Sub-Saharan Africa has witnessed the end of foreign colonial rule, the rise and fall of autocratic political regimes, and the disappearance of statist command economies. The challenges were to turn populations into coherent nations owing allegiance to the state; to democratise the state structures that govern these populations; and to liberalise the rules that regulate economic transactions. An important source to assess these prospects are the views and attitudes of ordinary Africans. This essay reflects on the original data derived from a crossnational research project. Nine African states were surveyed between 1999 and 2000. An attempt is made to gather some propositions from the analysis of the data. Many present serious challenges to common wisdom about African politics. It appears that the process of nationbuilding has created coherent political communities with high levels of national identity; that democratising the state in Africa builds on existing indigenous demands from ordinary Africans; and that economic liberalisation proceeds in the face of a mixed set of values about market and state.

Introduction

Over the last 40 years, sub-Saharan Africa has witnessed the end of foreign colonial rule, the rise and fall of autocratic political regimes, and the disappearance of statist command economies. Each of these momentous events has bequeathed equally extraordinary challenges. The first was to turn the populations living within the artificial boundaries carved out by the colonial powers into coherent political communities, or nations who owe their allegiance to the designated state. The second challenge was to democratise the state structures that govern these populations, and the third was to liberalise the rules that regulate their economic transactions.

The ability of states in sub-Saharan Africa to effect this "triple transition"¹ will largely determine whether the continent can bring about an African Renaissance and join the international community as full partners, or whether it will slip further into poverty, conflict and marginalisation. A fundamentally important source of information with which to begin an assessment of these prospects are the views and attitudes of ordinary Africans. To assess their views, an original set of data is used from a large-scale crossnational research project called the Afrobarometer. The data comes from systematic surveys of randomly stratified nationally representative samples in nine African states — seven in Southern Africa (Botswana, Lesotho, Malawi, Namibia, South Africa, Zambia and Zimbabwe) and two in West Africa (Ghana and Nigeria) — all conducted between July 1999 and July 2000. Trained enumerators conducted face-to-face interviews in local languages with a total of 14 975 respondents using questionnaires that contained a common set of core items.²

A caveat is in order about generalisations. Each country sample was drawn independently and randomly representing voting age populations.³ But the nine countries discussed here, which are all Anglophone territories that have recently undergone political transitions to multiparty systems, are not fully representative of the sub-Saharan subcontinent. It is not alleged that the findings in this essay can be extended to Francophone Africa, to the continent’s remaining authoritarian regimes, or to states that are imploding through civil war. If the occasional reference is made to ‘Africans’, it is with a more limited populace in mind.

This essay attempts to gather some propositions from the analysis of the Afrobarometer survey data. While undoubtedly requiring further refinement and testing, many of these propositions present serious challenges to prevailing common wisdom about African politics. At the broadest levels, three generalisations are apparent. First, it appears that, while far from complete, the process of nationbuilding since decolonisation seems to have created coherent political communities with high levels of national identity. Second, the process of democratising the state in Africa builds on existing indigenous demands from ordinary Africans. It is not an alien concept or process imposed from without. It is true that citizens are aware that African democracies remain far from complete, are less than satisfied with the performance of democracy, and do not invest state institutions with a high level of trust. However, while it is far from the ‘overwhelming majority’ often cited as a benchmark of democratic legitimisation,⁴ Africans extend roughly the same levels of legitimacy to their new political regimes as do publics in new democracies in Latin America and Central and Eastern Europe.

Third, while the process of democratisation is aided by a strong wave of public demand for democracy, the process of economic liberalisation proceeds in the face of a much more mixed set of values about market and state. In some respects, Africans support key principles behind economic structural adjustment; in others, there remain strong attachments to the leading role of the state in the economy.

Political community?

The existence of a pre-existing, consensually accepted political community is essential not only for political stability, but also for the consolidation of democracy.⁵ While modern democracy allows people to govern their own affairs through regular elections, political parties, and representative legislatures, democracy cannot indicate which people should be included in the process of ruling themselves in a given political unit. The identity of ‘the nation’ that is supposed to govern itself democratically is largely the result of a process of building psychological affinities between prospective citizens and the political territory where they live, its symbols and its prominent values.⁶

Given the diverse nature of the artificial societies lumped together by colonialism, most analysts have assumed that people identify first and foremost with their primary social or solidarity group and only secondarily — if at all — with the post-colonial national political entity.⁷ To the extent that they coexist, these identities are seen to be in tension with each other, if not mutually exclusive.⁸ In the surveys, these issues were addressed with two sets of identical questions. The first probed people’s feelings about their national political identity. The second measured their feelings toward their self-selected subnational identity. In contrast to the received wisdom, the results reveal an almost consensual, shared national identity in each of the nine countries surveyed. If these nations were once artificial, they are no more.

