ANGOLA'S POST-WAR RECONSTRUCTION

In order to understand the momentous transformations of the last eight years, we must first focus on the trajectory of the Angolan state until 2002. At independence in 1975, all major institutions of the colonial state imploded amidst civil war, the exodus of white settlers, and a flurry of foreign invasions. There was no institutional continuity in the civil service, courts, or the armed forces; bar the oil industry, all sectors of the fairly diversified late colonial economy perished immediately or were soon suffocated by socialist planning. The Eastern Bloc-supported MPLA, the precariously prevalent force in the three-way contest between liberation movements, lacked the human resources to tackle complex governance tasks. The implosion of the state apparatus was partly addressed by importing thousands of Cuban and East European technical advisers.Footnote ² But South African-supported rebel activity destabilised the rural world and limited the writ of the Angolan state. The progress of the UNITA insurgency during the 1980s confirmed the state's neglect, and solidified the MPLA society into an ‘archipelago of cities’ often only connected by air, which by 2002 constituted 80% of traffic to the interior (EIU 2008: 14). The MPLA state's dependence on the Western-controlled coastal oil industry, as well as the party's urban bias, also contributed towards the sidelining of the countryside by policy makers. From 1975 to 2002, the government of Angola was unable to hold most of its territory.

The 1991–2 UN-supervised peace process culminated in closely fought presidential and legislative elections, in which the contenders put across mutually exclusionary visions for Angola.Footnote ³ But these collapsed into Angola's deadliest round of warfare when UNITA leader Jonas Savimbi refused to accept defeat. The rebels occupied or destabilised an estimated 80% of the country, including its second city Huambo, and laid siege to other cities in the central highlands. Although initially caught off guard by UNITA's overrunning of the country, the MPLA leadership caught up, in large part due to the accretion of oil revenues to the state far in excess of the diamond revenues available to Savimbi.

In the shadow of the oil industry and the war's long ‘state of exception’, a new political economy came of age, characterised by strong presidential control and the prominent role of a coterie of advisers; the privatisation of state property to their benefit; and complex international networks of support that laundered the wealth of insiders while contributing through the sale of weaponry to the sustenance of the government's war effort (see Billon Reference Billon2001; Hodges Reference Hodges2004). A powerful, immensely wealthy and internationally well-connected oligarchy became a prominent feature of the MPLA society. During this period the state abandoned the welfare commitments it had imperfectly upheld until the end of the Cold War. The lingering provision of social services now emanated from presidential foundations in the guise of charity (Messiant Reference Messiant2001), and most day-to-day food distribution for internally displaced populations was managed by the UN. By the end of the war, the MPLA presided over what could be termed a successful failed state (Soares de Oliveira Reference Soares de Oliveira2007a). Angola's population was amongst the world's most deprived; the state was incapable of performing a host of sovereign functions, and largely uninterested in the governance of the hinterland and the administration of its people. Angola was the antithesis of the ambitious African state of the early post-independence years (a template the MPLA had never matched in practice but formally paid allegiance to). At the same time, the MPLA state had built the institutional nucleus for survival and prosperity in the form of a strong presidency, a capable national oil company, efficient armed forces and urban coercive apparatus (see Soares de Oliveira Reference Soares de Oliveira2007a, Reference Soares de Oliveira2007b). The rump state that the MPLA developed in the 1990s, its political economy and the people who put it together, must be understood as its character determined the brand of apparently very different post-war reconstruction analysed below (Berdal & Keen Reference Berdal, Keen, Mayall and Oliveira2011).

In the last years of the war, the MPLA leveraged its position as de jure government, growing counter-insurgency success, and UNITA's loss of foreign (especially US and South African) support, to win the international struggle for recognition. Given that the history of the conflict is now being actively manufactured by a victorious MPLA, this means that the first decades of independence are increasingly presented as the story of a ‘national’ state confronted by a parochial insurgency. Reality is far more complicated. The MPLA's legitimacy was low in many parts of the country throughout the war, and the party was assumed by many to be in the hands of a small, unrepresentative clique of mixed race and detribalised coastal Angolans. Especially after the late 1980s, the government was missing from the lives of many Angolans, save for the vicious role of marauding army units. Moreover, UNITA's claims to ‘stateness’ and investment in statist rituals, its real control of a small percentage of the population, and its popularity amongst others created something approaching parity between the two forces (Pearce Reference Pearce, Okumo and Ikelegbe2010: 368), with the MPLA state less than a real state, and the UNITA insurgency more than an insurgency. In this sense, 2002 was a foundational moment for the MPLA itself: the moment in which one of the country's competing political forces vanquished the last major challenge (UNITA having been preceded by the FNLA and the Portuguese colonialists) in its bid for the control of post-colonial Angola.

Moving to the reconstruction years, there are three dimensions to the government's post-war agenda that I discuss in turn. These are the creation of a durable political order based on MPLA hegemony; the political economy of reconstruction; and the party's long-term ‘development vision’ for the country. The following pages make no claim to an exhaustive portrait of political developments in the past eight years, but merely provide a general characterisation of the government's approach.

A HEGEMONIC PROJECT

Having won the civil war, the MPLA wanted to remain the hegemonic political force in Angola for the next generation. MPLA leaders did not have a static understanding of their society, and immediately realised that many aspects of wartime political and economic life would have to be reconfigured. The imperative was and remains to reconstruct and modernise in a way that would preserve the status quo. In the short-term, this was accomplished by the political neutering of UNITA. Many of its cadres were co-opted into the new dispensation with cosy, if politically irrelevant, sinecures. A general amnesty on war crimes was declared, with little consideration given to other forms of reckoning with the past such as a truth and reconciliation commission, but resulting in a quicker assimilation of the vanquished.Footnote ⁴ UNITA itself never recovered from military defeat, and dependence on the state purse for post-war activities limited its actions further.Footnote ⁵ The easing of UNITA into the MPLA world is only the most evident example of a decision to co-opt potential challengers and redistribute some political and material opportunities to buy acquiescence from fractious constituencies.

