Abstract
While power-sharing arrangements are often commended for establishing peaceful relations between major ethnic groups, they are also criticised for excluding ‘Others’. Nevertheless, more complex forms of party competition can emerge in power-sharing systems, including parties representing the dominant communities (‘ethnic parties’) seeking to engage Others (‘non-dominant’ groups). Drawing on semi-structured interviews with parties from Northern Ireland, we examine the extent to which dominant parties reach beyond their core ethnic constituencies, how and why. We consider increasingly salient non-sectarian issues, such as marriage equality and abortion, and explore how ethnic parties have sought to respond to these debates. We consider whether liberal forms of power-sharing influence the willingness of dominant parties to advance inclusion of non-dominant groups. Our findings suggest that under favourable conditions, flexible power-sharing can create space for incremental moves by ethnic parties to reach out to constituencies beyond their core, gradually moving the system towards more inclusive representation.
In November 2015, the Northern Ireland Assembly voted by a narrow margin – 53-52 – to extend the right to marry to same-sex couples. Subject to a Petition of Concern motion, a mechanism whereby legislation must gain assent from both nationalist and unionist parties, the legislation failed to pass. Among the majority vote, only four came from unionists, effectively vetoing the proposal. Marriage equality is an issue-area that, ostensibly, should span the ethnic divide. Yet the issue is profoundly ethnicised with Northern Ireland’s largest unionist party, the Democratic Unionist Party (DUP), opposing gay marriage, and the nationalist parties, including the largest, Sinn Féin, vocally in support. Accordingly, a voter who supports both LGBTQ (lesbian, gay, bisexual, transgender and queer/questioning) rights and Northern Ireland’s existing constitutional status would appear to have nowhere to comfortably park their vote.
A woman’s ability to secure safe, legal and local access to abortion is similarly a hotly contested issue in Northern Ireland, which until 2020 had some of the most restrictive reproductive laws in Europe. When Sinn Féin started to liberalise its position and support abortion up to 12 weeks gestation, DUP leader Arlene Foster claimed that some nationalists were defecting to the DUP as a result (Independent, 2018). Supporters of a woman’s right to choose, regardless of their views on the constitutional question, struggled to find a political party to call home.
These examples highlight a central dilemma within consociationalism: its tendency to marginalise identities and issues beyond the ethno-divide. Consociationalism starts from the assumption that no single group should be privileged in governing institutions and that broad inclusion is needed to ensure democratic and stable outcomes (Lijphart, 2008; McGarry and O’Leary, 2005, 2009, 2013). Institutionally, consociationalism includes executive power-sharing, proportionality, mutual vetoes and group autonomy (Lijphart, 1977). These institutions can either predetermine which groups share power (corporate consociation) or they can allow groups to self-determine the terms of their participation (liberal consociation) (McGarry and O’Leary, 2007: 675). Under liberal rules, voters can self-identify and are not constrained onto ethnic voting rolls. Party mobilisation may proceed on an ethnic basis, but this is not mandated beforehand.
Despite this starting assumption about the need for inclusion, emergent research has begun to highlight a major blind spot in consociational theory and practice: what happens to those groups and individuals not directly included in the power-sharing system? Increasingly, the catch-all term ‘Others’ is used to describe this diverse range of groups and individuals, including those who belong to an ethnic group not recognised in the agreement, those whose political identity is framed around an aspect of their social identity other than ethnicity and those who prefer ideological forms of political participation (Agarin et al., 2018). Others potentially face institutional barriers to their representation, participation and rights provisions in consociations.
Thus far, research in this area focuses on how Others mobilise in response to their exclusion (Agarin et al., 2018; Nagle, 2016, 2018; Stojanović, 2018). Nagle (2018), for example, outlines three available strategies for Others: hegemonic compliance, whereby consociational logic is recreated within civil society organisations and movements; constructive engagement, where Others engage the system in order to push for power-sharing reforms; and active resistance, where they actively reject power-sharing. By contrast, there has been limited attention to how included parties respond to the exclusion of Others. In this article, we ask how ethnic parties – the dominant included actors in consociational systems – confront the exclusion of other issues and identities. Ethnic parties are those that draw support primarily from a specific ethnic group and which are organised to make demands on behalf of that group (Chandra, 2005). This presupposes that they are unlikely to protect the interests of those beyond their core. Yet, as members of the government, parties are also expected to govern for all once in power. As veto-wielding actors, ethnic parties also have an outsized role in directing the legislative agenda and as such have significant power to determine the fate of Others, who as a ‘statistically marginal and heterogenous group’ (Stojanović, 2018: 361) are unlikely to hold veto power or to reach a critical mass of representatives in office on their own. How the included treat the excluded is thus an important, if neglected, area of research on consociationalism.
This line of inquiry mediates between two contrasting perspectives on consociation’s ability to include Others and to adapt over time to represent those newly emergent forms of political expression. The first perspective paints a rather deterministic picture of consociation as division institutionalised, accusing it of seeing all politics through an ethnic lens (Nagle, 2016). This ‘all politics is ethnic politics’ perspective argues that little political space is available for Others under consociational rules. By mobilising on the basis of ethnicity, ethnic parties compel voters to pick a ‘side’ or risk exclusion. Such parties do not present universalistic platforms; their purpose is to ‘secure political, economic and cultural benefits and protections for the ethnic group’ (Gunther and Diamond, 2001: 183). Policy tends to be characterised by ‘seepage’ whereby issues unrelated to ethnicity – such as economic development, trade union affairs, land or tax policies – become ethnicised (Horowitz, 2000: 8). Consociationalism, this suggests, perpetuates ethnic politics at the expense of those who do not subscribe to the identities included in government. Stojanović summarises this first perspective: consociationalists ‘are willing to sacrifice the political rights of a statistically small group of citizens whose political accommodation is not necessary for maintaining peace and stability’ (Stojanović, 2018: 361; see also Brass, 1991: 334).
