Wilful entanglements: Extractive industries and the co-production of sovereignty in Mozambique

Mining in Africa: Eroding soils and sovereignty?

The idea that ‘neoliberal’, globalised capitalism is undermining or eroding state sovereignty, eventually rendering sovereign states irrelevant and incapable of regulating capital and labour flows (Comaroff and Comaroff, 2001: 28, e.g.) appears to remain a compelling narrative. In this reading, the formation of enclaves at the heart of power leads to the ‘discharge’ of the state’s sovereign functions that results in the exhaustion of the territorial model (Mbembe, 1999: 112).

Extractive industries are often held up as a case in point. Whereas mining was socially ‘thick’ in the 1950s and 1960s (Ferguson, 1999), extractive industries in Africa are today enclaved and of little benefit to the country and society at large: oil is ‘socially thin’, and mining ventures increasingly work like offshore oil, walling themselves off from the surrounding social fabric. Capital jumps globally from point to discrete point, bypassing the nation-state altogether (Ferguson, 2006: 378–80). Such enclaved (Mbembe, 1999), parcellised (Mbembe, 2000), or fragmented (Ong, 2006) sovereignty has historical precedents in the colonial-era concession areas and chartered companies (Hansen and Stepputat, 2006: 308; Hönke, 2012: 111) and is experiencing a revival today.

While benefits to societies have indeed often been limited, much of the literature on African states under the conditions of globalisation has ‘overstated the case that global economic (and other) forces have crippled the state [… as] many international and transnational forces have propped it up more than they have sabotaged it’ (Migdal and Schlichte, 2005: 9). Neoliberal enterprise needs functioning regulatory frameworks and has strengthened states (Comaroff and Comaroff, 2001: 30) – although such state-making is as much an international project as a national one, ‘denationalized’ (Sassen, 2006), and oriented towards the facilitation of investment and business activity (Bracking, 2016: 21; Chalfin, 2010: 39; Mosse, 2005: 4). Rather than seeing sovereignty regimes necessarily eroding under the impact of ‘globalization’, core state functions are redefined as Africa is constantly being respaced (Engel, 2009; Engel and Nugent, 2010).

Despite these conceptual advances, the idea of ‘fragmented’ or ‘parcellised’ sovereignty often appears to be based on a notion of sovereignty as a finite whole, a sort of zero-sum game whereby any takeover of governance functions by corporate actors reduces state sovereignty. This is problematic, especially in Africa, because it reinforces the idea of weak or failing states outsourcing governance functions to companies, charities, or NGOs. Deploying the vocabulary of outsourcing against a backdrop of neoliberalism is seductive because it resonates with our own political sensitivities. A state-centric view on sovereignty has also obscured how actors commonly attributed to different spheres and scales, spanning foreign and national governments, international aid agencies and corporate actors, jointly produce practices usually subsumed under the functions of state sovereignty: service provision, the setting of fiscal and regulatory standards (Peters et al., 2009), revenue generation, and the physical control of people and space. I am less concerned with the de facto wielding of power over life, death, and the wellbeing of its subjects, an Agambenian take on sovereignty central to debates about state violence and the subject body (Cooper-Knock, 2017, e.g. Hansen and Stepputat, 2006), but rather the flip-side to such states of exception, the establishment of the normative and the normal: the claims to authority, control, and the extension of regulation and jurisdiction over a given collectivity (Comaroff and Comaroff, 2006: 35). A processual take on shared sovereignty elucidates how these diverse players – some with inscribed sovereignty, some with de facto, situational power to influence its production – shape and reconfigure the state.