Army generals, for instance, were essentially given a free hand in large swathes of the interior to engage in resource extraction and land grabs (Péclard Reference Péclard2008: 14). The Angolan government proceeded with its own version of disarmament, demobilisation and reintegration (DDR), consisting in the disarmament and demobilisation of UNITA forces and the integration into the Forças Armadas de Angola (FAA) of only a small number of UNITA soldiers.Footnote ⁶ The FAA itself stayed at close to wartime levels at 110,000 men throughout the rest of the decade, and the same applies to the paramilitary rapid-reaction police forces, which have remained constant at 10,000 men. Defence expenditure actually increased 2·5 times between 2005 and 2009, at first sight a dubious allocation of resources (IISS 2010). The maintenance of one of the largest and most competent armies in sub-Saharan Africa is explained by the fact that it had become central to the exercise of power in Angola and the broader region. Its disbanding was no longer an option.

In regard to civil society, the government understood that a measure of toleration was part and parcel of its ambition to foster an electoral suffrage system where the MPLA would reign supreme, as well as giving it an improved international standing. As the reconstruction years proceeded (and especially in 2007 and 2008), civil society actors were allowed a modicum of leeway. The growth of print media outlets (and from late 2008, TV Zimbo) is an important development in this regard even if their quality is often disappointing.Footnote ⁷ However, the salient characteristic of this period is not the MPLA's acceptance of dissenting views, but rather its consistent penetration of civil society organisations and professional associations, often dependent on the largesse of presidential foundations, achieving de facto political control while extolling their pluralism (Messiant Reference Messiant2001; see also Messiant Reference Messiant, Chabal and Vidal2007: 114). The same applies to the media, which are often owned or otherwise controlled by regime notables. Progressive vocabularies verging from ‘decentralisation’ to ‘transparency’ have likewise been selectively appropriated by the government with a corresponding record of muddled and compromised implementation at best.

The broader goal of the MPLA, as the late Christine Messiant (Reference Messiant, Chabal and Vidal2007: 121) aptly put it, was to create ‘an authoritarian hegemonic dispensation adapted to multiparty electoral politics’. While engaging with a limited version of democracy that does not tamper with the established mode of exercising power (mirrored in the increasing pluralism within the MPLA itself), the MPLA has deepened the ‘confusion between party, state and administration’ (Péclard Reference Péclard2008: 9, 13) in manners both symbolic and material.Footnote ⁸ The regime's landslide electoral victory in 2008 −82% of the votes in a peaceful election deemed fair by observersFootnote ⁹ – was followed by a new, markedly presidentialist constitution in early 2010, according to which the president will be indirectly elected by the National Assembly. This allows President dos Santos a further decade in power.Footnote ¹⁰

In this context, overt coercion is rarer than weary acceptance or even support for the MPLA, with party membership having increased from about 60,000 to 4 million between 1990 and 2004, in a country of some 16 million people (Vidal Reference Vidal, Chabal and Vidal2007: 145).Footnote ¹¹ Despite lingering hostility and/or ambivalence, the MPLA has never been as popular. Its current status as a catch-all party with nationwide support (if with an unreconstructed sociological core) is not a remainder from a past when the party had at most a plurality of supporters in some areas of the country. It is a concoction of the post-war years. The results of the long-delayed 2008 legislative elections showed that Angolans are exhausted after four decades of conflict and keen on predictability in their lives. This willingness to countenance MPLA domination is underpinned by the occasions in which the government does not hesitate to deploy the means of coercion. In addition to the unresolved Cabinda conflict, the armed forces and the police have been used in the violent deportation of tens of thousands of illegal migrants, mostly from the DRC (IRIN 6.10.2009). These instances allow the government to demonstrate its ruthlessness when challenged.

THE POLITICAL ECONOMY OF RECONSTRUCTION

The extent of the country's destruction in 2002 cannot be overstated (see UNSA 2002). Large cities such as Huambo and Kuito had practically been obliterated. More than 300 bridges were destroyed, the three major railways were unusable, and the damaged electricity system intermittently covered only a small part of the country (World Bank 2005: 2–3). An estimated 10 million landmines scarred the landscape. There were 4 million internally displaced persons and 450,000 refugees in neighbouring countries. More than a million people were dependent on food aid. Even in the few areas that had not seen much combat, such as Luanda, the state of disrepair of the mostly colonial-era infrastructure was considerable. In response to this, the Angolan government has a Long-Term (2025) Development Plan. Although never fully disclosed,Footnote ¹² the agenda is known to be premised on a ‘top-down, accelerated high-technology economic development, with heavy emphasis on investment, big projects and borrowing to build infrastructure. It is a vision of master planning resting on a premise that the state can solve the nation's ills’ (Vines et al. Reference Vines, Shaxson and Rimli2005).

The major tenet of the reconstruction process was the rebuilding of infrastructure. This consumed the greater part of available resources. Infrastructure, especially roads and bridges, was essential to jumpstart Angola's non-oil economy. There was also a political imperative to break out of the coast and the cities, and effectively occupy the national territory and set up a civil administration in the hinterland. The government's sensibility and the population's needs were aligned on this early emphasis, which resulted in unambiguous successes such as the rebuilding of much of the coastal and highlands road networks. But general reconstruction expenditure (as well as that for the 2010 Africa Cup which Angola hosted) has taken place through questionable procurement processes and in the absence of budgetary oversight. Quality control was non-existent, and construction companies appointed on account of their insider status rather than on the basis of merit or cost effectiveness. Less than four years after delivery, shoddy quality marred prestige projects such as the Chinese-built Luanda General Hospital (Jeune Afrique 6.7.2010). Reconstruction has been exceedingly expensive and failed to deliver value for money. A 2007 study of Angolan public expenditure warned of the danger of promoting poorly prioritised and unsustainable ‘white elephant’ projects to the detriment of more pressing needs, and recommended increased expenditure in education, health and agriculture (World Bank et al. 2007: 5–6).