Advocates of the second perspective suggest that while corporate consociation can reify ethnicity, liberal forms – those that do not specifically name the groups to share power – hold the capacity to accommodate shifts in ethnic salience and allow society to transition towards a more civic form of representation. Indeed, the ‘all politics is ethnic politics’ perspective has been tempered by research suggesting that ‘ethnic outbidding is not inevitable’. Mitchell et al. (2009: 403) employ the term ‘ethnic tribune party’ to reflect the sort of party that can be ‘intransigent about identity’, yet prepared to share power (Mitchell et al., 2009: 403). Garry (2014) meanwhile describes the phenomenon of ‘ethnic catch-all voting’ whereby citizens may come to support parties from a rival ethnic bloc when they perceive that party capable of representing the interests of all communities. Even if ethnicity is the primary basis of voter mobilisation, this does not preclude space for Others or assume that ethnic parties will de facto ignore them. McGarry (2017: 282) summarises the second perspective: ‘liberal consociations need not privilege ethnic identities, and are more likely to create political space for previously weak and marginalized identity groups than conventional majoritarian systems of executive formation’.
While the inclusion of Others is an important consideration in all power-sharing arrangements, Northern Ireland is an apt case for assessing these contrasting perspectives. First, Northern Ireland most closely embodies the logic of liberal consociation. If any consociation is able to transcend the ‘all politics is ethnic politics’ critique, it is a liberal consociation, which is more likely to ‘treat as equally as possible individuals who subscribe to rival national identities, to no national identity, to nested national identities, or have other salient public identities which cross-cut national lines’ (McGarry and O’Leary, 2009: 37). This case study thus allows us to consider whether liberal consociation facilitates the conditions under which ethnic parties reach out to Others. Second, having adopted consociationalism with the signing of the Good Friday Agreement in 1998, Northern Ireland has a long track record of power-sharing practice. We are particularly interested in the 2007–2017 period during which there was a decade of uninterrupted power-sharing with two formerly ‘hardline’ parties at the helm, the DUP and Sinn Féin, both of which have significantly moderated their platforms over time (Mitchell et al., 2009). This 10-year period provides an ‘extended period of inter-group cooperation’ (McGarry, 2003: 295) for assessing whether political space for other issues and identities has opened up over this period. Third, this 10-year period of relative stability came to an abrupt end in 2017, leading to a 3-year suspension of the power-sharing institutions. During this time, Northern Ireland experienced a series of political shocks – ‘relatively sudden changes [which] affect the distribution of resources and power in a country’ (Kuperman, 2017) – including a corruption scandal in which the DUP was implicated, the death of Sinn Féin Deputy First Minist Martin McGuinness and a referendum result in support of the United Kingdom leaving the European Union despite a majority in Northern Ireland voting to remain. This later period allows us to consider if political space did widen to include other issues and identities in the 2007–2017 period and whether that space was resilient in the face of political crisis. It may well be that ethnic parties are prepared to reach out to Others when their inter-ethnic relations are running smoothly, but that their willingness retreats in moments of inter-ethnic tension.
This article offers a corrective to both perspectives. It gives some credence to the second account, which acknowledges that ethnic politics will likely dominate, especially in the short term, but that liberal consociational institutions can support a politics of reaching beyond the core. As parties learn to govern together, they can be expected to moderate their stance. Alongside growing ethnic moderation, a more complex party system configuration may develop. New parties representing new identities, ideologies and interests may form and more ‘bread-and-butter’ policy issues may rise in salience. While ethnicity may be the dominant cleavage, under this perspective, there remains political space, however small, for other forms of political mobilisation to develop. Zuber (2012: 932, 934) refers to ‘nested competition’ in this context, whereby ethnic parties in divided societies compete for votes not only within their own ethnic bloc, but also across blocs, with civic parties and on cross-cutting dimensions.
We find that there has been adaptation over time to accommodate an increase in salience of other issues and identities, but that that movement is highly constrained, contingent and precarious. Consequently, we find the first account overly deterministic. Yet, our findings also temper the optimism of the second account, which underestimates the ethnicising effects that (even liberal) consociation can exert and the vulnerability of this space when it does emerge. Such findings are of comparative relevance as Northern Ireland is not the only consociation grappling with how to include Others. The reasons for reaching out may differ across cases, whether it be in response to court challenges, such as the European Court of Human Rights rulings regarding Bosnia and Herzegovina’s presidential election rules, or to mobilisations from below as with the October Revolution protests in Lebanon, or to party system realignments as in North Macedonia. Nonetheless, the challenge is a shared one. As Stojanović (2018: 360) suggests, the exclusion of Others in consociations ‘raises important normative and legal issues for liberal democracies’. As a theory of democracy, consociationalism must consider these questions of inclusion and exclusion.