This is not a radically new approach: there is a body of scholarship that recognises – also specifically for the case of Mozambique – how the entanglements of INGOs and state actors in the health and education sectors (McKay, 2016) or nature conservancy (Lunstrum, 2013; Symons, 2016) co-produce new forms of ‘articulated sovereignty’:

Rather than an abstract concept that a state clearly possesses or lacks, [sovereignty] is better understood as a set of attributes, competencies, and powers that are actively and routinely produced through a series of unequal interactions and negotiations with other actors, including other state and extra-state actors. (Lunstrum, 2013: 2)

What I seek to do here is to extend a processual take on the state, its imaginaries, and its sovereign functions to the interactions around the development of the extractives sector. Rather than lamenting the growing irrelevance of the state, the crumbling of sovereignty, or the effacement of political logics against economic logics under the onslaught of global corporate forces, it is more interesting to chart which kind of state emerges (Magnon-pujo, 2011: 3) from the invocations and implementations of development through the efficient management of the extractives sector.

Development through extractives?

The mid-2000s commodity boom, fuelled by heightened demand from China and India, has been one of the key drivers of resource extraction-dependent African economic development over the past decade (Taylor, 2016: 16). Mozambique, a donor darling and testing ground for neoliberal reforms since the civil war ended in 1992 (Hanlon, 2000; Pitcher, 2002), has for the past few years been anticipating a coal and liquefied natural gas (LNG) boom that would transform the country from a donor-dependent ‘emerging country’ into an ‘energy giant’ and ‘African Lion’ economy. The government hopes to turn Mozambique into a leading coal exporter; ‘discoveries’ of natural gas reserves in the Rovuma Basin estimated to be the world’s third largest contiguous field further created a climate of bonanza, with major international corporations jostling to secure exploration rights. These developments have taken place in the context of the opening up of a new extractive frontier across East Africa, from South Sudan to Mozambique. This was the result of sustained high commodity prices and high demand for natural resources, c. 2004–14 (Hendrix and Noland, 2014: 3), coupled with technological advancements, and, not least, advice from the World Bank and the United Nations Economic Commission for Africa, UNECA (Lange, 2011: 233–4). In fact, the Rovuma gas reserves had been discovered by the Portuguese colonial regime, and been confirmed by Soviet tests in the 1970s and 1980s; however, only more recent progress made their exploration commercially viable. ⁴

With the downturn in world market prices for coal and hydrocarbons since 2014, many ambitious projects have been put on hold. This led to an economic crisis in Mozambique, compounded by recent political and security developments, which in themselves were in large part triggered by the expectations of a natural resource boom. ⁵ Still, as I elaborate below, state, corporations, and civil society actors are in broad agreement that the state has to be made ready for the next, ‘inevitable’ cyclical upturn in commodity prices, and improve its ‘weak’ capacities to manage the nascent extractive sectors in an efficient, market-friendly, and transparent manner. The question of how to develop the sector best is not only technical but political, depending on delicate negotiations between unequal actors, who all advance their own interests in these negotiations: The Mozambican state and private sector lack the technical expertise to develop these economic activities independently, and depend on foreign investors to come in and kick-start the actual production. Investors, in return for their large-scale investments, expect a stable business environment and fiscal concessions to mitigate risks to their ventures and increase their return rates. Opposition parties demand a share of promised revenues, while civil society associations are trying to influence the process, backed by international advocacy groups.

In advancing their positions in these negotiations, these varied actors are subject to policy pressures and the coercive power of (international) institutions, but they also tap into and manipulate competing and overlapping dominant normative frameworks. These include the call for transparency in natural resource extraction and good governance of public finances, the efficiency of markets and the private sector opposed to a public sector scripted as, at best a handmaiden, at worst an obstacle to free capital flows, and the promise of accelerated development through the impact of extractives (encapsulated, for example, in the Africa Mining Vision 2020, promoted by UNECA). Charting how various actors deploy these political imaginaries can yield new insights into the ideas and practices of contemporary African states (cf. Schubert et al., 2018).

