QUALITY OF DEMOCRACY IN AFRICA: THE IMPACT OF OPPOSITION PARTIES. PROF. DIRK KOTZé DEPARTMENT OF POLITICAL SCIENCES, UNIVERSITY OF SOUTH AFRICA

Transcription

1 QUALITY OF DEMOCRACY IN AFRICA: THE IMPACT OF OPPOSITION PARTIES PROF. DIRK KOTZé DEPARTMENT OF POLITICAL SCIENCES, UNIVERSITY OF SOUTH AFRICA INTRODUCTION Quality of democracy as a focus of scientific research appears to be a natural progression from the scholarly focus on consolidation of democracy, which in turn is a progression from the transition and democratization literature. It is also associated with the third and fourth waves of democratization. In the existing literature on the quality of democracy, starting with Vanhanen (1990), Lijphart (1999) and later Altman & Pérez-Liñán (2002), the Journal of Democracy (October 2004) and Diamond & Morlino (2005), and O Donnell, Vargas Cullell & Iazzetta (2004) a clear distinction between two approaches emerged. The one approach to assessing the quality of democracy assumes that democracy is a global phenomenon with more or less similar characteristics. Democracies or states can therefore be compared or assessments on the same criteria can be made. Such assessments are normally quantitative and often involve ranking of countries. Examples of this approach are the Freedom House ratings, the Polity index, Lijphart (1999) and the Vanhanen index. The second approach to assessing the quality of democracy emphasises the specific contexts of countries and argues that the most appropriate assessment of the quality of democracy has to be done by the citizens and not by disinterested researchers. The approach is therefore qualitative, context- and country-specific and perceptional or constructivist. Examples of it are the State of Democracy framework designed by the International Institute of Democracy and Electoral Assistance (IDEA, 2008) and the Democracy Index on South Africa designed by the Institute for Democracy in South Africa (IDASA) (Calland & Graham 2005). As will be explained later, this paper follows the tradition of the first approach, though not denying the inherent merit of the second approach. The perceptions of citizens are exceptionally important for how they experience the quality of democracies. One justification for preference to the first approach, is Staffan Lindberg s observation about renewed recognition for Rustow s (1970) argument that democratic behaviour produces democratic values, and not the other way round (Lindberg 2006: Preface). Behaviour should therefore receive primary attention. The empirical dimension of behaviour can be compared, while the perceptional nature of values is less suitable for comparison. 1

2 Studies of the quality of democracy look at most of the world. For instance, Vanhanen 1 included 147 states in his index, Lijphart 2 included 36 states, Dahl s Polyarchy included 114 states, though his focus was more on categorisation of regime forms. More focused studies include these on established democracies (especially the OECD) 3, post-communist Europe 4, Latin America, South Korea and Taiwan 5, and India and Bangladesh 6. Studies about Africa are much rarer. Dirk Berg-Schlosser 7 looked at Africa in general, while Lijphart included Botswana and Mauritius in his 36 states. Lindberg (2006) concentrates on the impact of elections on the quality of democracy in African states. A specific focus on Africa in this article is therefore warranted. PROBLEM STATEMENT Two sets of criteria very influential in assessing the quality of democracy, are those developed by Lijphart (1999) and by Diamond & Morlino (2005). Lijphart (1999: chapter 16) used the following criteria: women s representation, political equality, electoral participation, satisfaction with democracy, government-voter proximity, and accountability and corruption. Diamond & Morlino (2005: xii) identified the following: rule of law, participation, competition, accountability, respect for civil and political rights, and implementation of greater political equality. In both instances an explicit reference to opposition parties (beyond elections as a form of public participation) is virtually absent. The closest they came was in the form of contestation (Diamond & Morlino 2005: xvii-xix), operationalised as the type of electoral system, the constitutional and legal order, access to the mass media and campaign funding. Lijphart s (1999: ) electoral participation was operationalised as voter turn-out. 1 Vanhanen, Tatu The process of democratization: A comparative study of 147 states, New York: Crane Russak. 2 Lijphart, Arend Problems of democracy: Government forms and performance in thirty-six countries. New Haven & London: Yale University Press. 3 Hardmeier, Sibylle. Quality of democracy Democracy Barometer for Established Democracies. NCCR democracy barometer, University of Zürich. 4 Hutcheson, D The quality of democracy in post-communist Europe. Abingdon: Routledge 5 Doh, Chull Shin & Yun-han Chu The quality of democracy in South Korea and Taiwan: Subjective assessments from the perspective of ordinary citizens. Paper presented at a conference organised by the Center on Democracy, Development and the Rule of Law and the European Forum, Institute for International Studies, Stanford University, October 10-11, < Korea_Taiwan.pdf 6 Ganguly, Sumit. The quality of democracy: Assessing India and Bangladesh. Paper presented at the annual meeting of the American Political Science Association, Hilton Chicago and the Palmer House Hilton, Chicago, IL, September 02, 2004 < 7 Berg-Schlosser, Dirk Indicators of democracy and good governance as measures of the quality of democracy in Africa: A critical appraisal. Acta Politica 39: 3,

