In this article I ask whether referendums diminish the quality of democracy in modern liberal democracies or do they encourage greater participation and better deliberation in the political process? I review arguments in democratic theory for and against referendums will be and typologies of referendums. In my review of the good, bad and ugly referendums; case studies, I review Ireland’s experience of the good, bad and downright ugly in its abortion referendum; the “good” example of the referendums in Switzerland and the good, bad, ugly and perhaps downright catastrophic of referendums relating to the EU: those on European integration that took place in 2005 in France and the Netherlands and the UK’s “Brexit” referendum in 2016. It concludes by discussing voter competency in referendums.
In a referendum the people decide directly on some issue, rather than electing representatives to make decisions on their behalf (Gallagher et al, 2013). Between 1945 and 2010, over 660 referendums took place in the 30 countries in Europe surveyed by Gallagher et al with one country – Switzerland – accounting for two-thirds of them. The use of referendums increased steadily in the second half of the 20th century (Gallagher et al, 2013) due partly to the expansion of the European Union and its integration imperatives and the increased salience of post-materialist issues such as environmentalism (Bjorklund, 2009).
The concept of referendum refers to a wide range of institutions that generate a variety of political interactions. Within the political system, it is useful to make the distinction between decision-controlling and decision-promoting referendums. In decision-promoting referendums, the initiative to hold a referendum is made by the political actor who has formulated the policy proposal to be voted upon. These can either be parliamentary majorities or popular or citizen-initiated referendum procedures. Popular initiatives, most hosted in Switzerland, can be interpreted as decision-promoting since the organisation or popular movement that promotes a referendum also draws up the policy proposal that is to be the subject of the referendum (Setala, 2009).
Decision-controlling referendums can be categorised as abrogative or rejective; they are deployed as a check on a policy proposal already passed by a legislative body. Abrogative referendums are held on enacted laws, while rejective referendums are held on laws passed but not yet in force. Both types of decision-controlling referendums may be demanded by signatories to a popular petition (e.g. Italy and Switzerland) or by other actors such as parliamentary minorities or regional governments (Setala, 2009).
The ideological debate on the referendum also feeds into wider debates within democratic theory. The proposal engages with a theoretical model of four normative models of democracy: the representative mode, the associative model, the deliberative model and the participatory model. It also examines the nature of citizen participation within each model (Michels, 2009).
Arguments for and against the use of referendums of democracies are largely based on the potentially beneficial or detrimental effects of such polls on the operation of representative government. The case for referendums can be summarised as ‘maximising legitimacy, maximising the human potential of citizens and the ending of alienation and apathy.’ The arguments against amount to the lack of analytical skills on the part of ordinary citizens to make wise decisions; when elected officials make decisions they weigh preferences and weld legitimate group interests into fair and equitable policies for all; rights of minorities are more likely to be guaranteed by the decisions of representatives and the referral of divisive issues to the public through use of the referendum device will weaken the functioning and esteem of representatives and representative government ( Uleri, 1995). These arguments are heard most frequently in Great Britain where belief in the sovereignty of parliament and its concomitant right and duty to make decisions for the people has been almost sacrosanct (Crepaz and Steiner, 2011).
To test the validity of these competing arguments on the desirability of referendums, I now examine some case studies. The first concerns Switzerland which provides a relative success story for the referendum. Switzerland is a federalist country with 26 autonomous units of government or cantons. It has four official languages which cut across the Protestant-Catholic religious divide, no dominant capital and ancient democratic traditions around pastures form which the founders of the modern Swiss constitution in 1848 were able to draw upon. The people (not judges) have the ultimate right to determine the constitutionality of a particular law through referendum for which the only requirement is that 50,000 signatures be collected; the voters are also final arbiters on constitutional amendments and a minimum of 100,000 voters can also submit a constitutional amendment of their own (Crepaz and Steiner, 2011).
The greatest strength of the referendum is the legitimacy it gives to political decision. Although, notoriously, female suffrage was not granted by referendum until 1971, women’s participation in politics became rapidly accepted by male voters, and since 2007 women have occupied three of the seven seats in the Federal Council, the Swiss cabinet. Innovative ideas have passed by referendum such as the Alp Initiative concerning heavy trucks passing through Switzerland (Crepaz and Steiner, 2011.
Swiss voters displayed a maturity in not displaying xenophobia when rejecting constitutional initiatives in the 1970s proposed by an anti-aliens movement which would have forced hundreds of thousands of foreigners to leave (Crepaz and Steiner, 2011) Arguably such maturity was not in evidence in 2009 when an initiative to prevent the future building of minarets was endorsed in a referendum despite the opposition of Muslims and the major parties (Gallagher et al, 2011).