A corollary question that emerges is if and if so how far Mozambique is actually neoliberal. James Ferguson argues that neoliberalism in Africa is less the ‘Thatcher-Reagan assaults on the […] post-war welfare state [with] the deployment of market mechanisms and “enterprise” models to reform government, restructure the state, and pioneer new modes of government and subjectification’ (2010: 172), but rather the ‘dismantling of the state, privatising of assets under Structural Adjustment Programmes (SAPs) – [and] much less so the development of new technologies of government’ (p. 173). Based on the material presented here, I would nuance this position, as I seek to demonstrate how ideas of new, efficient governance modes do find resonance among state actors, even if this is coupled with a specifically Mozambican, post-socialist, take on the ‘good management’ of the sector for the ‘common good’ of national development. With regard to the scholarship on Mozambique, the article thus also picks up a discussion initiated by Anne Pitcher, who suggested that at a moment when, ‘a kind of “savage capitalism” has taken root [in Mozambique], it seems imperative that scholars evaluate the memory and forgetting of socialist principles in the context of present struggles’ (Pitcher, 2006: 90).

Historic entanglements

The province of Tete, the epicentre of today’s coal ‘boom’, was long a sleepy backwater along the Zambezi valley, wedged between Zimbabwe, Zambia, and Malawi. The provincial capital, Tete, served the Portuguese colonial regime as trading post and departure point for goldfields further inland (Kirshner and Power, 2015: 71), but the first large-scale development project in the province was the construction of the Cahora Bassa dam in the 1960s, as part of Portugal’s effort to develop and modernise its ‘overseas provinces’ and stave off demands for independence (Isaacman and Isaacman, 2013).

Coal deposits in Tete have been known since the 1850s and mined irregularly since the 1930s (Kirshner and Power, 2015: 71), but industrial mining only took off after Mozambique’s independence in 1975. When the newly independent Mozambican state, under the leadership of the then socialist vanguard party Frelimo, set out to create a ‘New Man’, freed from the shackles of tribalism, obscurantism, and superstition (cf. Sumich, 2015: 9), the country’s socialist brother states were ready to help, including the German Democratic Republic, with its ‘technical-scientific’ and ‘technical-cultural’ cooperation (Obernhummer, 2010: 9). Military cooperation had started during the liberation struggle, but technical assistance became central to develop the country’s infrastructure and skill base (decimated after the rushed withdrawal of the Portuguese administration and deliberate sabotage by the leaving colonists). This intensified after Frelimo declared Mozambique a socialist country in 1977.

Following trade agreements between Mozambique and the GDR in 1977 (Obernhummer, 2010: 17), the cornerstone of this cooperation was the education of thousands of Mozambicans in East Germany, many of whom continued to work in the GDR as contract workers after their studies. The Magermanes, as they are known, rally in Maputo every week to demand their pension funds, which vanished after being transferred to the Mozambican Ministry of Labour during the dissolution of the GDR (Allina, 2016; Schenck, 2016).

There were also many East German technical experts, cooperantes, who went to Mozambique to work there as lecturers and teachers, medical experts, agronomists, and engineers. This exchange was especially fruitful in mining, where the coal produced in Mozambique was exchanged for the assistance received, albeit in a macroeconomic context marked by much lower global commodity prices than in the 2000s. Following a methane explosion at the Carbomoc coal mine in Tete, German mining rescue experts were called in, which from 1978 on led to a formal collaboration to manage the coal mine. In 1980/81, 180 Germans were working at Carbomoc, mainly as instructors and heads of units (Krauße, 2016: 31) – leading the GDR illustrated weekly Neue Berliner Illustrierte to claim that ‘the demons [of colonial exploitation] are vanquished’ ⁶ – and by 1982 Mozambique had exported 450,000 tonnes of coal to the GDR (Obernhummer, 2010: 19).