3 Vanhanen (1990) provided more specific indicators closer to opposition politics by concentrating on two related factors, namely 1) participation in the form of voter turn-out (similar to Lijphart s); and 2) the degree of competition, operationalised as the share of the vote received by all the parties, except the largest one. Lindberg (2006: Preface) introduced specifically the impact of elections on the quality of democracy in the following words: [T]he analyses show that the inception of multiparty elections in a country instigates liberalization and that repetitive electoral activities create incentives for political actors by fostering the expansion and deepening of democratic qualities in the society.... The study also shows that a sequence of elections not only contributes to increasing the democratic quality of a political regime but also broadens and deepens civil liberties in the society. Though not yet fully developed, all of these are suggestions about the relevance of opposition parties (as the outcome of elections) and they provide a measure of justification for this paper s theme. It leads us towards the question: what are the specific dynamics in Africa that warrant more research? The events in South Africa in September/October 2008 which caused a major split in the ruling ANC, and which resulted in a re-alignment of almost all the opposition parties, have been hailed as a significant improvement in the quality of democracy in South Africa. It stands in contrast to the recent election outcomes in Kenya and Zimbabwe, where the opposition was so strong and the results so close, that the losers had enough confidence to challenge them. In terms of conventional democratic theory the expectation should be that in those countries with strong opposition parties the quality of democracy should be much better than in those where one party is dominant, such as in Botswana, South Africa or Senegal. This paper investigates a possible correlation between the strength of opposition parties in political systems in Africa, and the external assessment of those systems quality of democracy. At this stage it is already possible to observe that the strongest democracies in Africa (South Africa, Senegal, Botswana, Mauritius, etc) do not have strong opposition parties. On the other hand, states with problematic democratic histories (Zimbabwe, Kenya, Guinea-Bissau, Malawi, etc) have all diverse multiparty systems and strong oppositions. If a correlation between opposition and democratic quality is ostensibly absent in Africa, how can it be explained? The objective of this paper is to investigate the apparent contradiction, and to explore a formula in terms of which the quality of contestation can be determined. The focus here will be confined to states in Africa, and its conclusion will not claim universal relevance. CONCEPTUAL CONTESTATION At the conceptual level democracy has become so hotly contested that it holds serious implications for our understanding of the quality of democracy. Contestation includes the 3

4 opposing views on procedural and substantive (or liberal and social) democracy; it is also about the difference between citizens control over versus influence in government s decision-making (partly expressed in majoritarianism versus proportionality); and is about the purpose of democracy (as a means towards an end, or as an end in itself especially in the manner it structures state-society relations). Claude Ake (2003: 29) was very outspoken about the implications of these contestations for Africa: By all indications, the established democracies are not able to offer the developing countries clear and meaningful standards of democratization to relate to. This is because they themselves have no clear and meaningful standard of democracy, a condition arising from the growing alienation of the practice of democracy the Western ideology of democracy and the concerted effort by powerful interests to deracialize democracy by offering a profusion of definitions which trivialize it. This conceptual contestation has also relevance for our understanding of the relationship between the quality of democracy and: - conceptualising and measuring democracy: many works have as their objective to determine the essential characteristics or criteria of a democracy, and thereby being able to develop a check-list. According to Ringen (2007: 19-20), the Polity project 8 is a good example of an effort to measure democracy. Other similar efforts are Afrobarometer, IDASA and the IDEA projects. Does the meaning of democracy includes an qualitative assessment or does it merely identify and register the defining characteristics of a democracy? The question is whether the quality of democracy is determined by how well the identified characteristics function, or is it determined by other considerations? - consolidation of democracy: consolidation is concerned with the depth and maturity of a democracy. Is it therefore not per definition synonymous with the quality of democracy? Can we say that the more consolidated it is, the better its quality? Consolidation theories use perceptional, empirical (behavioural) and value criteria to determine the progress of consolidation. Should the quality dimension be linked to them, or to other? - good governance: closely associated with donor conditionality in the developing world, good governance often assumes a value-laden meaning in the form of neoliberal market restructuring, conservative fiscal policies and multiparty elections. It is often accused of undermining state capacity and responsibility in favour of the private sector and individual responsibility. Good governance is, however, also associated with prudent procedures, such as anti-corruption measures, sound management practices and effective delivery of public services. The Mo Ibrahim Index of African Governance, the New Partnership for African Development (NEPAD) and the African Peer Review Mechanism all use good governance as a central tenet in their 8 Monty G. Marshall, Keith Jaggers, (Ted Robert Gurr). Polity IV Project: Political regime characteristics and transitions, < 4