In eight of the nine countries, majorities ranging from 89% to 97% stated that they were proud to be called citizens of their country, and that they wanted their children to think of themselves as citizens of the country. In Zimbabwe, the proportions agreeing with these statements dropped to 84% and 78%, respectively. Even if Africans feel a strong personal connection to the nation-state, they may be less willing to extend equal citizenship status to others, particularly those in specific ‘out groups’. But in eight of nine countries, majorities ranging from 87% to 94% agreed that all people born in their country should be treated as equal citizens, regardless of the group they belonged to. Moreover, majorities ranging from 82% to 94% agreed that it was desirable to create one united nation out of all people living in the country. On this question, Lesotho was the exception with agreement indicated only by seven out of ten respondents.

However, none of this should be taken to suggest that communal identities are no longer important in African politics. People were asked: "beside being a [eg Nigerian], which specific group do you feel you belong to first and foremost?" In four countries, the most common answer was some ethnic or linguistic identity (Nigeria — 48%, Namibia — 45%, Malawi — 38%, and South Africa — 22%). Yet, occupational or class identities were the most frequently cited responses in Lesotho (60%), Zambia (48%) and Zimbabwe (37%). Botswana stood out in bold relief from the rest of the study with almost one-third of the sample replying that they did not differentiate themselves from others. Thus, while communal identities still feature prominently, the surprising frequency with which Africans call on economic or class identities challenges many assumptions about the supposed ‘primordialism’ of African politics.

Democracy

Nowhere during the worldwide ‘third wave of democracy’ was the move away from military and civilian dictatorship toward competitive elections and multiparty systems more unexpected than in Africa. Yet, during the 1990s, at least 40 countries across the continent experimented with political liberalisation, with 22 eventually reaching a democratic transition through a successful founding election.⁹ However, many of these new democracies are extremely fragile and some have reverted to authoritarian rule. Even when leaders have implemented regular elections and secured the conditions for political competition, pluralism and the protection of human rights, democracy remains far from consolidated. Given the relative novelty of representative, electoral democracy in Africa, a large part of this halting progress may be attributed to the resistance or indifference of ordinary Africans, to whom democracy is seemingly a largely foreign or unknown concept that has been imposed by foreign donors and aid agencies.

What does democracy mean to the average African?

If democracy is a Western concept imposed on unwilling Africans, it could be expected that:

people would have a limited awareness of it;

they would hold negative views of it; and

they would have culturally distinct under-standings of the term.

Thus, people were asked: "what, if anything, do you understand by the word ‘democracy’? What comes to mind when you hear the word?" Although the questionnaire/interview was always conducted in the language of the respondent’s choice, the actual word ‘democracy’ was always presented in English.¹⁰ In attempting to capture specific understandings of democracy, respondents were free to offer answers in their own words. In order to avoid overlooking any distinctive interpretations Africans might attach to the concept, all answers were recorded verbatim and the responses were coded after the fact.

First, Africans are quite aware of democracy. In the nine countries, an average of 75% could supply a definition of the term.¹¹ In eight countries, the proportions ranged from nine out of ten South Africans and Malawians, to two-thirds of Namibians. Only Lesotho falls outside this range, with just over one-half of all Basotho able to offer an understanding of democracy. Thus, by no stretch of the imagination can democracy be described as a strange and incomprehensible project to these Africans. It is important, however, to note that, while large majorities were able to provide some reaction, far lower proportions offered multiple responses (people were allowed to supply up to three responses). Approximately one out of five could give a second definition and only around one out of 20 could give a third. This suggests that, while there is a widespread awareness and appreciation of the term, the depth of awareness or the complexity of understanding should not be overestimated. While certainly not an alien concept, popular understandings of democracy may be quite vague or ‘thin’.

Second, the Africans interviewed overwhelmingly attached positive understandings to democracy — nine out of every ten responses. No more than one out of ten respondents in any country surveyed saw democracy as in some way ‘bad’, with the remainder describing democracy in neutral terms. Third, conventional academic wisdom holds that, to the extent that Africans think of democracy, they do so in terms of economic advancement,¹² communal welfare,¹³ or the collective freedom of the post-colonial nation from foreign rule.¹⁴ However, far from seeing democracy as a panacea for poverty, domination or intercommunal conflict, the surveys revealed that an average of seven out of ten people described democracy in political and individual terms, usually as civil liberties or personal freedom, popular participation in government, or voting and elections. Only one out of every 20 respondents saw democracy in terms of social and economic development, and less than one half of 1% made reference to group rights.