The government professed a commitment to the human element of reconstruction, and indeed raised expectations of service delivery in the run-up to the elections. However, average Angolans have not been the target of much government expenditure since 2002. Some of this owes to the limitations of Angola's human resources: the state simply does not have the manpower to cope with the expansion of statist tasks implied in the ambitious reconstruction blueprint. Incompetence, and not just corruption, is an important part of the story. That said, many of the policies pursued by the government seem designed to maximise rentierism and minimise oversight. After the earlier emphasis on reconstruction of transport infrastructure, the recent years have seen the government commit to a flashier and mostly needless building spree, including the project to redesign Luanda's main bayside avenue, a new airport, football stadiums, and assorted government buildings in Luanda Sul (Power Reference Power2011). Regardless of its overt justifications, the infrastructure boom helps justify ‘a patronage-based political system which operates to a large degree through government spending projects’ (Vines et al. Reference Vines, Shaxson and Rimli2005: 7).

This links to a pivotal dimension in the government's reconstruction agenda: the overtly political manner in which the state apparatus is used to provide insiders with opportunities for accumulation of vast fortunes, a process set in motion in the 1990s but decidedly deepened since 2002. In this regard, the government's promotion of the Angolanisation of the private sector through participation in foreign investment has nurtured an Angolan business class.Footnote ¹³ The president's family and assorted insiders have benefited hugely from this process (Villalobos Reference Villalobos2009; Africa Confidential 4.7.2008). Despite a language of entrepreneurship, enrichment remains dependent on access to political power. While genuine entrepreneurs exist, the bulk of those favoured are silent partners in foreign joint ventures who simply collect their share without getting involved in the management side of things. Arguably, the incentive structures of Angola's oligarchic capitalism work against the fostering of genuine entrepreneurship. President dos Santos' recent speeches against corruption may point to a limited willingness to shore up the system and prevent the disorderly leakage of the early reconstruction period.Footnote ¹⁴ But the government's commitment to strengthening institutions, human capital, the judicial and regulatory systems and the quality of the bureaucracy, all central for a better prosecution of the reconstruction effort, remains in doubt (UCAN 2010: 22).

The macro-economic story of these years is remarkable, with 15·5% yearly average growth between 2002 and 2008, or eight times the average for the period between 1980 and 2002 (2·1%), and significant taming of inflation (UCAN 2009: 8; see Table 1). Starting in 2006, the growth of the non-oil economy surpassed that of the oil economy. But its growth was premised on the availability and circulation of oil rents (Klein Reference Klein2010: 19). This dependence on oil, despite the government's stated goal to diversify the Angolan economy, was revealed by the 2008–9 financial crisis and consequent drop in oil prices: a 32·4% slump in oil revenues led to a 75·1% drop in public investment, a sharp decline in hard currency reserves, and serious payment delays to contractors, especially in the construction sector (UCAN 2010: 16). By mid 2010, the macro-economic situation had improved. But the limitations of the government's plan to transform the country through public expenditure without reforming the public administration, curbing corruption, diversifying the economy or improving education and other social indicators had been exposed.

Table 1 The Angolan economy, 2002–8

Source: ‘Crude oil including lease condensate (thousand barrels per day)’, US Energy Information Administration: Independent Statistics and Analysis, available at: http://www.eia.doe.gov/cfapps/ipdbproject/iedindex3.cfm?tid=5&pid=57&aid=1&cid=AO,&syid=2002&eyid=2008&unit=TBPD, accessed 4.2.2011; ‘GDP growth (annual %)’, The World Bank: Data, available at: http://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?page=1, accessed 4.2.2011.

THE MPLA'S DEVELOPMENT VISION

In a 2006 address, President dos Santos noted that ‘it is necessary that the population feel that there are constant improvements in medical assistance, education, commerce, transportation, water and electricity supply and in the respect for their rights’ (MAT 2004: 7). With the end of hostilities, previous shortcomings were retrospectively dismissed as resulting from the war, and the MPLA resurrected a vocabulary of state-building and social concern for average Angolans. Far from being objectionable, much of its public discourse is premised on broadly shared ambitions for the country. In policy terms, however, the key concern of the ruling party, at least in the first instance, is the creation of what it calls a ‘national bourgeoisie’ rather than the alleviation of mass poverty. There is an articulate minority within the MPLA that views this matter differently, but it does not have traction over policy. In the dominant agenda (which, incidentally, is an improvement on the micro-sociological sphere of concern of the war years), ‘there is little room for the poor, who are often seen as an obstacle to, rather than the primary focus of, development’ (Vines et al. Reference Vines, Shaxson and Rimli2005: 6). The World Bank (2006: 9) noted that ‘public delivery of social services is … skewed in favour of the urban rich’, but the same applies to most public expenditure. While recent data does not allow for authoritative judgements, there is a tacit consensus among observers that the war victims and the poor have not benefited proportionately from the economic growth and that poverty reduction is not part of the post-war story.Footnote ¹⁵

Beyond the important specificities of its historical trajectory, the MPLA's development vision can be understood in terms of a ‘high modernist’ mindset (Scott Reference Scott1998) fired up by the resources, and imaginings of endless possibilities, of an oil-rich economy, with which students of oil boom-era follies of the 1970s will be familiar (see Karl Reference Karl1997; Soares de Oliveira Reference Soares de Oliveira2007a). Some analysts have presented this development vision as outdated and wrong-headed but ‘sincere’, in the context of the prevalent MPLA ideas about the state, especially in view of its ‘leftist’ roots. Yet this is to underestimate how easily the ‘oil nomenklatura’ (Ennes Ferreira Reference Ennes Ferreira1995) adapted to what James Ferguson (Reference Ferguson2006: 39) called a ‘nongovernmental state’ during the 1990s. There is something contrived about this high modernism. The Angolan elite, like the oil sheiks of old, are keener on the end products of modernity than on its processes. There is no sustained attempt to build the management systems of the complex technology that is the modern state. There is an explicit nouveau riche assumption that expertise (just like luxury consumer goods) can be bought, if tempered by a long-term desire that enough Angolans may one day master some of these enviable skills. This Angolan-style modernist agenda prioritises the things the elite want: shopping malls, skyscrapers, private condominiums, marinas and other ‘visible badges of being an important oil state’ (author interview, July 2009; see Power Reference Power2011). Despite recurrent references to agricultural development,Footnote ¹⁶ and belated incentives for private investment in non-oil sectors and the provinces, public expenditure remains overwhelmingly focused on Luanda and, to a lesser extent, the four or five major urban clusters.