For this article, we conducted 22 semi-structured interviews between February and June 2018. Eleven interviews were conducted with political party representatives: 3 DUP, 2 Ulster Unionist Party (UUP), 2 Social Democratic Labour Party (SDLP), 2 Alliance and 2 Green Party. 1 Another 11 interviews were conducted with representatives of civil society organisations representing human rights, ethnic minorities, LGBTQ rights, reproductive rights and the Irish language. The article also incorporates analysis of four interviews we conducted with Sinn Féin in 2013 and 2015 and one with the DUP in 2013. This was supplemented by content analysis of key party documents. Party manifestos for the 2007, 2011, 2016 and 2017 Northern Ireland Assembly elections and the 2016–2021 Northern Ireland Executive Programme for Government were analysed and coded with respect to the issues of marriage equality, abortion and the Irish language. We paid particular attention in the interviews and documentary analysis to how the parties characterised their support for these three ostensibly non-ethnic issues. Each dominated political debate at various points during the 2007–2017 period and each has also been deeply contested by the ethnic parties, none perhaps more so than the Irish language. While intimately connected with Irish cultural heritage, it is also a language claimed as part of a shared linguistic heritage of the region (McCoy, 1997), with an increase in interest in the language among the Protestant community. 2 This documentary analysis was primarily used to inform the analysis rather than forming the substance of our findings. In what follows, we ask three questions about ethnic parties reaching beyond their core: To whom do they reach out? How do they do so? Why?
Ethnic parties and others in Northern Ireland: A changing relationship
Following 30 years of violent conflict, the 1998 Good Friday Agreement delivered power-sharing between nationalists and unionists within Northern Ireland and formalised relations between Britain and the Irish Republic. A devolved Assembly and Executive were established along broadly liberal consociational lines. Electors vote on a common roll under the Single Transferrable Vote, a system seen as proportional, relatively favourable to smaller parties and which held potential for both intra- and inter-bloc transfers (Jarrett, 2018: 90–91). The Executive is open to all parties with sufficient support, with portfolios allocated through the d’Hondt formula based on party strengths in the Assembly. There are some corporate elements: Members of the Legislative Assembly (MLAs) are required to designate as Nationalist, Unionist or Other to ensure guarantees for the two main communities. Key decisions require cross-community support (through ‘parallel consent’ or ‘weighted majority’) (Northern Ireland Act, 1998). Cross-community voting can also be triggered by a Petition of Concern, signed by 30 Members, which functions as an effective community veto. Such rules are perceived by some as an inequality between civic and ethnic representatives (Evans and Tonge, 2001), creating ‘two orders of Assembly members’: ‘those whose votes always “count” and those whose votes never do so’ (Wilford, 2010: 139). Five parties make up a 2 × 2 + 1 system: the DUP and UUP at the more extreme and moderate ends of the unionist spectrum, respectively; mirrored on the nationalist side by Sinn Féin and the SDLP. In the civic bloc, the Alliance Party represents the main player.
Initially, power-sharing was characterised by instability, centring on peace and security concerns such as weapons decommissioning. After a 5-year suspension from 2002–2007, the 2006 St Andrews Agreement gave way to 10 years of uninterrupted power-sharing (2007–2017). Indeed, following the 2014 Stormont House and 2015 Fresh Start Agreements, provision was made for parties opting not to take their seats in the Executive to form an official opposition – an option taken by the moderate unionist, nationalist and civic parties following the 2016 Assembly election. Yet, the post-2017 period saw the breakdown of the power-sharing institutions, precipitated by both internal and external conditions. The most notable external shock is the United Kingdom’s decision to leave the European Union following a referendum in June 2016, supported by 51.9% of the UK electorate, but only 44.2% of voters in Northern Ireland. With a division between nationalist and unionist parties and voters on the issue (Garry et al., 2017), Brexit has served to promptly kick the constitutional question – whether Northern Ireland’s future remains within the United Kingdom or in a united Ireland – back into play. The chasm between nationalist and unionist parties was widened further by a confidence-and-supply pact forged between the DUP and the UK Conservative Party following the 2017 UK General Election.
Internal conditions relate primarily to identity and broader social justice. While the main reason for Sinn Féin’s withdrawal from the Executive was DUP First Minister Arlene Foster’s involvement in the Renewable Heat Incentive scandal, the nationalist party also cited its partner’s refusal to agree to Irish language legislation and marriage equality. Analysis of manifestos from 2007–2017 shows that the Irish language had been a source of contention since the St Andrews Agreement, when the British government committed to an Act, but had become a totemic issue by the 2016–2017 elections. This reflects a gradual shift over the last 15 years from peace and security issues to those pertaining to identity and culture, such as flags and parading. At the same time, there has been an increase in salience of so-called ‘moral’ issues with a distinct liberal/conservative cleavage emerging. Issues such as marriage equality and abortion have moved centre stage in recent election campaigns, with the Assembly voting on the former issue on five occasions from 2012 to 2015 (see Table 1).
Marriage Equality legislative proposals, Northern Ireland Assembly 2007–2017.
MLA: Member of the Legislative Assembly; SDLP: Social Democratic Labour Party.