Democracy: Is there any alternative?

If democracy was an alien, imposed concept, it could be expected that there would be little popular support for it. On the contrary, there is widespread popular support for democracy, both in idealised terms as ‘the best form of government’ and when confronted with more realistic trade-offs. In six of the nine countries, wide majorities ranging from eight out of ten (Botswana and Nigeria) to two-thirds (Malawi) said that "[d]emocracy is always preferable to any other form of government." Support was more tentative, however, in South Africa (60%) and Namibia (58%), and was exceptionally low in Lesotho where only four out of ten (39%) chose this statement.

Only in Malawi did a substantial proportion (22%) choose the statement that, "in some circumstances, a non-democratic government can be preferable" (increasing to 30% in Malawi’s Central Region, the homeland and political base of Dr Hastings Banda, the country’s former strongman). Almost one-quarter of Basotho (23%) said: "For someone like me, a democratic or non-democratic regime makes no difference." Another quarter (24%) of Basotho and one out of five Namibians (19%) said that they did not know the answer, figures four to five times higher than in other countries. Because this question was asked in a variety of new democracies, it could be determined that the African average of 63% support is almost identical to that found in 1995 in new democracies in six Central and East Europe countries (65%) and four Latin American countries (63%).¹⁵

The surveys also posed more realistic trade-offs. Noting that the "current system of governing with regular elections and more than one political party is not the only one" their country has had, surveys asked respondents whether they "would be better off" if their country switched to another form of government. Five different alternative regimes were offered with which Africans are highly familiar.¹⁶ Popular rejection of non-democratic rule is solid and consistent. There is little "authoritarian nostalgia" in any of these nine countries. An average of 80% rejected the option of military rule. Allowing only one party to stand candidates for election or hold office (the one-party state) was rejected by 72%. An average of 68% rejected the return to a system of rule by a council of elders, traditional leaders or chiefs, while 75% rejected abolishing parliament and parties so that the president can decide everything (strongman rule). Finally, an average of 71% rejected the option of reverting to whatever regime preceded the present multiparty regime.¹⁷

Rejection of non-democratic alternatives is consistently strongest in Zambia and Nigeria (with an average of 86% rejecting these alternatives). Of concern, however, is that the rejection is consistently weakest in Lesotho, South Africa and Namibia (with a much lower average of 63% rejecting these alternatives). Yet, it is important to note that, even in Lesotho, there is relatively strong resistance to any return to the past, nor is there any desire to try a range of other imaginable non-democratic alternatives. If democracy is not yet "the only game in town,"¹⁸ ordinary people already say they prefer it to a wide range of abstract and concrete alternatives.

Democratic performance

While there is a high level of ‘demand’ for democracy across almost all countries surveyed, perceptions of the delivery or ‘supply’ of democracy by the state are not uniformly as high. Africans appear to see democracy as a work in progress. By wide margins, people felt that their present multiparty regimes deliver far more political freedoms and rights (speech, association, voting, and from arbitrary arrest) than the ancien regime. This holds true regardless of whether the previous regime was an indigenous one-party rule, military dictatorship, racial white minority rule, or foreign colonial administration.¹⁹ However, Africans are far less certain that multiparty government has improved their economic quality of life (measured as access to basic necessities, standard of living, crime and violence, equal treatment by government, and social equality).²⁰ In most places, a narrow plurality said things had improved, but in Zambia and Zimbabwe most respondents felt that things had stayed the same or become worse. Along with other evidence in the surveys, this indicates that Africans, in contrast to the typical wisdom, are quite capable of reaching independent evaluations about politics and economics. When set beside the widespread support for democracy reviewed above, this also suggests that people will not abandon democracy easily or quickly simply because it fails to deliver the economic goods.

When asked to assess the freeness and fairness of their last election, assessments of progress are more reserved. Only in Botswana did more than a simple majority feel that their most recent election was completely democratic. It is true that once those are added who say the election was "free and fair with minor problems," majorities in eight of the nine countries saw their elections as basically democratic, ranging from eight out of ten Batswana and Namibians to just over 50% of Basotho. Significantly, there is a sharp split in opinion over the disputed 1998 election in Lesotho where one-third judged it to be "completely free and fair," but one in five felt it was not free and fair at all. Views of the electoral process were worst in Zimbabwe where less than one-third of the public said their 1996 election was free and fair.