5 assessment of states. Is good governance therefore equal to the quality of democracy? Andrew Roberts (2005: ) wants to reduce the quality of democracy to the questions: How well do democracies work? How well do they produce responsive, accountable and representative governments? For him the central concern of democratic quality is the connections between citizens and their elected representatives. It is about institutionalising links between the preferences of citizens and actions of elected representatives. Therefore he concludes that governments responsiveness is the moral determinant of democratic quality. Roberts views do not resolve much of the uncertainty and contestation. The question is whether quality is determined by the presence of a range of relationships (citizens representatives), or whether it is about the intensity, maturity and depth of a democratic environment? The suggestion is made here that the quality of democracy depends on its intensity or density, its sustainability, the level of public satisfaction with democracy (in the form of public acceptance and public participation in it), its reach in society (how inclusive it is of the population, and how effective the state, represented by government, can disperse the democratic dividends also to the perifery of society, while at the same time it can enforce democratic behaviour). Quality is not about the presence of certain democratic characteristics but about their degree of success and the impact they make on society. In his work on Africa Claude Ake did not directly address the quality of democracy as a consideration. His main effort was to argue for a symbiosis between democracy and development. Democracy is in his mind necessary for the sake of Africa s development (Ake 1996, 2003). It implies that the quality of democracy will depend on the contribution it can make towards development. It furthermore implies that indicators of development will automatically be indicators of democratic quality (similar to Lijphart s logic on political equality, mentioned below). Ringen (2007: 18) argues that it is not enough to observe a state s economic potential: its performance, and the quality of life it produces, must also be included. Therefore the UN Development Programme s (UNDP) Human Development Index (HDI) is a good example of combining potential with performance. For Ringen it is an appropriate index, because it is based on a state s economic potential and on the standard of health and education. Emphasis on development re-introduces the debate about the research focus of democratic quality: should the focus be on the regime or government, or should it be determined by the members of society? For Ake the state is very important as a developmental state with a social responsibility, and therefore the government cannot be excluded. For Ringen the emphasis should be on the people and not the regime. In concluding this argument, and for providing a research basis for this paper, we embrace the criteria of quality identified by Lijphart (1999) and Diamond & Morlino (2005) but they are not sufficient for assessing developing states or young democracies. Three more criteria are therefore suggested, namely: 1) the quality of life, articulated by the HDI: the index utilises the assumption that the higher the quality of democracy, the higher will be the quality of life (which is a variation on the notion that democratisations leads to development). The HDI is a 5