The political salience of other issues and identities has increased since 1998. At the party system level, a growing number of parties have opted for the ‘Other’ designation. As O’Leary (2018: 235) notes, ‘Northern Ireland’s peace process and consociational experiment have in fact enabled novel opposition parties to organize in conditions of generalized security’. Since 1998, the Alliance Party, the Women’s Coalition, the Green Party, NI21 and People Before Profit have all designated as Other. 3 Recent elections have witnessed a moderate growth in support for civic parties as well as their diversification. The 2016 election saw civic parties win 12 out of 108 seats, an increase of 3 seats, and the breakthrough of People Before Profit, a socialist party eschewing ethnic labels. 4 Eleven seats were won by Other parties in the 2017 election, but in the context of a reduction in the number of seats to 90, this represented an increase in their share of first preference votes from 11.9% to 13.2% (ark.ac.uk; see Table 2 and Figure 1). Other parties also performed above expectations in the 2019 local elections, with Alliance seeing a 65% rise in its representation to win 53 seats. PBP and the Greens also performed strongly (BBC, 2019b). Alliance polled second-place in the 2019 European elections, surpassing Sinn Féin and winning its first seat in the European Parliament.
Party vote share (%) Northern Ireland Assembly elections 1998–2017.
Source: www.ark.ac.uk/elections/
DUP: Democratic Unionist Party; UUP: Ulster Unionist Party; SDLP: Social Democratic Labour Party; NIWC: Northern Ireland Women’s Coalition.
The number of seats in the 2017 Northern Ireland Assembly election was reduced from 108 to 90.
Party vote share Northern Ireland Assembly elections 1998–2017 (%).
As Other parties gain in popularity, this pushes ethnic parties to respond to a wider range of policy issues, prompting them ‘to update their platforms’ (O’Leary, 2018: 235). For example, Alliance, Greens, the SDLP and Sinn Féin have been at the forefront of the campaign for marriage equality, increasingly reflected in their election programmes from 2007–2017. Meanwhile, the DUP’s 2016 manifesto declares that the party ‘stood by its commitment to family values and marriage and will continue to do so’ (DUP Assembly Manifesto, 2016). Nevertheless, the other parties’ positions may ultimately push the DUP to soften its stance. Arlene Foster’s decision to attend a PinkNews event in June 2018, the first DUP leader to ever do so, may be an early sign. 5 ‘[W]e must take our message to places that perhaps may not be traditional to our cause’, Foster said (PinkNews, 2018). Coming on the heels of her attendance at a GAA match in the Irish Republic (Northern Slant, 2018), these gestures indicate a symbolic willingness to reach beyond the core. The DUP also elected its first openly gay representative in the 2019 local elections.
In addition to widening party space for Others, demographic trends suggest a changing electorate, one not exclusively beholden to the nationalist/unionist cleavage. The proportion of respondents to an annual survey identifying as neither unionist nor nationalist increased from 33% in 1998 to 50% in 2018 (ark.ac.uk, 2019). Referring to this category as ‘neither/nor’, Hayward and McManus (2018: 5; 11–12) found that in the 2017 Assembly election, 16% supported Alliance, 7% supported SDLP but 38% rejected all parties, concluding that, rather than forming the basis of an alternative politics, this group rebuffs the political options on offer. Notably, the proportion identifying as ‘neither/nor’ fell significantly in 2019 from 50% to 39%, a potential symptom of the polarisation of politics post-2017 discussed below (ark.ac.uk, 2020).
According to the 2011 census, the proportion of people living in Northern Ireland who were born outside the United Kingdom and Ireland rose from 1.6% in 2001 to 4.5% in 2011 – an increase of 199% (migrationobservatory.ox.ac.uk). Belfast had the highest proportion of these residents, at 18,414 (22.7% of the total), making up 6.6% of the city’s dwellers. Of these residents, Poland was the most common country of birth (24.2% of the total) and migrants from Eastern European countries that joined the EU after 2001 made up 44% (migrationobservatory.ox.ac.uk). Net migration reached its highest level in 10 years in 2018, with 23,600 people coming and 19,400 people leaving Northern Ireland (BBC, 2019a).
Our interviews with representatives of these Other communities, including activists and politicians, revealed a sense of marginalisation from, and frustration with, Northern Ireland’s political system, which they attribute in large part to the power-sharing structures. As one LGBTQ advocacy figure articulated:
. . . We just have hurdles which exist here that do not exist in other jurisdictions. And that’s largely down to the way that the Assembly is constructed . . . That system is actually not created to govern. It’s certainly not created to effect change. It’s created to slow everything down. (Interview with LGBTQ Advocacy Representative 1, 2018)
This creates an untenable situation. The ability of the dominant ethnic parties to engage with Others matters for the democratic inclusion of those who do not identify with the ethno-divide. It also matters from the perspective of the parties. From an electoral perspective, the ethnic parties face a shrinking voter base and mobilise amid an increasingly diverse and de-aligned population. Opinion polls show significant movement on issues such as marriage equality, with a 2018 Sky News poll showing 76% of respondents in favour of its legalisation (Sky News, 2018), and abortion with 71% of respondents to a Life and Times Survey agreeing that it is a women’s right to choose (ark.ac.uk, 2018). 6 From a constitutional perspective, with Northern Ireland’s future looking less certain in the Brexit context, moderate non-aligned voters could be pivotal to securing support for either a united Ireland or to maintaining union with Britain (Nolan, 2018).
These developments suggest a changing political landscape. As noted, the 2007–2017 period saw the growth in both the number and vote share of Other parties. There was also a rise of non-ethnic issues to the top of the political agenda, such as abortion, marriage equality and the Irish language. Welfare reform, legislated by the British Government in 2012, was also subject to sectarian contestation, with unionist parties supporting implementation in Northern Ireland while nationalist parties remained opposed, leading to a prolonged dispute between the power-sharing partners. Civil society activists for abortion rights, same-sex marriage and the Irish language interviewed also attested to an increasing mobilisation on these issues, particularly among younger generations. Representatives of each spoke of a solidarity between these three agendas as well as welfare reform and a cross-pollination between their movements (Interviews with LGBTQ Advocacy Representative 1, 2018; Abortion Rights Activist 1, 2018; Irish Language Activist 2, 2018). Changing demographics, as outlined above, were also important, due to new immigration to the region and the rise of the ‘neither/nor’ identity. This offers some compelling evidence for the ability of the power-sharing system to cultivate a more complex and fluid political landscape. How have the dominant ethnic parties responded?