When asked to assess the overall extent of democracy produced by their multiparty governments, there was nowhere a clear majority who felt their country was completely democratic. Again, once those are added who said it was largely democratic with minor problems, seven countries had majorities who said their country was largely democratic, ranging from eight out of ten Batswana to just one half of Nigerians. The situation was much worse in Lesotho where just over one-third of Basotho felt their country was wholly or largely democratic. Another 13% said there were major problems, and 17% said it was not a democracy. Equally problematic, just under one-third of the sample said they did not know how democratic the country was. Assessments were even more negative in Zimbabwe where just over one-quarter felt the country was largely democratic. Almost four out of ten stated flatly that the country was not a democracy.

Lastly, people were questioned about their overall satisfaction with the way democracy works in their country. Again, there was majority satisfaction in seven of the nine countries, ranging from a euphoric 84% of Nigerians (in a survey conducted shortly after the country’s return to civilian rule following the Abacha dictatorship) and three-quarters of Batswana. In contrast, just over half of the South African sample and less than four out of ten Basotho were satisfied with the way democracy worked in their country. In Lesotho, 27% said they could not answer the question, reflecting the tentative nature of the country’s current arrangement where an elected majority party government coexists uncomfortably with an appointed, multiparty interim political authority. Negative assessments were at their most extreme in the 1999 Zimbabwe survey where only 18% said they were satisfied with the operation of democracy. An equally large 17% refused to answer the question on its own terms and instead volunteered the response that "Zimbabwe is not a democracy," a quite unusual occurrence in the field of attitude research.

Given its recent dramatic changes and its importance on the continent, it is instructive to pause and focus on Nigerians’ assessments. Nine out of ten saw rapid improvements in political rights and liberties and 84% were satisfied with their young democracy’s early progress. Yet, only one-half felt their country was sufficiently democratic at the time of the January-February 2000 survey. Obviously, the survey caught a moment of enthusiasm and aspiration, yet one tempered with a sober assessment of how far the country still had to go.

When the demands for democracy are compared with the perceived supply of democracy, a range of different patterns become apparent. First, there appears to be a relatively satisfied equilibrium of demand and supply in five of the nine countries surveyed. In Botswana, and to some extent in Nigeria, this equilibrium occurred at high levels. In Malawi, Zambia and Ghana, this balance occurred at more modest levels.

Second, the surveys revealed an acquiescent disequilibrium in Namibia and, to a lesser extent, in South Africa. Namibians exhibited relatively weak demand for democracy, yet felt their country was governed democratically, and were relatively satisfied with the way democracy works in this one-party dominant system. It might be that the regime is providing more democracy than Namibians want. Or it might be that Namibians are willing to consume whatever output the dominant party produces that it chooses to call democracy. While the patterns are not as clearcut, South Africans’ support for democracy was modest and their rejection of non-democratic alternatives was the lowest among the nine countries in this study. Yet, significant majorities felt that their 1999 election was largely free and fair, and large majorities also felt that the introduction of non-racial democracy had brought advances in political rights.

Third, the September-October 1999 survey in Zimbabwe found a different and probably more ominous type of dissatisfied disequilibrium. Put simply, while Zimbabweans longed for democracy, they were adamant that they were not receiving it. This yawning gap between aspirations and reality foretold with great prescience the impending electoral fate of President Robert Mugabe’s draft constitution and of scores of previously safe ZANU-PF parliamentary seats. Short of a new survey there, it can only be speculated whether the subsequent defeat of the draft constitution in a popular referendum, followed closely by the meteoric rise of the opposition Movement for Democracy and its unprecedented success in the June 2000 election, have increased Zimbabwean’s perceived supply of democracy. Or has Mugabe’s refusal to enforce the rule of law and his party’s sponsorship of illegal farm invasions and electoral violence only further widened the breach between hopes and reality in this embattled country?

Finally, a pattern of apathetic low-level equilibrium is evident in Lesotho where minorities demanded democracy, and similar minorities felt the country is governed democratically. At the same time, it needs to be remembered that most Basotho rejected a range of non-democratic alternatives, and also thought they had experienced an impressive increase in political rights since the end of the military dictatorship. While Lesotho consistently ‘scores’ at the bottom or near the bottom in this study, the real problem does not appear to be a reservoir of authoritarian values, but rather confusion, or indifference in the face of the current interim arrangement.

