Consociationalism is often seen as having close affinities with corporatism; some consider it to be a form of corporatism while others claim that economic corporatism was designed to regulate class conflict, while consociationalism developed on the basis of reconciling societal fragmentation along ethnic and religious lines.
The goals of consociationalism are governmental stability, the survival of the power-sharing arrangements, the survival of democracy, and the avoidance of violence. When consociationalism is organised along religious confessional lines, it is known as confessionalism, as is the case in Lebanon.
Political scientists define a consociational state as a state which has major internal divisions along ethnic, religious, or linguistic lines, with none of the divisions large enough to form a majority group, yet nonetheless manages to remain stable, due to consultation among the elites of each of its major social groups. Consociational states are often contrasted with states with majoritarian electoral systems.
Consociation was a term and concept discussed in the 17th century New England Confederation with reference to the interassociation and cooperation of the participant independently self-governing Congregational churches of the various colonial townships of the Massachusetts Bay Colony (which were embedded in the civil legislature and magistracy). It was debated at length in the Boston Synod of 1662, at the time when the Episcopalian Act of Uniformity 1662 was being introduced in England.
Consociationalism was discussed in academic terms by the political scientist Arend Lijphart. However, Lijphart has stated that he had "merely discovered what political practitioners had repeatedly – and independently of both academic experts and one another – invented years earlier". John McGarry and Brendan O'Leary trace consociationalism back to 1917, when it was first employed in the Netherlands.
Indeed, Lijphart draws heavily on the experience of the Netherlands in developing his argument in favour of the consociational approach to ethnic conflict regulation. The Netherlands, as a consociational state, was between 1857 and 1967 divided into four non-territorial pillars: Calvinist, Catholic, socialist, and general, although until 1917 there was a plurality ("first past the post") electoral system rather than a consociational one. In their heyday, each comprised tightly-organised groups, schools, universities, hospitals and newspapers, all divided along a pillarised social structure. The theory, according to Lijphart, focuses on the role of social elites, their agreement and co-operation, as the key to a stable democracy.
Lijphart identifies four key characteristics of consociational democracies:
Consociational policies often have these characteristics:
- Coalition cabinets, where executive power is shared between parties, not concentrated in one. Many of these cabinets are oversized, meaning they include parties not necessary for a parliamentary majority;
- Balance of power between executive and legislative;
- Decentralized and federal government, where (regional) minorities have considerable independence;
- Incongruent bicameralism, where it is very difficult for one party to gain a majority in both houses. Normally one chamber represents regional interests and the other national interests;
- Proportional representation, to allow (small) minorities to gain representation too;
- Organized and corporatist interest groups, which represent minorities;
- A rigid constitution, which prevents government from changing the constitution without consent of minorities;
- Judicial review, which allows minorities to go to the courts to seek redress against laws that they see as unjust;
- Elements of direct democracy, which allow minorities to enact or prevent legislation;
- Proportional employment in the public sector;
- A neutral head of state, either a monarch with only ceremonial duties, or an indirectly elected president, who gives up his or her party affiliation after being elected;
- Referendums are only used to allow minorities to block legislation: this means that they must be a citizen's initiative and that there is no compulsory voting.
- Equality between ministers in cabinet, the prime minister is only primus inter pares;
- An independent central bank, where experts and not politicians set out monetary policies.
Lijphart also identifies a number of "favourable conditions" under which consociationalism is likely to be successful. He has changed the specification of these conditions somewhat over time. Michael Kerr summarises Lijphart's most prominent favourable factors as:
- Segmental isolation of ethnic communities
- A multiple balance of power
- The presence of external threats common to all communities
- Overarching loyalties to the state
- A tradition of elite accommodation
- Socioeconomic equality
- A small population size, reducing the policy load
- A moderate multi-party system with segmental parties
Lijphart stresses that these conditions are neither indispensable nor sufficient to account for the success of consociationalism. This has led Rinus van Schendelen to conclude that "the conditions may be present and absent, necessary and unnecessary, in short conditions or no conditions at all".
John McGarry and Brendan O'Leary argue that three conditions are key to the establishment of democratic consociational power-sharing: elites have to be motivated to engage in conflict regulation; elites must lead deferential segments; and there must be a multiple balance of power, but more importantly the subcultures must be stable. Michael Kerr, in his study of the role of external actors in power-sharing arrangements in Northern Ireland and Lebanon, adds to McGarry and O'Leary's list the condition that "the existence of positive external regulating pressures, from state to non-state actors, which provide the internal elites with sufficient incentives and motives for their acceptance of, and support for, consociation".
In a consociational state, all groups, including minorities, are represented on the political and economic stage. Supporters of the consociationalism argue that it is a more realistic option in deeply divided societies than integrationist approaches to conflict management. It has been credited with supporting successful and non-violent transitions to democracy in countries such as South Africa.
Brian Barry has questioned the nature of the divisions that exist in the countries that Lijphart considers to be "classic cases" of consociational democracies. For example, he makes the case that in the Swiss example, "political parties cross-cut cleavages in the society and provide a picture of remarkable consensus rather than highly structured conflict of goals". In the case of the Netherlands, he argues that "the whole cause of the disagreement was the feeling of some Dutchman ... that it mattered what all the inhabitants of the country believed. Demands for policies aimed at producing religious or secular uniformity presuppose a concern ... for the state of grace of one's fellow citizens". He contrasts this to the case of a society marked by conflict, in this case Northern Ireland, where he argues that "the inhabitants ... have never shown much worry about the prospects of the adherents of the other religion going to hell". Barry concludes that in the Dutch case, consociationalism is tautological and argues that "the relevance of the 'consociational' model for other divided societies is much more doubtful than is commonly supposed".