The effectiveness of referendum initiatives in Switzerland in providing checks is borne out by an analysis of Swiss national elections and national votes from 1971 to 2005 which found a clear gap between the citizenry’s policy preferences expressed in elections and those expressed in popular votes (Sager and Buhlman, 2009).
Prior to the historic vote this year to remove the anti-abortion Eighth Amendment from the Constitution and to usher in a liberal abortion regime permitting terminations in the first trimester of pregnancy, the experience of the mediation of Ireland’s culture wars on abortion and divorce provided textbook examples of the functioning of the referendum as a conservative device in politics (Gallagher, 1995) and so a direct opposite to the Swiss experience.
For, in contrast to Switzerland, Ireland is one of the most centralized countries in Europe and was until recently a fervently Catholic and monocultural society. It was in just such an environment that Ireland’s political elites acceded in 1981 to a demand by the conservative Catholic coalition, the Pro-Life Amendment Campaign (PLAC), for an amendment to outlaw abortion to be inserted into the Constitution. The Eighth Amendment proposal was eventually voted upon in September 1983 and passed by a majority of two-to-one of those who voted. The history of Irish abortion politics between the passing of the amendment or Article 40.3.3 and its deletion is pretty familiar to TPQ readers so there is no need to recount it. Suffice to say that court rulings of Article 40.3.3 led to a virtual shutdown of the activities of family planning and student welfare services in relation to abortion advice and counselling and the globally reported X-case in 1992 restraining a suicidal child rape victim for travelling to GB for an abortion and the death of an Indian national in childbirth in 2012 in hospital both the result of religiously fundamentalist interpretations of the law of the most grotesque kind. These events galvanised one of the most transformative movements in contemporary liberal democracy.
Referendums on European integration are often in effect “second-order national elections; they are second-order because they are low salience, and first-order issues of national politics tend to dominate the campaigns. Consequently, the electorate uses their votes to demonstrate their feelings towards their government, (Hobalt and Brouard, 2011).
This ‘kick the cat’ mode of “sticking one on the establishment” was definitely a considerable, if hard to quantify, factor in the Brexit referendum in 2106 Many of the toxic elements evident in the “Second Partitioning of Ireland (Hesketh, 1990) in 1983 were present in 2016: a nativist cry against agents of globalisation, cosmopolitanism and the national elites held to be acting as their midwives (The European superstate in 2016; Planned Parenthood in 1983; apocalyptic scenarios of 70m Turks and hordes of Middle Eastern refugees coming to Britain in “Breaking :Point in 2016 and the “blood dimmed tide” and “the abortion mills of England grind the bodies of Irish children in 1983; a most unideal speech situation: the seduction of millions into tabloid newspaper and some popular culture narratives (often derivative of World War II memories) of plucky, independent Britain in thrall to EU tyranny in 2016; in 1983 the discourse and practice of almost an entire nation crippled by an historically aggrandising Catholic Church; the blatant lying by the victorious side in 1983 and 2016 and the inability of the losing side to tell a compelling story in response being perpetually on the defensive and lastly, and arguably, the most harmful of all to the functioning of democracy, the dereliction of duty by the winning sides in 1983 and 2016 to detail a pathway to implement a referendum result unlike the winning side in Ireland in 2018 and the losing side in Scotland’s independence referendum in 2014. The consequences of the latter were that in Ireland the judiciary and in Britain the executive sought to impose on the country hard line and rigid interpretations of the Will of the People which generated counter-reactions from the losing sides (Repeal the Eighth Amendment, People’s Vote) who ironically seek to use the referendum device in pursuit of their aims.
The opportunity afforded by the referendum device to voters to register discontent in an arguable “safer”, less “destructive” sway was evident This phenomenon in the two referendums held to ratify the European Constitution Treaty (ECT) in France and the Netherlands held within days of each other on 29th May and 1st June 2005 respectively. In both countries’ ratification of the ECT was backed by centre-right governments and centre-left oppositions and high levels of public support. In both countries though, the ECT was rejected after differing campaigns in terms of length (Hobart and Brouard, 2011).