Institutional performance

With regard to the institutions of democratic governance, public attitudes overall are more varied and less positive. Among the wide range of public evaluations of government performance that were employed, perhaps the most important finding is that — with one or two exceptions — these institutions have failed to cultivate a widespread level of public trust. A sense of trust in institutions is an important part of the legitimacy that can provide a cushion to buffer a young political system against the inevitable shocks of short-term dissatisfaction with government performance.²¹

Surveys asked people about their trust in a range of different institutions. The more partisan institutions of government (president/prime minister, parliament, local government) consistently enjoy far lower levels of popular trust than the more technocratic, bureaucratic institutions of the state (army, police, courts, public broadcaster, electoral commission and government press — where it exists). In sum, the median African tended not to trust elected institutions, but to trust appointed institutions. This suggests that governments in most of these countries will continue to have a difficult time obtaining public co-operation, especially where compliance requires voluntary support.

It is noteworthy that, while presidents or prime ministers tended to enjoy higher levels of trust than parliaments or local governments, the ‘big man’ factor seemingly so prevalent in African politics can hardly be detected, at least in the public mind. Only in Namibia did a large majority (more than seven out of 10) say they trusted the president (about the same levels of trust in the president recorded in a 1998 Idasa survey in South Africa). The highest level registered anywhere else was in Malawi where 50% expressed trust in the president. In South Africa, trust in the president plunged by over 30% in two years, signalling the end of the honeymoon of Thabo Mbeki (Nelson Mandela’s successor). At its most extreme, only 19% of Zimbabweans said they trusted the president.

It is also worth noting that, along with the public broadcaster, the army tends to enjoy the highest levels of trust among state institutions in six of the nine countries. Only in the two countries emerging from military rule, Lesotho and Nigeria, and in South Africa where the military’s reputation was sullied by its role in enforcing the states of emergency in the 1980s, was the army trusted by less than a majority of citizens. Yet, as shown above, large majorities rejected any idea of abandoning democracy for military dictatorship, indicating clearly that Africans trust the military as soldiers, but not as politicians.²²

Institutional trust not only varies considerably according to the specific institution, it also varies across countries. In general, Botswana and Namibia’s institutions enjoyed high levels of citizen confidence (in the vicinity of 60% to 70% for most institutions). At the other extreme, trust was the lowest in Zimbabwe and Lesotho with only one-quarter to one-third expressing trust in most institutions. In contrast to the typical monochrome view of governance in Africa, there was substantial variation in public confidence in government. Where many analysts see the problems of ‘Africa’ as a single piece (perhaps best epitomised by the recent, now infamous, issue of The Economist, entitled ‘Africa: The hopeless continent’), the surveys revealed an extremely variegated view of African politics across many different areas.

Questions probing perceptions of official corruption underscored the need to draw finer grained distinctions when depicting government in Africa. Corruption is widely seen by analysts as endemic to African government. Thus, it is no surprise that two-thirds of Zimbabweans, and one-half of Zambians, Malawians and South Africans thought that all or most government officials were involved in corruption. Yet, only one-fifth of Namibians held such a view about their government. But these are perceptions — what about actual experience with official corruption? Conforming to the stereotypical view of African government, an average of one out of ten Zimbabweans said they had been asked for a bribe or a favour in order to get government assistance in housing, land, employment or basic services. However, in Botswana, where international air visitors encounter anti-corruption posters even before they get past customs, only one out of 100 citizens said they have encountered such demands from government officials.

Economic liberalisation

If economic liberalisation in Africa is to succeed, its principal tenets must at least not be sharply inconsistent with the values of the principal agent of economic transactions: the people. While citizens understand democracy and support the process of democratisation, can the same be said about economic liberalisation?

Awareness of adjustment

A wide variation was found in the extent of popular awareness of structural adjustment programmes (SAPs) in those countries that had such a programme. At the lowest extreme, only 13% of South Africans said they had ever heard anything about their government’s self-imposed package of stabilisation and adjustment policies called the Growth, Employment and Redistribution Plan (GEAR). This will come as a major surprise to many South Africans, especially those in the labour movement who have spent the last three years waging a political battle to reverse this package. Structural adjustment programmes enjoyed a higher, though limited public awareness among four out of ten in Zambia, Ghana and Nigeria and one-half of the public in Malawi. It was far more widely recognised in Zimbabwe where 85% said they had heard of it.

Some economic policy makers will probably read these as disappointing results. After two decades of sustained international efforts to induce African governments to reorient their economies toward the market, large proportions of intended beneficiaries claim to be ignorant that such a strategy even exists. On the other hand, policy makers in unstable, emerging democracies might view these findings as a blessing. These programmes usually require an undemocratic limitation on economic policy debate and choice.²³ Since they are usually negotiated under international pressure and behind closed doors, a widespread awareness of SAPs may undercut a government’s legitimacy. Thus, governments may find it to their advantage to conceal the concessions on economic policy they have made to international investors or bankers and to undertake ‘adjustment by stealth’, a strategy apparently adopted by the government of South Africa.

