Understanding the 2015 marriage referendum in Ireland: context, campaign, and conservative Ireland

ABSTRACT

On 22 May 2015 the marriage referendum proposal was passed by a large majority of Irish voters and the definition of marriage in the constitution was broadened to introduce marriage equality. This referendum is remarkable for a number of reasons: (1) it is uniquely based on an experiment in deliberative democracy; (2) the referendum campaign was unusually vigorous and active; and (3) the voting patterns at the referendum point to a significant value shift along the deep seated liberal conservative political cleavage of Irish politics. This article provides an overview of the background to the referendum initiative, the campaign prior to the referendum, and the key factors that drove voter turnout and preference. Based on a post-referendum survey, we find that while support for the government of the day, political knowledge, and social attitudes have the same effects as commonly found in other referendums, the variation among social classes was less prevalent than usual and door-to-door canvassing by the two sides of the campaign impacted through turnout rather than vote preference. The voting behaviour of the different age groups suggests strong generational effects.

KEYWORDS:

• Political campaigns
• electoral behaviour
• referendums
• constitutional convention
• marriage equality

Acknowledgement

We are grateful to the referees for helpful feedback and comments on an earlier draft of this paper. The usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCiD

Johan A. Elkink http://orcid.org/0000-0002-8275-533X

David M. Farrell http://orcid.org/0000-0002-0528-4335

Theresa Reidy http://orcid.org/0000-0002-5380-0015

Jane Suiter http://orcid.org/0000-0002-2747-8069

Notes

1. http://www.huffingtonpost.com/eamon-gilmore/is-same-sex-marriage-the-civil-rights-issue-of-this-generation_b_7467986.html, 29 May 2015 (accessed 1 July 2015).

2. At this meeting the Convention in addition recommended that relevant legislation should be introduced to address the parentage, guardianship, and upbringing of children in families headed by same-sex married parents. The government also acceded to this request and the Child and Family Relationships Act was signed in April 2015.

3. https://www.constitution.ie/AttachmentDownload.ashx?mid=c90ab08b-ece2-e211-a5a0-005056a32ee4 (accessed 1 July 2015).

4. ‘Emigrants take planes, trains and automobiles #HomeToVote’, Irish Times, http://www.irishtimes.com/life-and-style/generation-emigration/emigrants-take-planes-trains-and-automobiles-hometovote-1.2222151, 23 May 2015 (accessed 1 July 2015).

5. There is no permanent national referendum study. Piecemeal efforts to explore the dynamics of voter decision-making arise periodically, but almost always when a government loses a referendum vote on an issue of political significance, such as a referendum related to European integration (Nice I or Lisbon I) or most recently on Oireachtas inquiries (Marsh et al., 2012). On other occasions when governments win referendums or the referendum is of lesser political significance, they are not sufficiently exercised to commission basic research. To date, it has not been possible to secure funding for a permanent referendum study from the Irish Research Council.

6. For more on this study, see https://referendumstudy.wordpress.com.

7. This form of modelling avoids the ‘kitchen sink’ approach to regression analysis: adding all variables to the same model would result in inappropriate controls for variables early in the causal chain.

8. The final model was also evaluated with a set of dummy variables that reflect party support as an additional control variable. This did not alter any of the findings reported and party support itself did not have a significant effect on either turnout or vote choice. Results available upon request.

9. It is clear, however, that in the case of the Fianna Fail party there were quite serious internal differences (resulting in the high profile resignation of Senator Averil Power from the party in the days following the referendum).

10. Except, perhaps, indirectly, with the strength of the support for either the Yes or the No side impacting on the probability of turnout in the referendum (Elkink & Sattler, 2013).

11. The oldest age category is somewhat more difficult to compare, since ‘65+’ would include both those close to 65 and those much older.

12. See, for example, the analysis by Michael Gallagher at http://politicalreform.ie/2015/05/27/on-the-same-sex-marriage-and-presidential-age-referendums/, 27 May 2015 (accessed 1 July 2015).