Analysis of the voting choices in both referendums do not convey a uniform Euro-skepticism. For European attitudes became intertwined with domestic concerns. In France voters wished to convey their desire for a social Europe to evolve and fears of the evolution of neo-liberal European economy plus their discontent with the incumbent centre-right administration. Party cues were more salient in the Netherlands where the campaign was shorter and voters displayed more concern about loss of Dutch economy and greater enthusiasm for postmaterialist issues such as environmental protection (Hobart and Brouard, 2011).
In conclusion, no institution in modern democracies brings us closer to the ideal of ‘self-governance’ by the mass public than referendums (Hobolt, 2009). Any assessment of the impact of the referendum on the quality of representative government, I would argue, hinges on the competence of voters to make collective decisions. But the experience of the Brexit result in 2016 and of the repeal of the Eighth Amendment in 2018 completely overturned received wisdom on the nature of referendum contests and the deliberative culture in both the UK and Ireland. In the case of the UK; its traditional doctrine of the sovereignty of Parliament was turned on its head as Brexiteers aggressively the Will of the People against the “unelected” judges who temporarily reined in Prime Minister’ Theresa May’s head long rush towards a “Hard Brexit” by insisting on a “meaningful vote” after a case taken by the City of London investor Gina Millar. In the case of Ireland, the orthodoxy about the historic moral monopoly exercised by the Catholic Church on Irish discourse and practice (Inglis, 1998) not equipping the electorate to fully understand the potential repercussions of inserting a ‘pro-life’ amendment into the Constitution in 1983 was completely swept away in 2018. Then the broad-based consent across the main political parties, the high deliberative quality of the work of the Citizen’s Assembly representing such a diversity of Irish civil society, the generational and transformative impact of the women’s movement on Irish civic culture and, last but certainly not least, the effect of the personal stories of the women affected by the cruelty of the Irish abortion regime. The high levels of civic awareness in Switzerland made electors more ‘competent’ arbiters in referendums. In the case of European integration referendums, the political competence of the electorates was not to be doubted but the mixture of issue preferences conveyed in the ‘No’ votes and the starkness of such binary choices, the consequences of which are played daily in the UK’s existential crisis over Brexit, makes workable compromise solutions difficult for politicians. A possible solution may be the holding of simultaneous referendums across Europe in the future (Hobolt, 2009).
To answer my own question, referendums do not have to lead to the vetoing of minority viewpoints and the closing down of debate by the triumph of the General Will as happened in Ireland over the abortion issue after 1983. It is the question asked on the ballot paper and the fullest understanding of the consequences of the voter’s preference in the terms of the implementation of the decision that determines the democratic validity of the plebiscitary device not a lack of cognitive and intellectual capacity of the plebs.
Bjorklund, T (2009) The Surge of Referendums and the New Politics Approach in Setala, M. and Schiller, T (Eds.) Referendums and Representative Democracy. Responsiveness, Accountability and Deliberation. Routledge/ECPR studies in European Political Science Abingdon, Oxon: Routledge
Crepaz, M.L. and Steiner, J. (2011) European Democracies. Seventh Edition London: Longman
Gallagher, M., Laver, M. and Mair, P. (2011) Representative Government in Modern Europe. Fifth Edition Maidenhead: McGraw-Hill
Gallagher, M. (1996) Ireland: The Referendum as a Conservative Device? in Gallagher, M. and Uleri, P. (Ed.) The Referendum Experience in Europe London: Macmillan
Hobolt, S. (2009) Europe in Question. Referendums on European Integration. Oxford: Oxford University Press
Hobolt, S. and Board, S (2011). Contesting the European Union? Why the Dutch and the French Rejected the European Constitution. Political Research Quarterly, 64(2), 309-322
Hesketh, T (1990) The Second Partitioning of Ireland. The 1983 Abortion Referendum Dublin: Brandsma Books
Inglis, T. (1998) Moral Monopoly. The Rise and Fall of the Catholic Church in Ireland.
Michels, A. (2009) Ideological Positions and the Referendum in the Netherlands in Setala, M. and Schiller, T. (Eds.)
Sagar, F. and Buhlman, M. (2009) Checks and Balances in Swiss Direct Democracy in Setala, M. and Schiller, T. (Eds.)
Setala, M. (2009) Introduction in Setala, M. and Schiller, T. (Eds.)
Uleri, P. (1996) Introduction in Gallagher and Uleri (Eds.)
➽ Barry Gilheany is the author of a PhD thesis Post-Eighth Abortion Politics in the Republic of Ireland from Essex University, Department of Government. He is also the author of The Discursive Construction of Abortion in Georgina Waylen & Vicky Randall (Eds) Gender, The State and Politics Routledge, 1998.