Some contemporary experiences speak to the sensibility of the Angolan elite as they pursue their reconstruction drive. As with other states of the developing world, there is a superficial elective affinity with the Chinese political-economic model which provides an example of a reform process that strengthens, rather than weakens, the status quo. Interviews with politicians of a certain age with memories of Maoist China bring out their baffled admiration for the Chinese accomplishment. But engagement with the actual Chinese experience of economic reform and social transformation is shallow. Little by way of detail is known or understood, and few if any specific economic lessons are taken on board, perhaps because the Angolan elite believe that, rhetoric aside, these are of limited relevance for a rentier state with scarce human resources. The same applies to references to Luanda as ‘Africa's Dubai’ in the media, and in private and public statements by politicians. The Angolan elite (like, say, Venezuelan elites in 1970s Miami) are enamoured with the modernity of Dubai's built environment, and the speed with which it was achieved, rather than seriously engaged with the political economy of that city. There is no willingness to emulate it. While superficial, this fascination has serious consequences, with glittery modern buildings mushrooming all over Luanda, and scarce electricity supplies being rerouted to satiate their nightly consumption.

Finally, impacting on the MPLA's agenda is the only major period of conventional Angolan stateness: the late colonial modernising set-up that collapsed in 1975.Footnote ¹⁷ Its physical debris and promise of modernity were ubiquitous throughout the subsequent quarter-century during which Angolans inhabited an architectural time warp. Amongst members of the MPLA elite, an insidious ‘1973 nostalgia’ took root, to the extent that its reconstruction agenda now involves putting back together features (many obsolete) of pre-independence Angola. But it is at the level of discourse that this impact is clearest. Ambitious, indeed megalomaniac, settler imaginings of Angola as a ‘new Brazil’ have become central to the aspirations of the inheritors of the state, if now mixed with a brash oil-fuelled nationalism. The dominant strands of the MPLA differ from the elites of the defeated liberation movements, who were mostly mission-educated and only marginally related to the colonial state and Portuguese culture. The MPLA attracted the mestiço and detribalised urban elites who were intensely ‘portugalised’ (Messiant Reference Messiant2008: 40), held deep memories of pre-scramble domination of the coastal enclaves, and were often employed in the lower and middle ranks of the colonial civil service. Confusingly, this social group spawned the MPLA while many of its members were beneficiaries of the late colonial attempt to bring Angolans into the fold of Portuguese rule.Footnote ¹⁸ While it is an exaggeration to say that they felt any ownership with regard to the transformation of this period, they were its beneficiaries. It is not surprising that those high-modernist dreams for Angola would speak to them as well, and resurface at a later stage.

FOREIGN RELATIONS AND THE ROLE OF FOREIGNERS IN RECONSTRUCTION

It was not always clear that Angola's transition from war to peace would take place with a modest degree of external influence. The UN's engagement with Angolan conflict resolution started with a small verification mission following the New York accords of December 1988. After the 1991 Bicesse accords another mission (UNAVEM II) was tasked with ensuring that the parties carried out the agreed responsibilities. This was a notoriously poorly staffed and under-resourced mission with an unclear mandate to boot.Footnote ¹⁹ There was no external appetite for an intrusive mission, and little sign (other than the emphasis on elections) of the liberal ambitions that informed other UN deployments, while the warring parties wanted to calibrate its involvement at an innocuous level. The reassertion of UN authority during the 1993–4 Lusaka negotiations and the better resourced follow-up, UNAVEM III, seemed to point to a more conventional role (MacQueen Reference MacQueen1998: 405). But the government's 1998 decision to pursue a ‘war for peace’, i.e. to discard negotiations and strive for a military victory, meant that the UN was no longer needed as a mediator. Of course, the UN, and especially the UNSC, remained important as a forum for consolidating the criminalisation of UNITA and guaranteeing that other states would not come to its aid. But in Angola itself, the MPLA wanted a UN presence that did not obstruct the war strategy.

The UN's political sidelining did not mean irrelevance on the ground. Its operational commitments grew exponentially to 7,000 men and continued during the final descent into war, especially in regard to food and health provision and the needs of the millions of internally displaced persons. MPLA hostility was tempered by the benefits of a major UN system welfare commitment to the poor and needy (the same applied to churches and humanitarian NGOs), in a context where it had neither the means nor the willingness to play that role. In a pattern that continued in the post-war period, the government sought to calibrate this external involvement in a manner that was unthreatening to its political agenda and indeed assisted in its implementation.