Due to the civil war with Renamo (Resistêcia Nacional Moçambicana, Mozambican National Resistance), coal exploration in Tete ground to a standstill in the mid-1980s. Due to the failures of the planned economy and the war effort, Mozambique ‘capitulated’, in John Saul’s and Joe Hanlon’s words, to the IMF and the WB (Hanlon, 2000; Saul, 1993), and opened up the economy to privatisation. Following the GDR’s collapse in 1989, collaboration was wound down: most Mozambican Vertragsarbeiter (contract workers) had to return, and Mozambique transitioned to peace, multi-party democracy, and donor-dependent reconstruction.

After privatisations, the Carbomoc site was taken over by Minas Moatize from 1998 to 2013; after that the price slump operations halted, and Minas Moatize now only maintain a rump presence on the site. The engineers of Mozambique’s mid-2000s coal boom were new entrants. Brazilian mining giant Vale acquired a large concession in Moatize in 2004; the Australian company Riversdale’s Moatize concession was acquired in 2011 by Rio Tinto (Kirshner and Power, 2015: 72), and sold again in 2014, after the downturn in prices, to the Indian conglomerate ICVL for $50 million.

This brief overview of transnational expert cooperation in the coal-mining sector in Tete serves to rebalance a ‘boom’ narrative that dehistoricises Tete’s ‘long history of governance through private concessions’ (Kirshner and Power, 2015: 71, 72), and obscures Mozambique’s post-independence experience of mutually beneficial transnational entanglements around the mineral resource extraction. The shift from command to market economy is read mainly through the lens of the collapse of Frelimo’s socialist project (Kirshner and Power, 2015). In this perspective, in the move from socialism to capitalism ‘the state largely has withdrawn from economic decision-making’ (Pitcher, 2002: 237). As Anne Pitcher shows, this transition has been interpreted by observers as either a triumph of neoliberalism or, more critically, as ‘recolonisation’, in which external aggression und ubiquitous influence of IFIs have ‘derailed’ Frelimo’s socialist project, offering a gloomy forecast (Pitcher, 2002: 4), with an imposed ‘forgetting from above’ of socialist times (Pitcher, 2006). An alternative perspective holds that the transition has provided opportunities for ‘primitive accumulation’ for the party elite and the creation of a national bourgeoisie (Buur, 2014; Sumich, 2010).

From the view of state functionaries tasked with managing the mining sector, however, an alternative history echoes along in parallel, which questions that narrative. Although the German experts left and Carbomoc was privatised, the impact of this collaboration lingers on, with some ideals of technical expertise and probity promoted by socialist ideology remembered fondly. At the MIREME in 2016, nearly all heads of departments at the ministry had studied in Saxony during the GDR and were mining engineers and geologists trained at the Technical University Freiberg, in the Ore Mountain range near the Czech border. This was also true for some older mining personnel in the provinces, as well as in other ministries I visited, where former GDR students held positions of responsibility. Others had studied in Cuba, in the Ukraine, and in Czechoslovakia (including the current president, Filipe Nyusi, who studied engineering in Czechoslovakia). Carrying out this research with the institutional endorsement of the University of Leipzig opened many doors, with many fondly reminiscing how they had learned German there in their youth.

The coal boom from the mid-2000s on, and the perspective of developing the production of LNG, meant that companies subject to local content laws found it difficult to comply with the regulations and find adequately trained national workers. Donors and corporate actors were engaged in building up capacities in Mozambique to form a workforce capable of filling all the anticipated jobs in the extractive industries. This included reviving some old links, as MIREME had started a scheme to send 20 students a year to study in Freiberg. And, as a Mozambican development consultant quipped, some old cooperantes, who ‘were on the state payroll at 6 dollars a month during socialist times are now back, earning 600 dollars a day’.

Beyond these tangible links of cooperation, ideas of ‘good development’ and the greater good inherited from the time of state socialism can still be traced in how state functionaries reacted to the pressures and challenges of making Mozambique ‘ready’ for the expected upswing of coal prices and the anticipated LNG bonanza.