Perhaps the variation in popular awareness of SAPs might be a function of the host government’s rhetorical hostility to them. While the South African package of home-grown policies has attracted the ire of trade unions and the country’s not insignificant Communist Party, the package lies at the heart of the government’s strategy to achieve economic growth and create jobs. Even though the Nigerian package was introduced in 1986 by the government of Ibrahim Babangida with extensive public debate, it always claimed that its reforms were indigenous. In contrast, Zimbabwe’s ESAP (Economic Structural Adjustment Programme) has entered the public discourse in 1990 and has been a favourite bogeyman for the Mugabe government, which has openly claimed that the policy has been imposed on it by Washington-based international financial institutions. Another part of the explanation might be the shifting and cryptic labels attached to adjustment programmes, such as the change from ERP to SAP in Ghana (with a detour into a social safety net programme called PAMSCAD). While it was not asked, it is also probable that most South Africans are under the impression that the government’s key economic policy is still the state interventionist Reconstruction and Development Programme (RDP) on which it publicly based its 1994 election campaign and to which it still pays frequent lip service. The lack of mass familiarity may also reflect that some African governments have only partially adopted market reforms and implemented them inconsistently.

Even if SAP terminology is not broadly recognisable, citizens may still know what sectoral policies are supposed to achieve. In Nigeria and Ghana, surveys asked a specific question: "What, in your opinion, is the structural adjustment programme supposed to do?" Only one-third of respondents in either country could venture an answer, standing in marked contrast to the three-quarters of citizens in these countries who could attach a meaning to democracy.

Economic values: Market versus state

Even if hazy about the exact content of adjustment policies, people still hold basic economic values. Where do African citizens position themselves in the great debate about states and markets? Do they see themselves as autonomous economic actors, or do they rely on government as the chief provider of public well-being? Respondents were asked to make a choice on a standard survey item that is used in value surveys worldwide: either "A. People should be responsible for their own success in life," or "B. Government should bear the main responsibility for ensuring the well being of people." This item neatly summarises citizens’ orientation to economic values generally. By this criterion, Africans were fairly evenly divided with a crossnational average of just 51% supporting individual responsibility and 46% who thought government should provide well-being. These figures are similar to responses in seven former Soviet Republics in 1989 to a similar question where 51% favoured government provision and 49% emphasised individual responsibility.²⁴ Beneath these averages, however, lie sharp crossnational differences. Support for individual responsibility was the highest in Malawi (73%) and the lowest in Nigeria (42%) and Zimbabwe (37%) where clear majorities saw the government as responsible for public welfare.

The same type of division in values is evident in responses to a similar item. Overall, 50% felt that "the best way to create jobs is to encourage people to start their own businesses," while an average of 47% in each country agreed that "the government should provide employment for everyone who wants to work." Again, Malawians leaned most toward the market (72%) and Zimbabweans the furthest from it (45%).

While Africans may be split on basic economic values, their views about the division of public-private responsibility in various sectors reveal a persistent statism. In almost all sectors of the countries surveyed, people expressed doubts about the ability of private enterprise to replace public provision. This was particularly true in relation to the distribution of investments in socio-economic development. For each sector, respondents were asked whether the government, individuals, businesses, or some combination of these, should have primary responsibility. Across all countries, majorities felt that government had the primary responsibility for providing agricultural credit (75%) and building schools and clinics (73%). In each country, the largest proportion saw a leading government role in job creation (61%).

While there was greater crossnational variation with regard to marketing key export commodities, in every country except Malawi the largest proportion of people favoured a leading role for government (averaging 63%).²⁵ In every country except Botswana and Namibia, majorities wanted the government to take the lead in controlling crime (an average of 60%).²⁶ Only with regard to building houses did most people consistently prefer the private sector or some private-public partnership (government primacy is favoured by an average of 40%). The exception is South Africa where 63% saw the government as having the primary responsibility, possibly reflecting the ANC’s highly publicised plan to build a million houses in its first years of rule to redress the country’s massive housing backlog.

Given the abject failures of state marketing boards across the continent, it may be surprising that more Africans have not embraced private sector solutions in these sectors. Only small minorities thought that market-based institutions should be responsible for helping to market export commodities, extend agricultural credit, or build schools and clinics. Thus, while respondents might know that the state lacks administrative capacity, they considered the private sector to be even less reliable. As such, at least in the nine countries participating in the surveys, Africans remain ambiguous at best about the viability of markets.