To whom do ethnic parties reach out?
In our interviews, we asked ethnic parties how they reach out to voters beyond their core. Respondents identified three broad categories (see Table 3). The first category includes those from the other community – alternate community others. These voters may come into contact with ethnic parties of the ‘other side’ because they lack a comfortable political home within their ‘own’ bloc, perhaps because they hold opposing ideological positions from ethnic parties in that bloc or because they reside in an area where the other community forms the majority. They may also seek constituency representation from these parties for practical or strategic purposes, even if they continue to align politically with parties of their ‘own’ ethnic bloc. Second, party representatives identified external others or those from ethnic minority communities who have migrated to Northern Ireland. Third, they refer to ‘indigenous’ or internal others: those from Northern Ireland who have chosen to explicitly reject nationalist or unionist labels. Different representatives used different terms to describe these groups, but those characterisations fit broadly under these three headings.
Understanding of ‘others’ by party representatives.
The three categories were attributed varying degrees of legitimacy by those we interviewed. Alternate community others were treated sympathetically. Representatives from the DUP referred to their openness to Catholic nationalist voters who find themselves alienated from Sinn Féin or the SDLP due to their relatively liberal positions on marriage equality or abortion, while SDLP representatives welcomed unionist voters who may cross communal lines to support them for the same reason. Those from the ‘other community’ with whom parties engaged on practical, local issues through constituency casework tended to be discussed in a neutral but respectful manner (Interview with DUP MP, 2018). Yet, on the other side, alternate community others from within the representative’s own community were, in some cases, treated with suspicion. A DUP representative dismissed the idea that unionists formed a significant part of the Irish-speaking community and then claimed this marginal group – represented by the Turas project – was ‘being used’ by nationalist parties: ‘They’re middle class Protestants who want to come down and learn Irish . . . There’s no Irish language class, no Protestants attending Irish language classes, in South Down’ (Interview with DUP MLA 2, 2018). External others were also presented in sympathetic terms, depicted by some representatives as politically passive, willingly non-aligned, understandably apolitical and minimally engaged (Interview with SDLP MLA 1, 2018). Internal others were accorded the least legitimacy. Where acknowledged, they tended to be characterised as a relatively small proportion of the population, seen mainly as urban and middle class (Interview with DUP MLA, 2018). The ethnic parties generally shared the belief that the system already has some institutional capacity to accommodate Others, pointing to the presence of civic parties in the Assembly, rather than reflecting on their own parties’ capacity to reach out to these voters (Interview with SDLP MLA 2, 2018). Some did, however, acknowledge that these individuals were not always well represented in the current system, with the community designation rule and petition of concern seen as problematic for Others (Interview with SDLP MLA 2, 2018).
How do ethnic parties reach out?
Our research found that, during the 2007–2017 period, ethnic parties had taken some initial steps to reach out beyond their respective cores. This reaching out, while rather preliminary, occurs at three levels: policy, constituency and electoral (see Table 4).
Strategies of engagement with others reported by parties.
With reference to policy, some representatives reported increasing contact and alliance with alternate community others on key policy issues, particularly marriage equality and abortion. Some contact with individual voters on these matters was recounted through written correspondence and street campaigning (Interview with DUP MP, 2018). Yet, these relations primarily occurred on a formal level with advocacy and campaign groups. One DUP MLA described having good relations with Catholic-aligned advocacy groups on abortion. In the run-up to the 2018 referendum on the legalisation of abortion in the Irish Republic, he reported attending a fundraising dinner for campaign group Both Lives Matter, which brought together ‘folk from devout Catholic and strong Evangelical Protestant backgrounds’ (Interview with DUP MLA 2, 2018).
It was at the level of constituency work that the parties depicted the greatest contact with Others, notably alternate community members. Every ethnic party representative reported conducting casework for constituents from the other community and several emphasised a rise in this contact in recent years. One DUP MLA estimated that 20%–25% of his caseload came from members of the nationalist community, a phenomenon he saw as increasing (Interview with DUP MLA 1, 2018). This work generally concerned ‘bread-and-butter’ issues from planning to welfare to health care. The reasons speculated for constituents approaching a party from the alternate bloc included their perceptions of the efficacy of that particular Member and their access to government. In the words of one SDLP MLA, constituents have become increasingly ‘retail’ and go to the person they believe will deal with their problem most efficiently (Interview with SDLP MLA 2, 2018). She also suggested a potential gendered aspect; some constituents seemed more likely to seek her help than contacting a more senior male constituency colleague for whom they would vote (Interview with SDLP MLA 2, 2018). That representative reported significant contact from members of migrant communities – ‘external others’ – who can face additional challenges in accessing public services (Interview with SDLP MLA 2, 2018). Most representatives referred to the increase in constituency casework from Others as a demand to which they were responding as opposed to one which they sought out. In terms of broader local outreach, a number cited constituency visits to schools of the other denomination and to ethnic minority organisations and projects.