6 combination of economic growth, life expectancy, literacy and educational indices, and is one of the most authoritative indicators of the quality of life. It can be related to Lijphart s contention that economic quality is a valid proxy for political equality, since political equality is more likely to prevail in the absence of great economic inequalities (Lijphart 1999: 282); and 2) the level of good governance, including the absence of corruption: the quality of the public sector, of the government in general, and their ability to take decisions in the best interest of the greatest part of the society, combined with their ability to provide services that will enable maximum development in the society, is arguably the ideal notion of good governance. It implies also minimisation of hindrances to governance, such as corruption, nepotism, inefficient management and political instability. Useful indices in this regard are Transparency International s Corruption Perception Index and the Mo Ibrahim Foundation s Index on African Governments; and 3) absence of electoral conflicts or crises: in recent years reverse waves of democratization appeared in states with promising prospects for democratisation, because of electoral crises. Examples are numerous, including Lesotho (1998), Kenya (2007) and Zimbabwe (2008). Conflicts during elections, such as those regularly appearing in Zanzibar, undermine their democratic quality. The absence of electoral conflict is presumed an indicator of political tolerance and of the quality of multiparty contestation. A related indicator is media freedom during elections (or detention of journalists). In order to avoid an imbalance in this paper in favour of elections and contestation, only the first two criteria will be integrated into the rest of the study, while the latter one will be subsumed in the proposed notion of the quality of contestation, following below. In view of the fact that this article proposes an investigation of a possible relationship between the quality of contestation and the quality of democracy, the connection between elections and opposition parties automatically demands attention. It is briefly introduced in the next section. ELECTIONS AND THE OPPOSITION Adam Przeworski is well-known for his view that democracy is a system in which parties lose elections or a regime in which those who govern are selected through contested elections (Przeworski et al. 2000: 15-16). The question is: what enhances the quality of such a democracy? Przeworski et al. did not address it directly. They utilised the concept contestation when an opposition exists that has the opportunity of winning elections in the near future. In their view, contestation (by means of elections) should be characterised by (1) ex-ante uncertainty, (2) ex-post irreversibility and (3) repeatability (Przeworski et al. 2000: 16). Przeworski is uses the reductionist view that all aspects of democracy can be derived from elections and contestations. He does not, however, present a qualitative assessment of contestation beyond the mere presence of specified procedural criteria. 6

7 Andrew Roberts (2005: 358) takes it a step further by connecting elections to the quality of a democracy. For him, democratic quality depends on the connection between citizens and their elected representatives. The connection concerns the nature and strength of the links between mandates, responsiveness and accountability which in effect, capture the Lijphart/Diamond-Morlino criteria of democratic quality in the notion of elections. G. Bingham Powell, Jr (2000: 4) also supports the notion that competitive elections, more than any other feature, identifies a democratic political system. But in his mind they are not the only instruments of democracy, and there are serious disagreements about elections. They are partially normative, partially conceptual and partially empirical. Though this paper does not concentrate on electoral systems as an independent variable in determining the quality of democracy, most of the discussions about elections and democratic quality are in the form of the debate involving Lijphart (1999, 2008), Bingham Powell (2000), Doreenspleet (2005) and others about majoritarianism versus proportionalism. The debate is about how elections and the choice of electoral systems can optimise democracy. Lijphart is a strong proponent of proportionalism (or consensus democracy) as the optimum form of democracy. The choice between majoritarianism and proportionalism is not based on objective criteria, and therefore it raises the question: does the difference in the procedure of elections, or the difference in allocation of political power in a procedural sense, or its procedural impact on the formation of governments, make a difference in the quality of democracy? The answer is arguably embedded in the protagonists perception of the power relations in society, and therefore how these relations can potentially be organised by means of elections. Is the society plural in nature, does it need to recognise minorities, are the minorities ascriptive or descriptive in nature, does Przeworski s ex-ante uncertainty apply to the results of elections? Are the relations determined by class, race, gender, culture, regionalism or history? The fundamentally ideological and existential nature of the different society conceptions makes this debate perennial in nature. The approach followed in this article leans towards proportionalism (but not necessarily consensus ). At the same time it should be made clear that majoritarianism and proportionalism are not necessarily mutually exclusive. Mozambique is a good example where the two are present in the same system. South Africa is also not far away from it. For the purpose of this discussion it is necessary to refer to some of the well-known functions of elections, because only then can the quality impact of elections be considered. Three perspectives on the functions are presented: the contestation perspective, the watchdog perspective and the minority representation or coalition perspective. 1) Contestation: The contestation perspective of elections can be separated in two: the democratic value of the process of contestation, and the value of outcomes of contestation. The first one emphasises the importance of public participation in which both the parties and voters can express their freedoms and political rights. Elections are important symbolic moments to reaffirm the social contract and the sovereignty of the voter. Elections outcome is less significant, because the argument is that the process legitimised the outcome. Whether the outcome is therefore a single-dominant party regime like in Japan, Botswana or South 7

