Consociational Democracy in Multiethnic Societies

CONTENTS

Introduction

1 The Concept of Consociational Democracy
1.1 Theoretical Background
1.2 Executive Power-Sharing
1.3 Segmental Autonomy
1.4 Proportional Representation
1.5 Mutual Veto Rights
1.6 Establishment of Consociatonal Democracies

2 Collected Critiques
2.1 Definition Deficiencies
2.2 Flawed Analysis? The Role of the Case Studies
2.3 Insufficiently Democratic?
2.4 Explanatory Weakness and Non-Applicability
2.5 Conclusion

3 The Favourable Factors for Consociational Democracy
3.1 Structure-oriented Factors
3.2 Actor-oriented Factors
3.3 The Status of the Factors
3.4 Conclusion

4 Making It Work
4.1 Executive: Parliamentary versus Presidential Systems
4.2 Over-representation versus Proportional Representation
4.3 The Modality of Veto Rights
4.4 The Form of Autonomy
4.5 Arbitration
4.6 Working towards Favourable Factors

5 Conclusion

References

Introduction

The central question of this paper is almost as old as the mere concept of consociationalism: to what extent can consociational democracy serve as the appropriate democratic form to divided and multiethnic societies? Soon after Arend Lijphart and Gerhard Lehmbruch had depicted consociational democracy as a viable alternative to majoritarian forms of democracy (Lijphart 1968; Lehmbruch 1967), a fierce debate about its wider applicability took root (See Nordlinger 1972, Daalder 1974, Barry 1975, Lijphart 1977, and Lustick 1979). Lijphart and Lehmbruch had presented the Netherlands, Belgium, Austria and Switzerland as cases of consociational democracies - but would the concept work in countries that are divided more deeply?

For a good reason, the issue regained relevance and interest of scholars during the 1990s. Intra-state wars took their tolls on an unprecedented scale, many of which had an ethnic dimension. Out of 226 armed conflicts recorded for the period between 1946 and 2002, 116 of them have occurred after the end of the cold war in 1989, all but seven of them being intra-state conflicts (Eriksson/Wallensteen/Sollenberg 2003: 593). It requires no statistical analysis to acknowledge ethnic divisions as one of the most serious sources of today’s violent conflicts. In this context, it has been asked whether consociational democracy is a suitable and appropriate model to accommodate the diverse interests and cultures of groups in a multiethnic society. Is it a sustainable model able to prevent conflicts from turning violent? Should it be part of peacebuilding efforts in a post-conflict society? If applied, how should a consociational design look like? The literature on these questions is abundant (See Gurr 1993, McGarry/O’Leary 1993, Lijphart 1995a, Harris/Reilly 1998, Schneckener 2002b).

In order to answer the central question as to what extent consociational democracy can assume a viable democratic model for multiethnic societies; this paper relies exclusively on existing literature - rather than on independent analysis of a few cases studies. Thereby, it seeks to provide a general theoretical review on the issue. The analysis is structured in four parts.

Chapter one describes the concept of consociational democracy as outlaid by Lijphart, presents definitions and central components. Chapter two summarises the critique to the model, ranging from the deficient definitions, to the claim that most of the original cases do not fit the model, to the contention that consociationalism would be insufficiently democratic. The focus of chapter three rests on the ‘favourable factors’; it looks at the conditions under which the development of consociational democracy can flourish. Both structure- and actor-oriented factors have been object of dispute, and Lijphart himself has modified them substantially over time. The final chapter then surveys crucial aspects to a sustainable consociational layout in a given case. According to Lijphart, there exists “no single consociational blueprint” (Lijphart1982:175). Nevertheless, lessons can be drawn from past and present cases that are worth considering in potential future applications.

1 The Concept of Consociational Democracy

Arend Lijphart defines consociational democracy as a model with four characteristics: executive power-sharing or “government by a grand coalition”, a high degree of segmental autonomy, proportionality as the standard of political representation, and mutual veto rights (Lijphart 1977: 25).¹ The first part of this chapter presents the historical and theoretical background to the concept, while subsequent parts are to illustrate each of the four features. The final part focuses on Lijphart’s explanation of how consociational democracies come into being.

1.1 Theoretical Background

The decade of the 1960s saw a major debate among political scientists over the question of how far the stability of political regimes requires a homogenous or integrated political culture. The prevailing view, as exemplified by Gabriel Almond (See Almond 1956; Almond/Verba 1963), regarded the “Anglo-American” model of a two-party system with a majority government and a homogenous, secular political culture as inherently stable and efficient. In contrast, “Continental European” democracies with its multiparty systems, coalition governments and heterogeneous social structures were seen as inherently weak and unstable. However, in a 1956 article Almond differentiated between the Anglo-American, the Continental European, and a third category, consisting of Scandinavian and the Low Countries. According to him, these countries combined “some of the features of the Continental and the Anglo-American” systems, and stood “somewhere in between the Continental pattern and the Anglo-American” (Almond 1956:392- 393). As a Dutchman who had studied and taught in the U.S., Lijphart focussed on this latter and only vaguely analysed category, rejecting the idea that merely Anglo-American type democracies were capable to deliver stable and effective democratic government.