On the question of electoral mobilisation, however, most representatives did not see this cross-ethnic contact translating into a significant vote, at least not in the short term (Interview with Sinn Féin MLA 2, 2013). Every ethnic party interviewed reported receiving some electoral support from Others (based on anecdotal experience), referring primarily to alternate community others. One Sinn Féin MLA noted, ‘I’ve had people in here who are from the Protestant community who are voting Sinn Féin because of our strong position on equality’ (Interview with Sinn Féin MLA, 2015). The potential for electoral support from members of the other community potentially ‘displaced’ on the basis of policies like marriage equality and abortion was noted by a number of representatives. A DUP MP reported receiving supportive communications from conservative Catholic voters about his opposition to equal marriage and abortion:
I would get more in correspondence where conservative Catholics would write and say, ‘Look, politically I don’t have a home because the SDLP and Sinn Féin no longer represent my views’, and they just say, ‘Look, I want to let you know I’ll be voting for you’, or, you know, ‘I share your views’. (Interview with DUP MP, 2018)
Yet, this vote was generally depicted as marginal: estimated as ‘in the hundreds, not thousands’ (Interview with DUP MP, 2018) or ‘a couple of hundred’ (Interview with DUP MLA 2, 2018). As one DUP MLA put it: ‘I have a very good relationship with the conservative Roman Catholics but I’m not silly enough to think that that translates into a large vote’ (Interview with DUP MLA 2, 2018).
Members from the moderate nationalist and unionist parties reported greater success in appealing to voters of the other community. Interestingly, however, one liberal UUP MLA classed these voters as unionist, by definition, in voting for him:
I get quite a few votes from people who live in nationalist areas or who designate themselves as Catholics . . . I think a lot of that is because they’re slightly left of centre. And they’re quite happy with the state that we have at this moment in time. Which makes them Unionists in this [sense].
Likewise, he was doubtful that nationalist-identifiers would vote for him based on his liberal stance on abortion and marriage equality:
I support Sinn Féin’s policy on same-sex marriage . . . But I couldn’t vote for them because of it – because of their policy on a united Ireland . . . So, I don’t think I get many first preference votes from those who would want a unified Ireland. (Interview with UUP MLA 2, 2018)
A more nuanced picture emerged on lower-order vote preferences. UUP and SDLP MLAs reported significant vote transfers from civic parties and moderate parties of the other bloc. One SDLP member said she received a positive response from UUP voters to a statement by then-UUP Leader Mike Nesbitt during the 2017 election campaign encouraging them to cast their second preferences for the SDLP (Interview with SDLP MLA 2, 2018). These representatives saw the potential for a moderate electoral alliance as being on an upwards trajectory up to 2016, yet felt it had been dealt a blow by the polarising events since 2017. This falls in line with evidence from research on inter-bloc transfers in the 2011 elections, which shows that while Sinn Féin, the DUP and UUP received negligible cross-ethnic transfers (2% each), 16% of the SDLP’s inter-party transfers came from across the divide (Barry and Love, 2011: 8–9). Jarrett (2018: 102) marks the SDLP as unique among the four main ethnic parties in its willingness to use STV to attract transfers from voters of the other bloc. While not requiring ethnic parties to reach out, STV does offer them some security in doing so: they can appeal to their ethnic base but still pick up lower-order preferences from Other voters.
When it came to voters who had migrated to Northern Ireland, it appeared that there is little systematic or sustained effort to engage these ‘external others’ electorally. These communities were depicted as highly heterogenous, their support often contingent on what the majority party happened to be in their area. A number of representatives alluded to the difficulty of capturing the ethnic minority vote, not least due to the transient nature of these residents and their perceived lack of political engagement. The SDLP reported fielding candidates of Portuguese and Polish nationality in the 2011 local elections in Armagh, but to limited electoral effect (Interview with SDLP MLA 1, 2018).
Ethnic party representatives generally had least to say about reaching out to those who rejected both nationalist and unionist labels – so-called ‘internal others’. In electoral terms, they associated this group primarily with civic parties or with non-voting. One SDLP MLA admitted that, ‘[Some parties], the Greens in particular, are pitching to that audience . . . but nobody’s doing it very strategically’ (Interview with SDLP MLA 2, 2018). A UUP MLA echoed the opportunity these voters represented electorally, believing that many shared his party’s goals for social progress under the constitutional status quo. Yet, he expressed the challenge of engaging them under the unionist banner:
If we could explain what we want . . . without mentioning the word ‘unionist’, people would actually see ‘Oh, that’s what I want!’ (Interview with UUP MLA 2, 2018)
To what extent is this engagement with Others an active party strategy? Sinn Féin is the only party that appears to have had a central initiative for cross-community engagement, having launched a Charter for Unionist Engagement in 2007 (sinnfein.ie). The SDLP has, however, criticised the party more recently for abandoning this agenda (BBC, 2016). SDLP MLAs depicted their party’s outreach as based less on ‘over-arching’ strategy than long-standing cross-community work ‘below the radar’, mostly at local level (Interviews with: SDLP MLA 1, 2018; SDLP MLA 2, 2018), an approach echoed by the DUP, though one representative reported distributing business cards to households in nationalist majority areas during constituency rounds (Interview with DUP MP, 2018). On the electoral side, while welcoming support from Others, few representatives reported actively campaigning for those votes. In a 2013 interview, one DUP MLA highlighted the risk that engaging with voters from nationalist areas may be perceived as coming at the expense of unionist areas (Interview with DUP MLA, 2013). Similarly, one SDLP MLA said visits to organisations like the Orange Order and non-Catholic faith groups were not usually publicised due to their sensitive nature (Interview with SDLP MLA 1, 2018).