8 Africa, or a coalition government in the Netherlands or Germany, its democratic nature is not questioned. The second perspective emphasises elections in terms of outcomes: how political power is distributed by means of elections? Regular turn-over of governments is therefore essential to democracy as a counter for the negative effects of majoritarianism (compare Lijphart and Dahl). What is the balance of power between political parties as a result of the election? How strong is the official opposition? How diverse is the party political spectrum? How effective is the legislature in relation to the executive? 2) Watch-dog: The watch-dog perspective is premised on the separation of powers doctrine especially separation between the legislature and executive. Elections are key instruments in maintaining this separation, because every election is an opportunity to strengthen the opposition or change the government. Elections can therefore directly be related to the strength and relevance of parliaments. The watch-dog notion assumes that opposition is the primary instrument for enforcing a government s accountability to elected representatives. The assumption continues that a strong opposition strengthens a parliament in relation to the executive. Hence, the stronger the opposition, the better is the quality of democracy. A complication in this perspective is presidential elections. They strengthen the separation of powers between the executive and legislature but they also complicate executive accountability to the legislature. If a president becomes unpopular but does not act unconstitutionally, a legislature has limited power to sanction the president, except withholding budgetary approvals and similar decisions. In relatively new democracies where most legislatures are weaker than the executive, directly-elected presidents can weaken them more, even if the opposition in such legislatures are very strong. Parliamentary systems or hybrid presidential systems provide more opportunities for the legislature to be strengthened. The South African Constitution, for instance, provides not only for presidential impeachment, but also for a parliamentary motion of no confidence in the President. Almost a year ago the governing ANC recalled the President, he resigned and a new President was elected by Parliament. A presidential system entrenches a president against such parliamentary intervention. 3) Minority representation / coalition: Pluralism and consociationalism are sensitive to prevent distortion of the societal spread of power by a too dominant majority. The challenge is therefore how to accommodate diversity not only in political participation (competitive elections) but also in representation (or influence on government). An obvious focus area is therefore the electoral system (see Lijphart and Bingham Powell). Another relevant area is the presidential versus parliamentary system, that determines legislative-executive relations. The more politically homogenous a society is and the more entrenched Przeworski s ex-ante uncertainty is, the more a government opposition polarisation can be reconciled with democracy. The more heterogeneous the political system is, and the more political identities are ascriptive, the more is the demand for minority representation or coalitions a condition for democracy. For this study s purpose the quality of democracy will be determined not only by the presence of 8

9 a strong official opposition party, but by the degree to which the legislature and executive accommodate the diversity of a society. Our focus moves therefore to an investigation of the nature of contestation and how it is translated into a representative contestation process. For this purpose we use the notion of the quality of contestation. QUALITY OF CONTESTATION : A FORMULA Democratic practice is most of the time concerned with government-formations (majority or coalition governments) and the presence of a strong opposition party. Contestation insofar as it determines the remainder of the opposition spectrum is largely inconsequential for democracy, especially for its majoritarian variation. As already mentioned earlier, in most of the African states classified as free by Freedom House, there are neither coalition governments nor a strong official opposition party. On the other hand, the strongest opposition parties are found in partly free or not free states such as Sierra Leone, Kenya and Zimbabwe. The assumption used in this paper is therefore that the relationship between the governing and strongest opposition parties is not sufficient as an indicator of the quality of democracy or the quality of contestation. A broader perspective on contestation as part of democracy will have to be investigated. Our intension is not to present the quality of contestation as an alternative for the criteria identified by Lijphart (1999), Diamond & Morlino (2005) and others, but to explore possible relationships between contestation and the other criteria. Excluded from this study is the possible impact of electoral systems on the quality of contestation, as well as popular perceptions about elections and opposition as a determinant of democratic quality. This study can be criticised for an ingrained bias against two-party systems, particularly in the proposed formula below. Seventeen African countries have been identified for the purpose of this study, namely Benin, Botswana, Ghana, Guinea-Bissau, Kenya, Lesotho, Liberia, Mali, Mauritius, Malawi, Mozambique, Namibia, Senegal, Sierra Leone, South Africa, Zambia and Zimbabwe. They include all nine free states on the continent identified by Freedom House. Initial observation of the party political situation and election results in all of these states did not reveal a clear indication of their state of democracy. A more systematic analysis is therefore warranted. The first step is to determine the gap between the first and second parties in elections. The gap indicates how intensely the competition for the status of government is. We prefer to call this index the intensity of competition and it is presented as follows: A = Intensity of competition a = (percentage of strongest party) (percentage of 2 nd party) As indicated already, index A is insufficient, because those states with the highest intensity scores (such as Sierra Leone, Malawi, Zimbabwe and Guinea-Bissau) are not the most democratic states in Africa. Additional variables are therefore required. 9