The need to improve

Between 2008 and 2014 the commercialisation of coal deposits in Tete province and the auctioning of LNG blocks in the Rovuma basin, off the coast of Northern Mozambique, fostered a boom climate of heightened investor interest, and the development of infrastructures for the export of raw materials. For coal, investors like Vale (Brazil), Rio Tinto (UK/Australia) and Jindal (India) developed dormant and new mines around Moatize, Tete, and production took off, from 31,000 (metric) tonnes in 2010, to 179,000 tonnes in 2011, up to 3.26 million tonnes in 2012 and 4.26 million tonnes in 2013. ⁷ 2015 saw a 23% drop in production over the previous year, with several companies suspending their operations, or selling off shares in their ventures. While other operators scaled down because of the price dip, Vale, producing 600,000 tonnes of coal per month in early 2016, decided to open a second mining site to increase production to 750,000 tonnes per month and compensate for lower prices with higher output.

The development of LNG took longer, because of its capital- and technology-intensive nature. Following successful exploratory drillings in 2010, the first blocks were auctioned off to Anadarko (US) and Eni (Italy), though at the time of research, final investment decisions were not yet taken. The price slump had rendered these investments more risky again, and operators were seeking guarantees and concessions from the government in return for their commitment. Accordingly, the estimated production start date had been pushed back from 2017 to 2022, further compounding Mozambique’s economic woes. Moreover, because the state companies set up to manage the state shares in these ventures did not have sufficient capital, these shares had been paid for by loans from international oil companies themselves. This meant that until initial costs were recouped, all revenues would flow to the investors. In the best-case scenario the state would start benefiting in 2030. Nonetheless, industry analysts remained optimistic about Mozambique’s prospects. Rystad Energy, for example, a Norwegian consulting firm, affirmed that if the challenges of developing the oil and gas sector were tackled well, Mozambique would become Africa’s largest hydrocarbons producer by 2050. ⁸

Despite an adverse economic climate at the time of research, there was among my interlocutors a general consensus that the country had to be prepared for the next, ‘inevitable’ cyclical upturn in commodity prices, when investments and revenues would flow again. Most of my interviewees said that the natural resource boom, and the arrival of big multinational corporate investors, had changed the rules of the game, and that Mozambique’s state institutions and infrastructures (transport, power generation) were, in their current form, ill-equipped to deal with the requirements of globalised capital. As Bram Büscher writes, ‘Mozambique does not have the infrastructure or the skills necessary to mine or develop its own mineral and energy resources […], which makes it dependent on outside investors, [and it has] no idea exactly how many natural and mineral resources the country possesses, nor the means to verify the numbers investors and companies come up with’ (2015: 6; see also Kirshner and Power, 2015: 69). The state had to improve its capacities and be made ‘ready’ to deal with this changed environment and manage the windfall from extractives for the greater good of the Mozambican state.

To illustrate these challenges: one of the MIREME buildings in downtown Maputo, where decisions about the management of the sector are taken, and where the activities of global corporations should be monitored, inspected, and regulated, had been without running water for the past eight years, and elevators only worked sporadically. Ricardo, a Brazilian consultant, also told me how the previous Minister of Mines, Esperança Bias, had said in a speech that Mozambique would soon produce 100 million tonnes of coal annually. As he told me, ‘I raised my hand and asked, “do you know what the global coal market is, annually?” “No”, was the reply. “Well, it’s 300 million tonnes – there’s no way that Mozambique could just capture a third of the world market”. I was thrown out. Later they asked me back in…’.