At the World Congress of the International Political Science Association 1967 in Brussels Lijphart and Gerhard Lehmbruch presented two papers that challenged the basic linkage between political stability and social structure. Having worked independently of each other, they showed case studies of countries that were politically stable in spite of political and social fragmentation. According to their findings, the Netherlands (see Lijphart 1968), and Austria and Switzerland (see Lehmbruch 1967) had in common an elite behaviour of co-operation rather than competition.

Accordingly, Lijphart adopted the term ‘consociational democracy’.² After having provided an explanation for political stability in the culturally fragmented societies of Austria, Switzerland, the Netherlands and Belgium (see Lijphart 1969), other cases were found and analysed which fitted the consociational model. Among them were Lebanon 1943-1975, Cyprus 1960-1963, Malaysia 1955-1969, Surinam 1954-1975 and Colombia 1958-1974.³ More recent studies focussed on Northern Ireland (1973/74 and since 1998), South Tyrol (since 1972), Bosnia-Herzegovina (since 1995), India (since 1947) and South Africa (since 1994).⁴ Lijphart and others refined the concept and concentrated on the factors conducive to the advent of consociational democracy, thereby enhancing not only the explanatory, but also the normative power of the theory (Instead of many: see Bogaards 1998). Latest publications looked at the concrete design that a successful consociational framework should feature (See Schneckener 2002a, 2002b, van den Borghe 2002). Thus, the focus of literature on consociational democracy, including Lijphart’s, has shifted over time from explaining political stability in plural societies to the theory’s normative power. To Kenneth McRae, this comes as no surprise. He comments:

“The Lijphart message was an exciting one because of its fundamental optimism. It offered a normative model that could be tried in a wide array of situations of intergroup conflict, even when prospects for success appeared bleak. Its only apparent limit lay in the capacity and the will of leaders of the groups involved to understand the situation and devise appropriate solutions by negotiation. Since these limits can never be known in advance, the consociational approach can lay claim to a very broad mandate” (McRae 1991:96).

1.2 Executive Power-Sharing

The first and most important element of consociational democracy is government by grand coalition, which Lijphart defines as a device by which “the political leaders of all significant segments of the plural society cooperate […] to govern the country” (Lijphart 1977: 25). By implication, consociational democracy thus lacks significant opposition. Lijphart stresses this point by contrasting executive power-sharing⁵ with the majoritarian Westminster-type democracy. While the British model features a style of leadership that is competitive and adversarial, it is coalescent in the consociational one (Ibid). The grand coalition can come in various shapes. Parliamentary systems fit consociational arrangements most easily. Here, the executive is formed by a cabinet, which consists of representatives from all groups. The archetypical grand coalition is represented in the Swiss Federal Council (Bundesrat). Since the inception of a so called “magic formula” in 1959, its seven members roughly proportionally represent the strength of their parties as well as religious and language backgrounds, and make decisions by consensus (See Lehmbruch 1967).⁶ Another, though less sophisticated case of a grand coalition cabinet could be found in Austria between 1945 and 1966 with proportional delegations of the two major parties, the Socialist SPÖ and the Catholic ÖVP (See Lijphart 1977:32, Lehmbruch 1967:20-26). In Belgium and the Netherlands, the shape of executive power-sharing is different. Although there has never been an institutionalised form of grand coalition cabinets (involving all major parties), most of the cabinets have been larger than minimal winning size (See Lijphart 1977: 32). Furthermore, informal arrangements such as permanent or ad hoc “grand” councils constituted another element of executive power-sharing (See ibid).⁷

Despite their greater compatibility with parliamentary systems, executive power-sharing arrangements can be and have been formed in presidential systems as well. A case in point is Lebanon from 1943 to 1975, where the constitution ruled that the president must be Maronite, the prime minister Sunni, and the cabinet be composed according to a roughly proportional formula. However, since the presidency is widely regarded as the most important and prestigious office, such a regulation is not unproblematic: there can always be only one president (See Lijphart 1999b: no page number).

1.3 Segmental Autonomy

Segmental autonomy entails minority rule: rule by the minority over itself in the area of the minority’s exclusive concern. This “logical corollary to the grand coalition principle” (Lijphart 1977: 41) is also phrased principle of subsidiarity. On all matters of common interest, decisions are made by all the segments together with roughly proportional degrees of influence. On all other matters, decisions and their execution are left to the separate segments. The delegation of legislative and executive powers to the segments (e.g. ethnic groups), together with the proportional allocation of government funds, represents a strong stimulus to the various segmental organisations. Thus, it deepens the segmental cleavages, or, to use Lijphart’s phrase, it “increases the plural nature of an already plural society” (Ibid: 42).

Segmental autonomy can be based either on territoriality or personality. Where groups mutually exclusively inhabit regions of a country, where the cultural, ethnic or linguistic borders coincide with territorial ones, autonomy is arranged easily: each area with the group populating it is granted (territorial) autonomy. A special form is federalism - although federalism can of course be applied in non-plural societies as well. Where segments are geographically too interspersed, segmental autonomy can be established on personality. For example, non-territorial segmental autonomy has played a significant role in the realm of cultural affairs, especially education and communication, in the Netherlands, Austria and Belgium (See Lijphart 1977: 44).