How is such engagement experienced by Others? Interviews with the LGBTQ, abortion rights, Irish language and ethnic minority groups affirmed some level of contact with all parties, though least publicly with the DUP. While representatives generally welcomed this outreach, it did not come without risk. They were mindful of the perception of their being aligned to one or other ethnic bloc and some strategised to avoid this taint (Interviews with: LGBTQ Advocacy Representative 1, 2018; Abortion Rights Activist 1, 2018; Irish Language Activist 2, 2018; Ethnic Minority Rights Representative, 2018; Irish Language Activist 3, 2018). The ethnic minority sector appeared particularly vulnerable to the risk of ethnic contestation and co-optation. One representative of a religious minority community centre in South Belfast reported most engagement from Sinn Féin and Alliance, but increased engagement from all parties in recent years, including the DUP. In contrast to Sinn Féin, however, the DUP was less likely to publicise its visits. She attributed the motivation for party engagement to both strategic concerns – in recognition that the community makes up a growing proportion of the South Belfast voter base – but also genuine ‘compassion’ for the issues facing the community (Interview with Religious Community Organisation Representative, 2018).
Policy and constituency thus emerged as the primary levels of engagement with Others by ethnic parties, with more nascent engagement at the electoral level.
Why do ethnic parties reach out?
Ethnic parties have principled and strategic reasons for reaching beyond their core. The increasingly multi-dimensional political landscape offers one incentive to reach out; changing demographics another. That is, such parties have a finite base from which to recruit support and, in conditions of intra-ethnic competition, will reach saturation point. For unionists facing a shrinking share of the population – as one unionist party representative told us ‘in the next 15 to 25 years we will all be minorities’ (Interview with DUP MP, 2018) – there is an impetus to reach out. The same may be said for nationalist parties, which have seen a share of their traditional vote go to Other parties, such as Sinn Féin’s loss to People Before Profit in West Belfast.
A primary reason advanced by most representatives was principle: they were elected to represent all citizens in their constituency, therefore, engaging with those outside their community was their democratic duty. Representatives said they stood for certain ideological positions – primarily socially liberal/conservative and economically left/right – and sought to represent anyone in society who shared those views. One DUP MLA said of his opposition to abortion and marriage equality:
The nationalists would have a very different view on these sorts of things. Therefore, I feel it is important that I represent [my constituents’] views and the views of a significant [portion of] conservative Roman Catholics who also would hold similar views to me on this. (Interview with DUP MLA 2, 2018)
Sinn Féin MLAs claimed that on issues like welfare and education reform, their party was working for voters in working-class nationalist and unionist areas (Interview with Sinn Féin MLA 1, 2013).
Yet, in describing their motivations, representatives also betrayed strategic self-interest. A number acknowledged the importance of reaching out to Others for their parties’ electoral survival. For one SDLP MLA this represented a significant opportunity:
I think that’s the gap in the market here. I don’t mean it as cynically as that, but I just think there are many people who are increasingly, (a) sick of the identity politics and, (b) . . . there are a lot of people who strongly support the Union or strongly support Irish unity but just maybe [feel] that politics hasn’t served us very well . . . (Interview with SDLP MLA 2, 2018)
Another representative voiced this strategic motivation with reference to ethnic minorities, alleging that some other parties engage in clientelistic practices to ‘court’ these voters (Interview with SDLP MLA 1, 2018).
An ethnic party’s status as major or minor within their bloc may also influence their motivation to reach out. On one hand, it would appear that those minor (formerly major) parties, SDLP and UUP, would be more inclined to reach beyond their base given their relatively moderate positions and their diminishing electoral base. On the other hand, the major parties, Sinn Féin and DUP, have moved to dominant positions partly by moderating their stances and appealing to middle-ground voters. Reaching out to Others can arguably serve as a useful signal of their moderate credentials to voters within their own bloc. One political commentator attributed comments by former DUP leader Peter Robinson in 2011 that the party would begin to appeal to Catholic voters, to such motivations (Devenport, 2011).
Beyond electoral calculations, some representatives referred to reaching across as imperative to securing their party’s constitutional ambitions. One Sinn Fein MLA noted that, while it was not the party’s motivation for reaching out to unionists, a by-product may be that voters become curious about them and begin to see the benefits of Irish unification (Interview with Sinn Féin MLA 1, 2013). One UUP MLA, meanwhile, recognised that securing traditional nationalists’ support for the Union was essential to its preservation (Interview with UUP MLA 1, 2018).
Conclusion
We acknowledge that while power-sharing may be necessary under conditions of deep division and communal conflict, Others nonetheless occupy an unfavourable position in such systems. In this article, we have taken up the challenge put by Stojanović (2018: 361) who argues that ‘the consociational approach is ipso facto uninterested in the political status of those who do not belong to any of the significant segments’ by exploring the relationship between the included (dominant ethnic parties) and excluded (Others) in consociational systems. While much of the nascent literature on Others and consociationalism focuses – rightly – on how Others view power-sharing, we have considered the role of ethnic parties not because we want to privilege them in analysis or solutions, but because as veto-wielders they have an outsized role in determining whether and how Others can be included. As gatekeepers of inclusion, it is integral that we understand to whom ethnic parties reach out, how and why they do so. We have found that ethnic parties in Northern Ireland’s power-sharing system made incremental moves to engage with Others – including those of the other ethno-national background, immigrant communities and those who identify as neither nationalist nor unionist – and did so in the arenas of policy, constituency and elections, during the 2007–2017 period. Their motivations for ‘reaching out’, we argue, can be explained by both principled and strategic concerns. Events following the 2007–2017 period, however, might suggest that consociational stability is vulnerable to shocks, which tend to close the political space available to Others.