When the war ended in early 2002, the Angolan government's objective vis-à-vis the UN and Western donors was thus to maximise their contribution towards reconstruction while subordinating these resources to its political agenda. Angola's leaders, knowing well that they could not put the country back together without a major foreign input, called for a donors' conference to help them mobilise resources. However, the major Western donors were less than forthcoming. Concerns about egregious siphoning off of oil revenues and myriad insider deals had plagued international perceptions of Angola in the last years of the war. High-profile campaigns by organisations such as Global Witness (2002, 2004), Human Rights Watch (2004) and the Open Society Institute, among others, shed light on corrupt practices in the oil sector and self-enrichment by the Angolan elite. The international mainstreaming of activist agendas on transparency and the role of the extractive industries in armed conflict further dented the Angolan government's reputation (Benner & Soares de Oliveira Reference Benner, Soares de Oliveira, Witte and Goldthau2010). Perhaps more importantly, the IMF (2002) had become an implacable critic of the opaqueness of Angola's oil accounts, and its leaked documents pointed towards the routine disappearance of billions of dollars. Finally, there was scepticism in some quarters that an oil-rich country such as Angola should need a major external financial commitment towards reconstruction, with the resident UN coordinator calling on the government to share ‘a greater part of the burden’ (see Africa Confidential 14.6.2002, 11.10.2002).

It is important not to overstate the extent to which Western donors took these normative concerns on board (Soares de Oliveira Reference Soares de Oliveira2007a; Benner & Soares de Oliveira Reference Benner, Soares de Oliveira, Witte and Goldthau2010). Even at the 2002–4 height of international criticism of oil sector corruption in Angola, the enthusiasm of foreign oil investors, as well as their home governments, never wavered. Angola's woes remained a niche concern in the mainstream media. Yet there was enough momentary unease with the quality of Angola's governance to create a quasi-consensus amongst prominent Western states and the Bretton Woods institutions that a donors' conference should be postponed until the Angolan government could explain the whereabouts of missing oil revenues and commit itself to implementing some of the reforms put forward by the IMF. Potential donors also wanted to know that the Angolan government intended to spend its considerable resources in a developmental direction rather than expect foreigners alone to pay the reconstruction bill.

The Angolan reaction to this refusal has been described by prominent decision-makers in terms of ‘shock’, ‘huge pain’, and a ‘sense of betrayal’.Footnote ²⁰ From an Angolan perspective, the proposed reformist package was perceived as ‘economically risky and politically suicidal’ (Traub Reference Traub2006). The sovereignty-conscious Angolan leadership was unlikely to concede when faced with perceived international bullying. Many of the measures on the table were inimical to the sort of grip over the system, and especially oil revenue flows, that decision-makers wanted to retain at that stage. The Western emphasis on conditionality risked cancelling out the discretionary power over resources and the orientation of post-war reconstruction. Unsurprisingly, the donors' conference was postponed on numerous occasions from 2002 to 2004. In 2004, during a visit to the USA, President dos Santos apparently raised the issue with President George W. Bush, to no effect. In fact, the donors' conference would never take place and, by early 2005, it was the Angolan president himself who postponed its convening indefinitely (EIU 2004: 11).

What had changed in the meantime was the appearance of a major international partner, China, willing to deal with the Angolan government with none of the preconditions set forth by Western donors (see Ennes Ferreira Reference Ennes Ferreira, Alden, Large and Soares de Oliveira2008; Power Reference Power2011; for the broader context see Alden Reference Alden2007; Alden et al. Reference Alden, Large and Soares de Oliveira2008; Brautigam Reference Brautigam2009; Taylor Reference Taylor2009). In March 2004, the Chinese government extended credit lines to Angola, and very soon Chinese companies staffed by tens of thousands of imported labourers were deeply involved in the country, ranging from the oil sector to the reconstruction of infrastructure. By 2009, Chinese loans to Angola amounted to at least US$13.4 billion (according to some estimates, US$19·7 billion, see Vines et al. Reference Vines, Weimer and Campos2009: 5), and bilateral trade had grown thirty-five times in the eight years up to 2008 (EIU 2009: 19). Many of these deals were signed with the Hong Kong-based Chinese International Fund, an ostensibly private, opaque vehicle for Chinese investment. The Angolan end of the Chinese relationship was jealously controlled by the presidency, which created an Office for National Reconstruction (GRN, headed by insider General ‘Kopelipa’, presidential head of military affairs and one of country's most powerful men) to administer and disburse the Chinese monies.Footnote ²¹ Some experts point to factors other than the Chinese connection that better explain the increasing leeway of the Angolan government as the decade progressed, with emphasis on the increase in oil production (from 1 to almost 2 million bpd) and the exponential rise of the oil price throughout this brief period (Downs Reference Downs2007). These points are valid, but nor can one underestimate the extent to which the Chinese credit mattered in 2004, or the symbolic role of the Chinese arrival in the broader transformation of Angolan external relations.

For Angola's China partnership was part of a wider diversification of its portfolio of international relations away from over-dependence on the major Western powers, which were also the home countries of Angola's key investors. The Angolan government did this by pursuing three strategies. The first was the cultivation of privileged bilateral relations with states with which it previously had tenuous relations (e.g. India, South Korea) and, in some cases, the reactivation of old rapports (Russia, Cuba). The second was to join OPEC, the cartel of major oil-producing states. At the time, this decision was contested by some in the Angolan elite because it was seen as detracting from Angola's scope for deciding oil production levels. The presidential aspiration was for Angola to be part of an important group of states with clout in the international economy, thus placing itself above the sort of paternalistic intrusions the West reserves for ‘failed’ or ‘weak’ states, and with plenty of opportunities for statesmanship and summitry into the bargain.Footnote ²² The wider Angolan interest in multilateral organisations can be seen in this context.

A third and final aspect of the Angolan government's post-war international relations was an attempt to increase its external respectability after an earlier moment of bellicose refusal to engage in reforms. In order to diminish the clout of external critics, Angolan leaders have accepted a renewed engagement with the IMF, a partial convergence with international reform demands, especially in the area of oil sector transparency (see Hansen-Shino & Soares de Oliveira Reference Hansen-Shino and Soares de Oliveira2011; Isaksen et al. Reference Isaksen, Amundsen, Wiig and Abreu2007), and, culminating in the 2008 legislative elections, a degree of liberalisation of the public sphere. As already mentioned, this is consistent with the elite's desire to build a hegemonic yet internationally acceptable dispensation. Elements of a liberal agenda (and of its rhetoric in particular), if drained of their progressive implications, can be willingly deployed as tools in the consolidation of the regime's external credibility.