Support for adjustment policies

This does not mean that the debate about structural adjustment in Africa should be cast in black and white terms. Without the benefit of much empirical evidence, commentators often assert that Africans are either ‘for’ or ‘against’ (usually against) adjustment. But structural adjustment is a complex package of reforms that citizens do not have to accept or reject in wholesale fashion. In fact, people agree with some aspects of liberalisation. The typical structural adjustment package was broken down into four constituent policy elements that ordinary people would tend to encounter. Across all countries, more people said they would rather pay user fees "to raise educational/health care standards" (59%) than have free services "if it meant the quality of services is low" (39% overall). A lower, but still majority percentage indicated a willingness to pay higher prices in order to have a greater availability of goods on the market (50%) rather than supporting low prices at the risk of shortages (39%). Malawi and Zimbabwe stood out as exceptions where more people preferred lower prices even at the risk of shortages.

This unexpected balance of opinion in favour of adjustment shifted sharply with regard to public service retrenchment and the privatisation of public corporations. In seven of the nine countries, more people wanted the government to retain ownership of key industries (a crossnational average of 56%) rather than sell them to the private sector (36%). In eight of the nine countries, more people said that all public officials should keep their jobs even if the salary bill damages the economy (63%), than that there were too many public employees and that some should therefore be laid off (29%). The strongest pro-state sentiment was found in Botswana and Malawi, countries with small populations and long-standing traditions of public employment on which many extended family networks depend. Interestingly, support for retrenchment was the highest in Zimbabwe (51%) and South Africa (43%). The Zimbabwean responses may be a comment on the bloated nature of Mugabe’s party-state machine. In South Africa, public service retrenchment is one of the most contentious issues dividing the government from its trade union supporters, but the results suggest some level of public awareness of the need to reduce the size of the bureaucracy inherited from the apartheid system.

To summarise, the Africans participating in these surveys displayed bifurcated attitudes toward structural adjustment. Far from being wholly for or against economic reform, most Africans expressed discriminating views. In some respects, they supported market liberalisation, willing to pay school and clinic fees in order to maintain service standards, and to some extent, higher market prices. Against this willingness to ‘get the prices right’, the evidence indicates a strong popular resistance to reforming state economic structures. There remains strong attachment to the state supply of development goods and services. The key issue appears to be the persistent role of the state as a principal provider of employment in Africa. Even if public sector salaries are declining, the rents and perquisites associated with public service are apparently still attractive. The responses suggested that people did not believe that private sectors in their countries could deliver a commensurate supply of adequately paying jobs. Until the free market can demonstrate that it will perform better at job creation than the state, Africans apparently prefer to stick with the devil they know.

Conclusions

The nine countries covered in this study would appear to have progressed a good part of the way toward a successful ‘triple transition’. They seem to have moulded coherent nations who owe their loyalties to the political communities where they live. The artificial and arbitrary boundaries of the past have become the source of real, meaningful and widely shared national identities. As the struggles to democratise the state and to consolidate young electoral democracies continue, two important facts stand out. First, there is an almost consensual popular agreement on the identity of ‘the people’ who should be governed by these various states. Second, there is a more limited but still widespread agreement that they should be governed democratically. There is little nostalgia for going backward, or trying any other alternative form of regime.

But it is equally clear that, with one or two exceptions, these states have a considerable distance to move to satisfy popular demands for a fully democratic state in which they can place confidence to govern in the public interest. To the extent that Africa’s democratic and market experiments remain unfinished, delays in the consolidation of new political and economic regimes can be attributed more to enduring autocratic and rentseeking tendencies among leaders and the incapacity of the state, than to the intransigence or apathy of citizens.

The most pressing needs facing reformers, whether in Africa’s civil societies or in international donor agencies, are twofold. The first is to build capacity in state institutions to enable them to respond to public opinion and govern effectively. The second is to continue to support greater pluralism of interests and more diverse centres of power in the form of effective political parties, civil society organisations and independent media able to hold the leadership accountable and check autocratic and rentseeking tendencies.

Finally, while international donors and certain segments within the governments of these nine sub-Saharan states may be convinced that democracy and free markets go hand-in-hand, the people remain less than convinced. They do not totally oppose the objects of adjustment, nor do they totally accept them. But until they can see the evidence that free markets can do better than the state in ensuring a decent standard of living, economic liberalisation will proceed with a far smaller mass constituency than that accompanying political democratisation.

Notes

P Norris, Introduction: The growth of critical citizens?, in P Norris (ed), Critical citizens: Global support for democratic governance, Oxford University Press, Oxford, 1999, p 5.