Rinus van Schendelen has argued that Lijphart uses evidence selectively. Pillarisation was "seriously weakening", even in the 1950s, cross-denominational co-operation was increasing, and formerly coherent political sub-cultures were dissolving. He argued that elites in the Netherlands were not motivated by preferences derived from the general interest, but rather by self-interest. They formed coalitions not to forge consociational negotiation between segments but to improve their parties' respective power. He argued that the Netherlands was "stable" in that it had few protests or riots, but that it was so before consociationalism, and that it was not stable from the standpoint of government turnover. He questioned the extent to which the Netherlands, or indeed any country labelled a consociational system, could be called a democracy, and whether calling a consociational country a democracy isn't somehow ruled out by definition. He believed that Lijphart suffered severe problems of rigor when identifying whether particular divisions were cleavages, whether particular cleavages were segmental, and whether particular cleavages were cross-cutting.
Ian Lustick has argued that academics lack an alternative "control" approach for explaining stability in deeply divided societies and that this has resulted in the empirical overextension of consociational models. Lustick argues that Lijphart has "an impressionistic methodological posture, flexible rules for coding data, and an indefatigable, rhetorically seductive commitment to promoting consociationalism as a widely applicable principle of political engineering", that results in him applying consociational theory to case studies that it does not fit. Furthermore, Lustick states that "Lijphart's definition of 'accommodation' ... includes the elaborately specified claim that issues dividing polarized blocs are settled by leaders convinced of the need for settlement".
Critics point out that consociationalism is dangerous in a system of differing antagonistic ideologies, generally conservatism and communism. They state that specific conditions must exist for three or more groups to develop a multi- system with strong leaders. This philosophy is dominated by elites, with those masses that are sidelined with the elites having less to lose if war breaks out. Consociationalism cannot be imperially applied. For example, it does not effectively apply to Austria. Critics also point to the failure of this line of reasoning in Lebanon, a country that reverted to civil war. It only truly applies in Switzerland, Belgium and the Netherlands, and not in more deeply divided societies. If one of three groups gets half plus one of the vote, then the other groups are in perpetual opposition, which is largely incompatible with consociationalism.
Consociationalism focuses on diverging identities such as ethnicity instead of integrating identities such as class, institutionalizing and entrenching the former. Furthermore, it relies on rival co-operation, which is inherently unstable. It focuses on intrastate relations and neglects relations with other states. Donald L. Horowitz argues that consociationalism can lead to the reification of ethnic divisions, since "grand coalitions are unlikely, because of the dynamics of intraethnic competition. The very act of forming a multiethnic coalition generates intraethnic competition – flanking – if it does not already exist". Consistent with Horowitz' claims, Dawn Brancati finds that federalism/territorial autonomy, an element of consociationalism, strengthens ethnic divisions if it is designed in a way that strengthens regional parties, which in turn encourage ethnic conflict.
Consociationalism assumes that each group is cohesive and has strong leadership. Although the minority can block decisions, this requires 100 per cent agreement. Rights are given to communities rather than individuals, leading to over-representation of some individuals in society and under-representation of others. Grand coalitions are unlikely to happen due to the dynamics of ethnic competition. Each group seeks more power for itself. Consociationalists are criticized for focusing too much on the set up of institutions and not enough on transitional issues which go beyond such institutions. Finally, it is claimed that consociational institutions promote sectarianism and entrench existing identities.
The political systems of a number of countries operate or used to operate on a consociational basis, including Belgium, Cyprus (effective 1960–1963), Lebanon, the Netherlands (1917–1967), Switzerland, Ethiopia, and South Africa. Some academics have also argued that the European Union resembles a consociational democracy.
Additionally, a number of peace agreements are consociational, including:
- the Dayton Agreement that ended the 1992–1995 war in Bosnia and Herzegovina, which is described as a "classic example of consociational settlement" by Sumantra Bose and "an ideal-typical consociational democracy" by Roberto Belloni.
- the Belfast Agreement of 1998 in Northern Ireland (and its subsequent reinforcement with 2006's St Andrews Agreement), which Brendan O'Leary describes as "power-sharing plus".
- the Ohrid Agreement of 2001 setting the constitutional framework for power-sharing in the Republic of Macedonia.
In addition to the two-state solution, some have argued for a one-state solution under a consociational democracy in the state of Israel to solve the Arab–Israeli conflict, but this solution is not very popular, nor has it been discussed seriously at peace negotiations.
During the 1980s the South African government attempted to reform apartheid into a consociational democracy. The South African Constitution of 1983 applied Lijpart's powersharing ideas by establishing a Tricameral Parliament. During the 1990s negotiations to end apartheid the National Party (NP) and Inkatha Freedom Party (IFP) proposed a settlement based upon consociationalism. The African National Congress (ANC) opposed consociationalism and proposed instead a settlement based upon majoritarian democracy. The NP abandoned consociationalism when the U.S. Department of State came out in favor of the majoritarian democracy model in 1992. in Iran, the present government is based on consociationalism.