The upshot of this diversification strategy was a quick improvement in Angola's international standing. From the perspective of Western states, this translated into a newly placatory approach to Angola's governance issues such as corruption and dismal social indicators, and an emphasis on partly successful elements of Angolan reconstruction such as DDR, the rebuilding of infrastructure, and the 2008 elections. Conversely, the momentary alignment of Western foreign policies with NGO agendas deteriorated rapidly. Foreign visitors in recent years have flattered Angola's rulers and reconstruction drive, and paid next to no attention to previously pressing concerns, as exemplified by Pope Benedict XVI and US Secretary of State Hillary Clinton's visits in 2009. Western diplomats are now often critical of NGO ‘destructive criticism towards the Angolan government’, and some express marked antipathy vis-à-vis leading Western NGOs active on Angolan issues.Footnote ²³ Angolan NGOs have complained about a reduction in Western commitment towards formerly ‘hot’ issues in the country.Footnote ²⁴ Foreign diplomats speak of a ‘nuanced approach’ to Angola and admit that ‘a confrontational strategy in this country [will] get you nowhere’.Footnote ²⁵ This soothing posture can be partly justified by real improvements in Angola. But serious problems remain, there is often no discernible movement towards addressing them, and Western diplomats currently go to great pains to focus on ‘non-political objectives’Footnote ²⁶ and disregard touchier matters. Western foreign policies towards Angola have instead converged with the Asian (and also Portuguese and Brazilian) emphasis on business opportunities and no meddling in internal affairs.

FOREIGNERS IN RECONSTRUCTION

Angola's accent on national sovereignty and distrust of external involvement, in itself typical of many post-colonial states, is heightened by a history of foreign intervention and the more recent ‘normative’ undermining of the Angolan government's reputation by anti-corruption campaigners. Yet the government has no isolationist intent. As discussed above, there is a clear attempt to normalise and diversify Angola's external relations, and create privileged connections with as many useful partners as possible. The same applies to the role of foreigners at the domestic level. Angolan decision-makers understand that the country does not have the management systems and technical capacity to reconstruct its infrastructure or run the more elaborate post-war economy. As a result, Angolan reconstruction contains as much international input as in states with multilateral, UN-directed efforts. However, there are crucial differences. This external involvement occurs within a political context exclusively defined by the Angolan government; the key foreign role is played by the private sector and state-owned corporations rather than by international organisations or NGOs; and their involvement is premised on non-negotiable financial rewards to members of the elite who are unavoidable partners in foreign entrepreneurial activities. The Angolan government has thus tapped into external tools for reconstruction, but on its own terms.

The Angolan economy has for decades presupposed a key role for non-Angolans, as there are not enough skilled Angolans to man the state and the economy at the requisite level. The exodus of 350,000 Portuguese settlers in 1975 destroyed the non-oil economy. The functioning of the post-independence state was in turn only made possible by the presence of Eastern Bloc advisers. After the Cold War, these were replaced by a large number of foreign workers, UN agencies, NGOs and church charities. This scarcity of human resources also applies to the oil industry, which remains dependent on expatriate technical expertise. Other things that worked satisfactorily in the last years of the war did so because they were under management contracts with foreign firms. This was the case of the water supply in Soyo and Caxito, solid waste collection and port operations in Luanda, and ground handling services at Luanda airport (World Bank 2005: 4).

The advent of peace resulted in an explosion in demand for foreign labour in all areas, varying from reconstruction proper to the manifold services sector in the major cities. The key difference between the post-1991, and especially post-2002, period and previous decades is that this external role had previously always circumscribed Angola's margin of political manoeuvre. More recently, however, this external presence is orchestrated, and its political impact circumscribed, by Angolan elites. The Portuguese in particular have returned to Angola in great numbers, reaching an estimated 80,000 in 2008 (UCAN 2009), in tandem with the growing importance of post-war Angola for the Portuguese economy.Footnote ²⁷ While some are Angola-born, many are simply looking for business or employment opportunities, as are 25,000-odd Brazilian citizens. The number of Chinese workers in the country is the subject of considerable debate, with 50,000–75,000 a credible figure (UCAN 2009; Vines et al. Reference Vines, Weimer and Campos2009: 42). Smaller but visible presences include Malians, Lebanese, Indians and Spaniards, as well as assorted northern Europeans, North Americans, Israelis and South Africans.Footnote ²⁸

Foreigners are spoken of, and refer to themselves as, ‘service providers’ to the authorities, implementing whatever they are asked to do and benefiting from whatever opportunities come their way. Their business activity is subjected to overt or covert partnerships with Angolan insiders, without which they cannot enter the Angolan market or thrive in it. While essential for the reconstruction effort, the foreign private sector is not collectively organised. Opportunities for individual companies depend on the discretionary power of a handful of decision-makers, and work visas for expatriates are the subject of a complex and arbitrary regime resulting in a precarious presence on the ground. This makes foreigners both crucial to the reconstruction effort and powerless, marginal individuals without much political clout.Footnote ²⁹ Even high-profile players such as the major Brazilian and Portuguese construction companies, which are particularly close to the presidential palace, are better thought of as resilient courtiers than as consequential actors in their own right.

This marginality also applies to international organisations and foreign NGOs. This is partly due to their meagre financial resources when compared with the revenue stream of the Angolan government. But the key factor is the government's strategy of strictly defining the contours of any external entity's role in Angola and guaranteeing that it advances its agenda. The government's thinking regarding foreign development partners is therefore in continuity with its policy towards the foreign private sector. Donors, IOs and a number of NGOs have obliged the authorities, stressing their obsequiousness towards the government's agenda.Footnote ³⁰ A vital element of this relationship is that the internationals should not be seen to cast doubts over the government's reconstruction storyline.Footnote ³¹ As an international official remarked,

The result of having to play a ‘constructive role that does not embarrass the authorities’ (author interview, Luanda, June 2009) is a considerable abridgement of the scope for critical involvement. It is also in marked contrast to the 2002 UN blueprint for Angola's reconstruction, a sort of ‘Peacebuilding 101’ which did not shy away from criticising ‘the deeper institutional problems concerning the nature of governance in Angola’ (UNSA 2002: vii). Despite a still-important presence in Angola, then, the foreign development agencies have played a minimal role in defining the reconstruction agenda or even influencing the government's own plans.