In fact, I was repeatedly told that MIREME, prior to 2008, had been one of the least important ministries, chronically underfunded, and a holding track for disgraced functionaries, those who, because of their position within the ruling Frelimo party, could not be fired and had to be ‘promoted’ to Mineral Resources. This further contributed to the lack of technical expertise and professional ethics at the Ministry, according to interviewees. Things had gotten somewhat better under Bias, who had been instrumental in forging good relationships with foreign investors. The new Minister installed by President Nyusi, Pedro Couto (at the time of research), though a capable economist, had no previous experience in extractives, and repeated internal reshuffles had further disrupted routines. ⁹

Beyond mastering international markets and contract law, state functionaries themselves had to be transformed into newly efficient public servants to live up to the expectations raised by the promise of a resource bonanza (cf. Macuane et al., 2018). The Mozambican public administration, having inherited from colonial Portuguese and post-independence socialist bureaucracy an inordinate love of paperwork and hierarchy (Gonçalves, 2013), and plagued by years of under-funding and regular structural reshuffles, had to be updated to allow Mozambique to operate competitively.

Wilful entanglements

It is in this context of heightened investor interest and an almost constant barrage of expert knowledge about how to improve Mozambique’s management of the extractives sector that I carried out my fieldwork. I had hoped that my experience in commercial risk analysis would allow me to carry out participant observation at MIREME. When I shared this idea with Ricardo, he told me every government department received at least two experts per week and the last thing they needed was yet another (self-appointed) foreign expert volunteering their expertise. What ultimately opened doors was my fluency in Portuguese and German. This allowed me to accompany a German delegation of mining safety and rescue experts on their mission to Tete and Manica as an embedded translator, courtesy of the German development agency GIZ’s (Gesellschaft für Internationale Zusammenarbeit) programme at MIREME and the Inspector-General of Mines, Engineer Obete Matine.

The delegation’s first stop was former state mining company Carbomoc, which had four production sites in Tete from the 1970s to the 1990s, and which, with the help of the GDR, had a dedicated, professional rescue corps. After privatisation, this site was taken over by Minas Moatize, who ran the site between 1998 and 2013; since the price slump operations had been halted. The five gentlemen who were still manning the site and received our delegation knew Obete Matine well; he had worked for Carbomoc after his return from studying in the GDR in the 1980s. They were all seasoned miners, in their mid-to-late fifties; three of them also trained in the GDR. They told us that Carbomoc had a well-organised rescue plan and apparatus, though the equipment was now, in the words of the German experts, ‘history’. The private operator reduced rescue capacities and now that operations were halted, nothing was happening anymore. Still, a risk remained, as even with inactive coal mines there was a latent risk of exploding methane bubbles.

As we drove over to Vale, Matine told me that during the times of socialism, the division between state and company was blurred, with Carbomoc providing a number of state-like services to local communities. This explained, according to him, why the communities nowadays expected similar services and shared responsibility from the private corporate actors.

At Vale we met with the head of safety. Regarding rescue they appeared well prepared: there were trained medical staff and facilities onsite, airlift provisions in case of major injuries, and the rescue team seemed motivated and capable. They had trained first aiders in all units, and there was a clear system of individual and unit responsibilities. Vale’s safety and rescue unit was also regularly called upon to reinforce local fire-fighting capacities, as Moatize district, despite being a highly industrialized zone, had no fire department of its own. The Vale people cited cases of a fire at a nearby petrol pump, the burning down of a communal market, joint drills at Tete airport, and detailed the interaction with local communities when, a few months before, two teenagers had drowned in a water-filled abandoned mining hole that people used for washing.

Despite the evident professionalism of the staff, Vale’s rescue department operated with three firemen per shift, for a total of 12 firemen for the whole concession, which even to a layperson like me sounded rather low. The fire-fighting expert of the German delegation was immediately suspicious; he asked how the risk assessment was made, and how the number of brigade members was calculated. The Vale staff freely admitted that they knew three people was too little:

In the absence of a national norm, we followed the US norm NFPA 1710 [National Fire Protection Association] and came to the conclusion that we needed 11 men per shift. We communicated this to the national director and to corporate HQs in Brazil, but they said that because of the current economic situation they could not increase our personnel and equipment.