The sample sizes for each country are as follows: Botswana: 1 200, Ghana: 2 004, Lesotho: 1 177, Malawi: 1 208, Namibia: 1 183, Nigeria: 3 603, South Africa: 2 200, Zambia: 1 200, and Zimbabwe: 1 200. The margin of sampling error is plus or minus 3% (2% in Nigeria, and 2.2% in South Africa). Fieldwork was conducted by national research institutions affiliated with the Afrobarometer project. The authors are grateful for research funding from the National Science Foundation and the United States Agency for International Development. The surveys in Botswana, Lesotho, Malawi, Namibia, South Africa, Zambia and Zimbabwe also comprise the Southern African Democracy Barometer, a project within the larger Afrobarometer supported by USAID’s Regional Centre for Southern Africa.



Samples were designed using a common, multistage, stratified, area cluster approach. Random selection methods were used at each stage, with probability proportional to population size where appropriate. Sampling frames were constructed in the first stages from the most recent census figures or projections available, and thereafter from census maps, systematic walk patterns, and project-generated lists of household members. With the exception of South Africa, each country sample was self-weighted and sufficiently representative of national characteristics on key socio-economic indicators (gender, age, region) that post-weighting was not necessary.



J Linz & A Stepan, Toward consolidated democracies, Journal of Democracy 7, April 1996; L Diamond, Is the third wave over? Journal of Democracy 7, July 1996.



Linz and Stepan, ibid.



E Gellner, Nations and nationalism, Basil Blackwell, Oxford, 1983; D Rustow, Transition to democracy: Toward a dynamic model, Comparative Politics 2, 1970.



See A Lijphart, Democracy in plural societies: A comparative exploration, Yale, New Haven, 1977; W Connor, Ethnonationalism and political instability, in J Gagiano & H Giliomee (eds), The elusive search for peace: Israel, South Africa and Northern Ireland, Oxford University Press, Cape Town, 1990.



Connor, op cit.



M Bratton & N van de Walle, Democratic experiments in Africa, Cambridge University Press, New York, 1997.



The exceptions were Namibia where Oshivambo interviews used the recognised word oDemocracy and Botswana where researchers felt that people were more likely to be familiar with the Setswana phrase describing democracy.



Reported crossnational averages are calculated as the raw mean of aggregate country distributions. This has the effect of weighing each country equally regardless of sample size or actual population size.



C Ake, Democracy and development in Africa, Brookings Institution, Washington DC, 1996, p 139.



P Chabal & J P Daloz, Africa works: Disorder as political instrument, Indiana University Press, Bloomington, 1999, p 130,



C B MacPherson, The real world of democracy, Oxford University Press, Oxford, 1966.



New democracies barometer IV (1995) cited in W Mishler & R Rose, Five years after the fall: Trajectories in support for democracy in post-communist Europe, Studies in Public Policy 298, 1998, p 13; Latinobarometro (1995) cited in J Linz & A Stepan, Problems of democratic transition and consolidation: Southern Europe, South America, and post-communist Europe, Johns Hopkins University Press, Baltimore, 1996, p 222.



This approach is inspired by R Rose, W Mishler & C Haerpfer, Democracy and its alternatives: Understanding post-communist societies, Johns Hopkins University Press, Baltimore, 1998, chapter 2.



Rule by traditional leaders was not asked in Nigeria, the option about strongman dictatorship was not asked in Ghana, and the item on returning to the previous regime was not asked in Ghana and Nigeria.



Linz & Stepan, Problems of democratic transition and consolidation, op cit.



This scale was not asked in Ghana. The item on freedom from arbitrary arrest was not asked in Nigeria.



This scale was not asked in Ghana. Only the items on equal treatment by government and standard of living were asked in Nigeria.



D Easton, A systems analysis of political life, University of Chicago Press, Chicago, 1965, p 273.



Christie Keulder is thanked for this point.



G O’Donnell, Do economists really know best?, in L Diamond & M Plattner (eds), The global resurgence of democracy, Johns Hopkins University Press, Baltimore, 1999.



A Finifter & E Mickiewicz, Redefining the political system of the USSR: Mass support for political change, American Political Science Review 86(4), December 1992, pp 857-874.



Oil in Nigeria, cocoa in Ghana, diamonds in Namibia and Lesotho, tobacco in Botswana, Malawi and Zimbabwe, gold in South Africa, and copper in Zambia.



Item not asked in Nigeria.

ROBERT MATTES is the manager of the Public Opinion Service of the Institute for Democracy in South Africa (Idasa).

MICHAEL BRATTON is a professor in the Department of Political Science at Michigan State University.