ASSESSING THE ANGOLAN EXPERIENCE

Six key factors made possible the Angolan approach to reconstruction. The first is the outright military victory of the MPLA. The second is the material basis for autonomous decision-making and aloofness from external pressures: Angola's oil endowment clearly puts it in a different category from other poor and aid-dependent states. Third, the leadership of the MPLA is cohesive, pragmatic and highly competent in managing the international politics of the reconstruction period. The party itself is far from homogeneous, but its historical divisions are circumscribed by the power of the presidency, and lack the salience of previous decades. The fourth factor is the lack of a well-organised and adequately funded internal constituency that articulates an alternative liberal agenda. The MPLA's dominance is compounded by a historically weak, and always harassed, civil society, which places little pressure on the government in this regard. Fifth, there is the growing role of international partners that eschew conditionality in their relations with other states, and are willing to put material resources and political support at the service of bilateral relations with post-conflict states. The case of China highlighted above needs to be put in a broader context where non-OECD states, including democracies such as Brazil, South Africa and India, routinely dissent from the Western emphasis on liberal reconstruction. These alliances have played a crucial role in preserving Angolan domestic autonomy. Finally, there is the lack of commitment by Western donors to their own purported liberal agendas, which are frequently negotiated away when set against political or economic matters deemed more pressing.

This paper started with a broad characterisation of Angola's reconstruction drive from 2002 to 2008. I mostly did not address post-2008 developments or the overall likelihood of success. A return to war in Angola is not possible. Yet uncertainty remains about Angola's trajectory despite two key accomplishments, namely the creation of political stability, and the rebuilding of infrastructure and resulting partial reintegration of the national territory. As shown by the 2009 slump, the continuation of this partly virtuous course will be a function of the uninterrupted flow of oil revenues. Improvements also depend on the ‘vicissitudes of internal transformation’, and whether Angola can go much further with an unreconstructed state apparatus and insufficiently public-spirited elite.Footnote ³² Forged by a traumatic experience of war, fuelled by ample resources, and masterminded by the country's leading politico-military actors, this is a state-building drive firmly rooted in Angola's historical trajectory. Angola is picking itself back into a shape of sorts, with its institutional and political economy legacies (Soares de Oliveira 2007a) providing both a repertory for action and limits to what is politically possible.

TOWARDS AN UNDERSTANDING OF ILLIBERAL PEACEBUILDING

International peacebuilding missions, although not all alike and evolving across time, share a number of liberal peace assumptions that place electoral democracy and the free market at the forefront of reconstruction efforts. Their goal is not the restoration of the status quo ante, but the building of a society and polity on (more) liberal lines. This commits international peacebuilders to a diversity of policies, ranging from the promotion of the rule of law and a free press to incentivising privatisation and coming to terms with war crimes. A large literature examines the normative and practical shortcomings of liberal peacebuilding (see for instance Barnett et al. Reference Barnett, Kim, O'Donnell and Sitea2007). Despite their flaws, internationally accepted discourses on reconstruction, and the very real financial and sometimes coercive pressure exercised by external parties, have conditioned the boundaries of what is politically acceptable in many instances of post-war reconstruction. In cases such as Sierra Leone, Mozambique, Namibia, Liberia and the DRC, to name but a few, the centrality of this external agenda is undeniable, however poorly implemented in practice.

Angola provides an example of a substantially different process of reconstruction, where local elites define the sort of post-war society they want to build. This and other instances of illiberal peacebuilding remain improperly theorised and are rarely placed in a comparative framework. Yet there are many cases of post-war reconstruction with no (or easily evaded) international supervision and/or conditionality, and scant regard for liberal precepts. Rafik Hariri's Lebanon rebuilt itself under heavy-handed Syrian suzerainty (Perthes Reference Perthes1997), while Iranian money flows since the 2006 war favour the Hezbollah parallel state. The Sri Lankan government's crushing of the Tamil Tigers resulted in the subordination of the defeated minority.Footnote ³³ Eritrea's bellicose attitude towards outsiders has given it considerable leeway in defining policy. Even states initially constrained by Western donors have managed to dilute external influence to the point of insignificance. Hu Sen's Cambodia evaded the shackles of the once-pervasive international presence. In post-1994 Rwanda, a tightly organised RPF has been able to use Western guilt over the genocide, together with Paul Kagame's one-time reputation as a reformist, to pursue an authoritarian agenda (Ansoms Reference Ansoms2009). The point of this brief reference to other experiences is not to suggest that they are alike: they present wide variations in anything from developmental ambition to the level of state repression. But they share important features, and are amenable to a comparative approach rarely attempted thus far, despite the availability of vibrant and high-quality case-study literatures.

In addition to the previously mentioned gaps in the study of autonomous recovery, there is much equivocation in the existing literature on the normative character of these exercises. Not infrequently, contemporary domestic state-building is superficially explored, less for the sake of understanding it in its own terms, and more in order to bash liberal peacebuilding and exogenous state-building. A degree of soul-searching vis-à-vis Western-led interventions is certainly necessary in view of a disappointing record (see Mayall & Soares de Oliveira Reference Mayall and Soares de Oliveira2011). But much of what is written on the subject romanticises autonomous recovery and overstates its rootedness, while equivocating about the character (both normative and in terms of developmental potential) of the political projects at the heart of it. Unsurprisingly much of this literature focuses on micro- or medium-level processes such as grass-roots deliberative democracy, traditional justice and clan/tribal gatherings, and mostly leaves out central-state level questions. This ignores the fact that most consequential processes of autonomous recovery are state-centric and macro-level. Such an approach also neglects the dysfunctions at the heart of these political systems that led to war in the first place. It posits ‘the indigenous’ as somehow inherently legitimate in societies which have, after all, faced serious and as yet unresolved crises of legitimacy.