Our analysis of the 2007–2017 period lends some support to the claim that liberal consociation can facilitate the opening of political space beyond the ethno-divide and that ethnic parties operating in such structures can and do reach out beyond their core. Indeed, without overstating the extent to which they did so, the ethnic parties were beginning to take advantage of these opportunities in the examined 10-year period. Moreover, the receding of peace and security issues during this time saw the rise of alternative issues, including abortion, welfare reform and marriage equality. Other developments include a moderate increase in support for civic parties and the formation of an official opposition at Stormont which opened up the space for cooperation between moderate nationalist, unionist and other parties. The ‘normalization’ of politics over this period formed a recurring theme in interviews. A member of the then-UUP leadership reflected on the success of this short-lived opposition arrangement following the 2016 election, his party’s positive relations with SDLP leader Colum Eastwood and the potential for the two parties to form a ‘shadow Executive’ that could model an alternative, moderate coalition to voters (Interview with UUP MLA 1, 2018). It appears that the conditions which make ethnic parties more likely to cooperate across the divide may also make them more likely to reach out to Others and that these phenomena may be mutually supportive.
If our analysis ended with the 2007–2017 period, we might have good cause for supporting the second perspective on power-sharing noted earlier. The note of optimism regarding normalisation, however, was followed up in many interviews by lamenting the contraction of political space, especially for Others, and increasing polarisation in response to the series of shocks experienced by the region, not least of all the Brexit referendum and collapse of the devolved institutions. That is, the 3-year power-sharing suspension knocked much of the progress on reaching out off course. This brings us back to the ‘all politics is ethnic politics’ thesis.
There are two ways that this thesis plays out in our case. First, the logic of ethnic representation retained a firm hold. Many of those we interviewed tended to employ an ethnic lens in understanding Others. When asked about voters beyond their core constituency, the first response was to refer to those of the ‘other side’, meaning either nationalist or unionist (so-called alternate community others), or those who had migrated to Northern Ireland (so-called external others). There was also a sense from moderate ethnic and civic parties that, in spite of moderating public opinion, insecurity vis-à-vis ‘the other side’ remains a key driver at the ballot box, putting a brake on that shift translating into electoral change. As one UUP MLA put it, voters are ‘moderate 364 days of the year’ – the exception being election day (Interview with UUP MLA 1, 2018).
The second constraint on reaching out relates to the tendency to fall back on old habits in moments of crisis. Events subsequent to 2017 suggest that liberal power-sharing arrangements display a vulnerability to ‘shocks’ and other crisis moments, which served to re-entrench the ethnic parties in their respective positions. For example, the preference-swapping deal so celebrated by the SDLP and UUP in 2016 was not recreated in the 2017 election, with many within the UUP resisting their leader’s call to exchange preferences with the SDLP. Similarly, despite positioning the party as the leading supporter of marriage equality, including at one point making it a pre-condition for the restoration of power-sharing, Sinn Féin was quick to drop this demand when it looked like a deal on an Irish Language Act was imminent in early 2018.
This has implications for the precarious position of Others in polarising conditions. Given the veto power held by ethnic parties in power-sharing systems, it is in the strategic interest of Other communities to have the dominant parties advocating for their rights. Yet, such parties have often proved themselves to be unreliable partners – reaching out in stable conditions but retreating to their ethnic trenches in challenging times. During such times, not only do political issues unrelated to ethnic divisions struggle to reach the top of the legislative agenda, they are sometimes seen as items over which to barter with the ‘other side’ as was the case with marriage equality.
This suggests avenues for future research. As a first step, it would be important to consider whether other liberal consociations offer indications of reaching out as well. North Macedonia would be an apposite case for comparison, especially focusing on the ethnic Macedonian party, the Social Democratic Union of Macedonia (SDSM), which has begun to expressly appeal beyond its ethnic core. A second strategy would be to compare liberal consociations to their corporate consociational counterparts, such as Bosnia and Lebanon, in order to test the contention that the former bears stronger capacity to accommodate Others. Finally, while we have focused on how and why ethnic parties reach beyond their core, a third avenue would be to consider the question of when this happens. This would enable further empirical testing of our claim that the political space for Others cultivated under stable liberal consociational rules nonetheless contracts quickly in the face of shocks. Following any of these tracts would help to reiterate a central claim of this article: power-sharing may be necessary in order to support democracy amid deep division, but it need not ‘sacrifice’ the political rights of Others in the process.
Footnotes
Acknowledgements
We would like to sincerely thank the participants in this research for their valuable contributions to this paper. We would also like to thank Timofey Agarin, Laura Sudulich and Kayla Smith for helpful comments on earlier drafts of this article, as well as our colleagues at the IPSA World Congress of Political Science, Brisbane 2018; the APSA Annual Meeting, Boston 2018; the PSA Annual International Conference, Nottingham 2019; and the Villanova University Political Science research in progress seminar, January 2019.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research was carried out as part of the Exclusion amid Inclusion: Power-Sharing and Non-Dominant Minorities’ project at Queen’s University Belfast, funded by the Economic and Social Research Council.
