referendum observation missions in Hungary, it is worth taking a look at the referendum on EU quotas for relocating asylum seekers. The government opposed the quotas and was campaigning heavily against them. Although the referendum result was invalid because of the low turnout, the prime minister nevertheless declared victory because among valid ballots, 98 per cent said no to the EU quota system. The fairness of the procedure was highly controversial, since the referendum question was deceptive and unconstitutional. Moreover, state-sponsored xenophobic propaganda backed the referendum and the government was using state offices to mobilise voters and threatening non-compliant local officials with funding cuts.
While the Turkish and the Hungarian political systems differ in many aspects, there are similarities between the two referendum procedures. Both countries were in a state of emergency when the referendums were held. The governmental campaigns were one-sided, and the processes were manipulative. Consequently, neither the Turkish nor the Hungarian electorate were adequately informed on the matters to be decided. It is also true in both cases that the opposition campaigns were starved of coverage by state-run media and the governmental campaign rhetorics cast their opponents as terrorist sympathisers. In short, these referendums did not conform with international guidelines.
It appears that international soft law does not provide a sufficient legal basis for evaluating constitutional referendums. The aim of the currently existing good practices is not to serve as a standard with normative strength, but as mere guidelines for national law-makers. Moreover, consultative bodies like the Venice Commission and the OSCE certainly do not have the power to impose sanctions or force a re-run of the vote when the constitutional referendum goes against international good practices. And we can hardly find any judgments of the ECtHR concerning referendums; thus, there is a very little chance of the recent application of the Turkish opposition succeeding.
It is time to consider making international referendum law a distinct body of law. But how should international norms regulate constitutional referendums? In order to answer this question we need to investigate the history and the recent practice of constitutional referendums. It is equally important to identify the legal developments. The existing international soft law may form the basis of the international legal standards on constitutional referendums. And setting out the theoretical foundations and the justifications for the establishment of international referendum law is also essential.
Formulating international legal standards by which states can hold a constitutional referendum would be beneficial in many ways. It would prevent states from holding constitutional referendums during situations of emergency. In a state of emergency the normal functioning of state institutions and the mass media may be limited, and very often there are limitations on the exercise of political freedoms. Clearly, under such circumstances the democratic process of holding a referendum cannot be guaranteed. Additionally, an international referendum law would prevent states from using constitutional referendums to restrict the fundamental rights of vulnerable minority groups (LGBT people, refugees, etc.).
There is a need to develop international referendum law which would govern national referendum laws. The newly adopted international legal standards would list the essential procedural and institutional as well as substantive requirements for holding a fair and democratic referendum. Observing the internationally accepted procedure would mean outcome of a referendum could be accepted as legitimate both within the country and beyond.