1.4 Proportional Representation

Lijphart vaguely defines the principle of proportionality by its two functions. First, it is a method of allocating civil service appointments and funds among the different segments in proportion to their numerical strength (See Lijphart 1977: 38). Proportionality serves as a manifestation of the grand coalition and can entrench all parts of the administration, military and other public services. As a “neutral and impartial standard of allocation” (Ibid: 39) it removes a large number of potentially divisive problems. Second, proportionality means that all groups influence decisions in proportion to their strength. It is thus a refinement of executive power-sharing: not only should all significant segments be represented in decision-making organs, but they should also be represented proportionally. Again, the Swiss ‘magic formula’ provides a classical example.

Two variations of the principle of proportionality are depicted by Lijphart: the deliberate overrepresentation of small segments, and parity of representation. Both are devices that give added protection and security to small segments, and can be sensibly deployed in societies whose segments are of unequal size (See ibid: 41).

1.5 Mutual Veto Rights

The final characteristic of consociational democracy is the right of each segment to veto a decision. It thus represents negative minority rule. Where decisions in a grand coalition are made by majority vote, minority segments can easily be outvoted. When such a majority decision harms the vital interests of a minority segment, this defeat might be regarded as unacceptable and could lead to the collapse of inter-segmental elite co-operation. Only if each segment holds veto rights, it has a complete guarantee of political protection (see ibid: 37).⁸ The great danger of the veto rights, however, is that it can lead to minority tyranny and subsequent immobilism. Acknowledging this risk, Lijphart argues that for three reasons it is not as serious as it appears. First, he states that “the too frequent use of the veto by a minority is not very likely because it can be turned against its own interests, too” (ibid). Second, the very existence of the veto as a potential weapon would give a feeling of security, in turn leading to a reduced probability of its actual use. Third, he contends that each segment would recognise the danger of deadlock as a consequence of unrestrained veto use (See ibid). All points are disputed, as will be shown further below. Nevertheless, a standard practice exists in consociational democracies that appears to reduce the likelihood of immobilism: the linking of several divisive issues and their simultaneous solution through reciprocal concessions to a package deal.⁹

1.6 Establishment of Consociational Democracies

The question arises? How do consociational democracies come into being in the first place? Arend Lijphart views the rational choice of elites as central, which have "an awareness of the dangers inherent in segmental cleavages and a desire to avert them" (Lijphart 1977: 100). According to him, this awareness may result from a previous conflict that has forcefully shown the perils of segmentation and leads elites to self-restraint and coalescence (See ibid). In his 1977 book he acknowledges another factor put forward by other authors: traditions of elite accommodation. (See Lijphart: ibid; Lehmbruch 1967, Daalder 1974: 618).However, he stipulates these traditions as a merely favourable factor and not an explanation.

[...]

¹ In later articles Lijphart further differentiates between essential primary characteristics (executive power-sharing and autonomy) and supplementary secondary characteristics (proportionality and veto rights). See Lijphart 1995b, 1999b.

² The term ‘consociation’ is derived from the Latin ‘consociatio’, describing a close and firm link, e.g. a durable link between members of a co-operative. See Schmidt (2000): 327. Lehmbruch’s originally used term ‘ Proporzdemokratie ’ was soon abandoned for ‘ Konkordanzdemokratie ’, the German equivalent for consociational democracy.

³ For the case of Surinam see Dew (1972); for Colombia see Dix (1980); for Lebanon see Lehmbruch (1974); for Lebanon, Cyprus and Malaysia see Lijphart 1977: Ch.5 (pp. 142-176).

⁴ For literature on the case of Northern Ireland see McGarry (2002) and Schneckener (2002b); for South Tyrol and Bosnia-Herzegovina see Schneckener (2002a, 2002b); for India Lijphart (1996) and Lustick (1997), for South Africa Lijphart 1985, Laitin 1987, Lijphart 1999b.

⁵ The terms ‘executive power-sharing’ and ‘government by grand coalition’ are used interchangeably, as they are by Lijphart himself. See for example Lijphart 1995b, 1999b.

⁶ The ‘magic formula’ gave two seats each to the FDP, CVP and SP, and one to the SVP. After an electoral gain of the SVP in October 2003 the 2-2-2-1 formula remains intact, though with the shift of one seat from CP to SVP it has lost its proportional ‘magic’. [http://www.nzz.ch/dossiers/2003/bundesratswahlen (19.12.2003)].

⁷ For example, informal temporary grand coalitions of party leaders were formed in the Netherlands in 1917 and in Belgium in 1958 in order to settle the deeply divisive issue of state aid to religious schools. See Lijphart 1977: 33.

⁸ The mutual veto can be either an informal understanding or a formal rule, as Lijphart shows with the cases of Switzerland, the Netherlands (both informal), Austria (formal) and Belgium (formal on certain issues). See Lijphart 1977: 38.

⁹ Package deals are also termed logrolling, or, in Austria, Junktim. See Lehmbruch 1967: 26-29.