Such entanglements came up time and again, in unexpected places. Mozambican functionaries stressed their need for training and capacity-building to donors and consultants. However, donors and experts also repeatedly told their Mozambican counterparts how good it would be if the Minister of Mines transmitted to the President the importance of these programmes, so that the President would express Mozambique’s continued interest in this technical collaboration at the upcoming bilateral negotiations between the two countries. Both to the technical experts and the programme officer of the development cooperation it was key that the Minister underscore how it would be ‘counterproductive to have it [the natural resource capacity building programme] diluted in a larger development cooperation programme pillar like local governance’¹⁰. Reversing facile assumptions about power relations in the development game, foreign development professionals requested the intercession of Mozambican political actors with their own political decision-makers to ensure the continuation of their capacity-building activities. By jointly working towards an instrumental notion of efficiency and good governance of the mining sector, they produced stateness to promote their own goals.

The same holds for entanglements between corporate, state, and civil society actors. All the company representatives speaking at the National Dialogue underscored their determination to uphold their commitments to social responsibility and the country’s development despite necessary cuts. Some of my interlocutors viewed this with a more sceptical eye, saying that community consultations organised jointly by the Ministry of Land, Environment and Rural Development (MITADER), companies, and civil society were little more than a fig leaf, or as they say in Portuguese, ‘for the English to see’ (para o inglês ver), to give a semblance of legitimacy to the relocations. Relocation processes are disruptive and in some instances accompanied by violence, and local communities are often the ‘weakest link’, but as Madeleine Fairbairn (2013) has shown, consultations and the zoning out of lands for commercial activity allow civil society actors and local administration players to improve their position.

The Centre for Public Integrity (Centro de Integridade Pública, CIP) – an organisation of investigative journalists, who are often quite critical of government policy – who have been part of the civil society representatives in the EITI (Extractive Industries Transparency Initiative) process since 2004, lamented that the initiative was really a pact between government and industry with civil society legitimating the proceedings. However, they also saw the development of the extractives sector as an opportunity for civil society to improve their own capacities as interlocutors of the government:

Low prices are the best moment to create a strong institutional framework. We’ve been wrong-footed before; now we have to sit and think very well how to benefit from the new, coming super cycle. We have to look at our weaknesses … to overcome the crisis, and lay down the bases to avoid the resource curse and transform the extractives into a blessing. We have an opportunity now¹¹.

In Tete, contrary to Vale and government promises of better living standards, improved infrastructure, employment and income perspectives, etc., the region’s inhabitants have instead been facing land grabbing, resettlements, non-transparent agreements and police violence, expressing the ‘tensions in the investment–development–dispossession nexus’ (Büscher, 2015: 13; HRW, 2013). However, the creation of a ‘Platform of Civil Society for Natural Resources and Extractive Industry’, where civil society actors come together to articulate their claims, has placed these issues on the national agenda, leading the government to set up a commission to address the matter (Brito et al., 2012). Some corporate informants accused civil society actors of acting in bad faith and building on lands scheduled for zoning out to receive compensation, or of using their advocacy to extort the company for private rather than collective benefits. However, civil society activists from various NGOs reported how the arrival of mining and LNG multinationals led them to adapt to the challenges of large-scale extractive industries. They had raised their own capacities to grasp the legal and technical implications of dossiers, adopting a more ‘constructive’ approach in their interactions with government administration and parliament (cf. Tchatchouang, 2018). They felt they had improved local residents’ capacities, educating communities on their rights, mounting networks of citizen reporting and collective action. Despite the potentially detrimental environmental and social impact of extractives, they recognised the prerogative of the state to direct the economic growth of the country through the development of the sector, and saw this kind of capacitação (capacity-building) as beneficial for the exercise of citizenship. What we see here is that even ‘civil society actors’ who, by the structural role they take up in public debates are often critical of government and industry, end up ‘capacitated’ by the processes triggered by investment into the extractive sector, while recognising and shoring up the state’s authority.