10 In view of the fact that the gap between party 1 and 2 is insufficient as an indicator, and considering the minority representation/coalition perspective discussed earlier, an alternative is to determining how many parties are involved in closely contested elections. The alternative will present an indication of how inclusive the political system is, how much tolerance is displayed for a diverse range of political views, and that parties are able to mobilise support in elections for those views. An index is therefore suggested for the number of parties within 15% of the first one. The focus is therefore on the major parties and it excludes inconsequential parties. The dividing factor of parties is increased by 1 to produce calculations which can differentiate small degrees of distinction (especially in instances where the gap between the first and second parties is larger than 15%). In order to develop a relationship between the intensity of competition and the number of parties in close contestation, index A is divided by the number of parties to produce index B: density of competition. Density suggests how close to each other the contesting parties are: the expectation is that the more and the closer they are, the more it contributes towards teh quality of democracy. It is articulate in the following equation: B = Density of competition Table 1 below shows that Sierra Leone has the highest competitive density, followed by Malawi, Guinea-Bissau, Zimbabwe, Liberia, Ghana, Mauritius and Zambia. Amongst them only Ghana and Mauritius are free societies. On the other hand, Senegal, Namibia and South Africa, as free societies, perform much worse on the density index. Due to this inconclusive situation additional indices should be explored. The obvious challenge is to determine the quality of contestation. The first option was to divide index B into index A. Because the indices in A range between 1.19 and 74.6, they distort the equation s results. Hence, after much testing, the following equation appears to be more sensitive for all the variations: C = Quality of contestation Table 1 shows that Sierra Leone, Malawi, Guinea-Bissau, Liberia and Zimbabwe score the highest on this index as partly free or not free societies. Senegal, Namibia, Mali, South Africa and Lesotho score the lowest, while all of them are free societies. In order to compare the proposed quality of contestation index for parliamentary elections with other indices of democratic quality, the following are included: 10

11 the Mo Ibrahim Index on African Governance: this index measures good governance in Africa. One of its indicators participation and human rights is used for this comparison; Freedom House s indicators are well-known, namely political rights and civil liberties, rated on a 1-7 scale. It classifies states as free, partly free or not free. For the purpose of this study, the category free is considered as more or less democratic; the Human Development Index: as indicated earlier, it reflects the quality of life based on economic growth, literacy and life expectancy variables; the Corruption Perception Index: corruption is one of the indicators of good governance, effective accountability and parliamentary oversight. All of them are incorporated in table 1 below: TABLE 1: QUALITY OF CONTESTATION IN COMPARISON WITH OTHER INDICES state parliamentary contestation Mo Ibrahim Index (2006) 9 Freedom House (2006) 10 UNDP (2005) 11 Transparency International (2005) 12 A B C Participa tion PR CL status HDI CPI Benin F Botswana F Ghana F Guinea PF Mo Ibrahim Foundation Ibrahim Index of African Governance < 10 Freedom House Freedom in the World Map of freedom < 11 UNDP Human Development Report 2007/2008: Fighting climate change: Human solidarity in a divided world. New York: UNDN/Palgrave Macmillan < 12 Transparency International Corruption Perception Index < 11