More importantly, a romantic understanding of autonomous recovery misstates the extent to which the political imaginations and reconstruction blueprints of victorious forces are, more often than not, high-modernist, patronage-based and illiberal. The cases mentioned above can seldom be understood in terms of a ‘traditional’ or ‘authentic’ agenda for reconstruction.Footnote ³⁴ The reconstruction agendas of the MPLA, the RPF and the Sri Lankan government are instead urban, elite-controlled processes focused on the limitation of popular mobilisation, the strengthening of the state apparatus, and the construction of a stable but non-egalitarian political order.Footnote ³⁵ When decision-makers think of a future for their countries, there are examples they seek to emulate, but these are not found in the workings of liberal societies (or, for that matter, in traditional, consensus-based ‘local practices’). The post-war establishment may bring improvements to the population, but typically the peace dividend flows primarily to the powerful and wealthy rather than to war victims and the poor masses.

The following is a list of recurrent characteristics of illiberal peacebuilding that will be useful for pursuing a comparative study of these processes: (1) military victory or hegemonic post-war oversight; (2) hegemonic election-running, designed to earn international support yet not representing a danger to the regime; (3) secretive formal or informal structures for running the reconstruction process;Footnote ³⁶ (4) reconstruction opportunities distributed among insiders and promotion of an oligarchic capitalism; (5) constitutional change to extend presidential powers and strengthen the status quo; (6) penetration/co-optation of civil society organisations (especially professional organisations) and the media; (7) acceptance of the situation by Western donors who carry on business as usual while deploying a vulgarised version of the ‘transition paradigm’ to explain illiberal practices; (8) a high-modernist vision and technocratic mindset, with much public expenditure on infrastructure and heavy borrowing; (9) a general amnesty, no ‘justice and reconciliation’, or else clearly manipulated victor's justice; (10) a peace dividend that favours the powerful, while poverty reduction is not a priority.

In reality, there are no perfectly illiberal reconstruction processes.Footnote ³⁷ Many of these states have done it ‘their way’, but have also engaged piecemeal with liberal agendas such as partial democratisation, ‘market reform’, dealing with war crimes, women's rights, and so forth. Especially when dealing with specific issues, governments are able to give foreigners the impression that they ‘share a [common] understanding of state failure and reconstruction’ which can be translated into a like-minded policy agenda (Englebert & Tull Reference Englebert and Tull2008: 119). They thus skilfully cherry-pick political and material resources available internationally (West/East, IOs, private sector, NGOs, etc.) and deploy as many as possible on their own terms. Lip service to some liberal themes may be useful internally to deal with rivals, and enough to let these states off the hook on other liberal concerns. But ultimately these political projects are at odds with liberal peacebuilding.

There are obvious merits to a domestic reconstruction strategy when compared with liberal peacebuilding: there is clear ‘local ownership’; there are no concerns about ‘exit strategies’; there is no risk of a decline in political commitment with the passing of time. The fabric of such efforts is made of really existing political actors of consequence with really existing political agendas, rather than the often inconsequential people liberal outsiders would like to see enact their ‘great post-conflict makeover fantasy’ (Cramer Reference Cramer2006: 245ff). The last two decades show that externally imposed conditionality rarely works, while international peacebuilding has struggled to create a sustainable peace, let alone ‘development’, however defined. Alternatively, reconstruction drives such as the one studied in this article do result in a form of state-building and political order.

Support for these efforts is therefore defensible – but only if their character is owned up to, and one is explicit about the trade-offs. During the Cold War, a long line of political scientists presented political order as the a priori imperative for sustainable development (see Huntington Reference Huntington1968), and were therefore sympathetic towards the lack of pluralism in Latin American and East Asian states. They emphatically argued that all good things don't go together, and that liberal arrangements were luxuries that poor and post-conflict societies could ill-afford. On the contrary, current-day apologists for locally managed reconstruction are rarely open about the often insalubrious and exclusionary content of the associated state-building projects, and the fact that the building of a ‘decent society’ (Margalit Reference Margalit1998) is not at their heart. Scholars are understandably fascinated by the ‘locally instigated’ and ‘largely popularly supported political structures’ of Somaliland (perhaps an example of a domestically initiated process that is genuinely inclusive, see Walls Reference Walls2009: 373–86), but elsewhere have ended up praising ‘local solutions’ merely on account of their locality.Footnote ³⁸ In an otherwise insightful contribution, Jeremy Weinstein (Reference Weinstein2005: 5, his emphasis) defines ‘autonomous recovery’ as a process whereby states ‘achieve a lasting peace, a systematic reduction in violence and post-war political and economic development in the absence of international intervention’. There is little mention of the fact that such processes of autonomous recovery seldom encompass individual freedom or a commitment to broad-based economic growth.

In sum, domestically led reconstruction contains important pitfalls and is problematic in its own right, especially in terms of its frequently illiberal character. Local ownership, and foreign subsidiarity to it, is a means to an end, rather than intrinsically praiseworthy regardless of the purpose it serves. Serious comparative analysis of illiberal reconstruction should thus beware of simplistic positive contrasts with international peacebuilding. Studied without illusions, illiberal peacebuilding is, to paraphrase Christopher Cramer's (Reference Cramer2006) understanding of state-building more generally, an essentially ‘tragic process’ whereby brutal or predatory processes may counter-intuitively participate in social transformation and the consolidation of institutions. It includes the possibility that, in some contexts, ‘progress’ is the reinvention of inequality rather than genuine emancipation.