12 Bissau Kenya PF Lesotho F Liberia PF Mali F Mauritius F Malawi PF Mozambi que PF Namibia F Senegal F Sierra Leone South Africa PF F Zambia PF Zimbab= we NF General conclusions about all the indices are, firstly, that in the case of Ghana, Benin and Kenya there is a relatively consistent correlation between all the indices. It implies that the proposed quality of contestation at parliamentary level does make a contribution towards assessing the quality of democracy. In the case of Sierra Leone and Guinea-Bissau a significant dissonance exists between the quality of contestation and the other indices. The dissonance is so significant that the contestation index appears to be unreliable. Contestation at parliamentary level does not correlate at all with other indices of democratic quality. In the case of South Africa, Senegal, Botswana, Lesotho and Mauritius the quality of contestation is less significant than the other democratic quality indices. Parliamentary contestation as an index is therefore in this instance also not reliable. The strongest correlations in the table are between HDI and CPI, and between the Mo Ibrahim Index/Freedom House and CPI, which makes corruption perceptions a very strong indicator. The weakest correlations are between Contestation and Mo Ibrahim/Freedom House, followed by Contestation HDI and Contestation CPI. 12

13 Additional explanatory factors clearly still need to be introduced. It is not the purpose of this study to explore them. A preliminary remark in this respect is the presence of a strong, directly-elected executive President as a possible factor. In all the instances of free societies, except in Senegal s, in which the quality of parliamentary contestation is low (i.e. South Africa, Botswana, Lesotho and Mauritius), the head of the executive is either a prime minister or an indirectly-elected President. In the case of Sierra Leone, the parliamentary quality of contestation index is 2.73, while the presidential contestation index is only While its parliament is highly contested, the President is not. It gives the executive unassailable power and suppresses a democratic parliament. CONCLUSION Assessing the quality of democracy is important, because it should provide us with an indicator of democracy s impact on society. The debate between the procedural and substantive notions of democracy is also preoccupied with this dilemma. Over and above moral judgements about democracy and procedural compliance with democracy s core criteria, the main question is whether democracy makes a difference; whether it improves the quality of a society? Many perspectives on democracy reduce its essence to elections and contestation. We are therefore challenged to identify the aspects of elections/ contestation that can serve as indicators of the quality of democracy. A formula is proposed to determine the quality of contestation (for legislatures). A limited correlation or relationship between parliamentary contestation on the one hand, and political rights, civil liberties, quality of life, corruption and good governance on the other hand, has been identified in the African examples. Strong opposition parties and highly contested parliamentary elections are therefore not necessarily a prerequisite for, or an indicator of, a quality democracy. Further qualifications within the notions of contestation and opposition have to be explored in order to determine their relationship within the quality of democracy. 13

14 BIBLIOGRAPHY Ake, Claude Democracy and development in Africa. Washington DC: The Brookings Institution Ake, Claude The feasibility of democracy in Africa. Dakar: CODESRIA Altman, David & Pérez-Liñán, Aníbal Assessing the quality of democracy: freedom, competitiveness and participation in eighteen Latin American countries. Democratization. 9:2, Bingham Powell Jr, G Elections as instruments of democracy: Majoritarianism and proportional visions. New Haven & London: Yale University Press Calland, Richard & Graham, Paul (eds) Democracy in the time of Thabo Mbeki: IDASA s democracy index. Cape Town: Institute for Democracy in South Africa Diamond, Larry & Morlino, Leonardo (eds) Assessing the quality of democracy. Baltimore: The Johns Hopkins University Press Doreenspleet, Renske Electoral systems and democratic quality: Do mixed systems combine the best or the worst of both worlds? An explorative quantitative cross-national study. Acta Politica. 40, IDEA. (Todd Landman, ed) Assessing the quality of democracy: An overview of the International IDEA framework. Stockholm: IDEA Lijphart, Arend Patterns of democracy: Government forms and performance in thirtysix countries. New Haven & London: Yale University Press Lijphart, Arend Thinking about democracy: power sharing and majority rule in theory and practice. New York: Routledge Lindberg, Staffan I Democracy and elections in Africa. Baltimore: Johns Hopkins University Press O Donnell, Guillermo, Vargas Cullell, Jorge & Iazetta, Osvaldo M. (eds) The quality of democracy: theory and applications. Notre Dame, IN: University of Notre Dame Press Przeworski, Adam, Alvarez, Michael E., Cheibub, José Antonio & Limongi, Fernando Democracy and development: Political institutions and well-being in the world, Cambridge: Cambridge University Press Ringen, Stein What democracy is for: On freedom and moral government. Princeton & Oxford: Princeton University Press Roberts, Andrew The quality of democracy (review article). Comparative Politics, 37: 3,

15 Vanhanen, Tatu The process of democratization: A comparative study of 147 states, New York: Crane Russak 15