Again, some of these entanglements reverse neat preconceptions about power relations between these different sets of actors. At the National Dialogue, the Vale representative spoke at length about the crisis, and how his company was affected by the high transport costs. He mentioned the ramp-up in production, and plans to sell shares in Vale’s transport infrastructure project, the Nacala Corridor (an upgraded rail link from Moatize to the port of Nacala), stating twice that Vale really needed the government’s support. In the Q&A, CIP asked what support he had in mind, but the speaker’s reply was evasive. In a later interview with one of Vale’s competitors I was told Vale was just finalising the sale of 50% of the Nacala Corridor, for which they stood to pay USD 300 million in capital gains tax to the Mozambican government, and that they probably wanted the government to reduce this. This shows parallels to a case in December 2017, when Eni sold 25% of its stake in LNG Area 4 to ExxonMobil for close to USD 2.8 billion, and President Nyusi, instead of using the newly established capital gains tax, allegedly personally negotiated a far lower tax take for the state. ¹²

Finally, according to some informants, the windfall from extractives and the increased scrutiny the state was under from having to comply with international good governance standards allowed them to adhere to new, higher standards of professional ethics, overcoming what they saw as the Mozambican public administration’s original sin, subservience to the ruling party. This included turning a blind eye to financial misdeeds and private enrichment of some Frelimo grandees. ¹³ ‘Imagine if the first revenues [of LNG] had started rolling in in 2018 – we would have had no capacities to audit them’, said an international consultant seconded to the Audit Court. Now, the price slump allowed Mozambican auditors to put stronger mechanisms in place. Julinho, an auditor at the Audit Court, told me that a 2014 law on the decentralisation of the court had in his view lowered audit standards, as provincial auditors were much more beholden to their superiors:

Luckily, the law was reverted in 2015. Audits at provincial level by provincial auditors don’t work. We have a very strong political system here [glances around furtively] – I am talking about the party in power – and this influence would be felt if the provincial governments were auditing themselves. If the auditors come from here, people there [in the provinces] don’t know you, so there’s not this kind of influencing. There, these are small provinces, so people already know each other, which can complicate the realisation of the mandate. So, the law was reversed in favour of the Tribunal; the law was in effect for not even one year, then it returned to the situation that we think is normal.

There are evidently material incentives for efficiency reforms in the public administration, and stories abound of functionaries travelling to the US for simple Excel trainings, and of handsome per diems skimmed off. Functionaries often see the capacitation workshops they are subjected to as ineffective and resource-absorbing (Sabaratnam, 2017: 70–71). However, the idea of development through the impulse of extractives equally ‘gives expression to a deep-seated will to civilise’ (Gould, 2005: 69), which resonates with the mission civilisatrice and the high modernist nation-building project of Frelimo, Mozambique’s ruling party (cf. Scott, 1998) of the late 1970s and 1980s – the creation of a homem novo (new man) and a modern society based on the principles of scientific socialism – and by extension, the urban middle class that makes up its historical social basis and the reservoir from which public functionaries are drawn (Sumich, 2013, 2015). In addition to producing state effects – new tax regimes and a capacitated bureaucracy – we also see a particular type of state politics being reinforced, in which international standards of efficiency are aspired to, even if in practice a dominant party still has leeway to undermine these (cf. Buur, 2014: 19).

Donors and expert consultants were generally in agreement that while MIREME still had serious capacity constraints, the Audit Court and the Revenue Authority had improved their capacities and professionalism in recent years, which they often credited to the pressures brought on by the extractives boom. Cláudio Dimande, the head of the World Bank-funded Mozambique Mining and Gas Technical Advisory Project (MAGTAP), was in charge of capacity-building projects in various public administration services and contracting consultants to help the state draft new laws and regulations. ‘This is a new sector that is extremely aggressive, very competitive, and every decision has very drastic consequences. Little by little, they [the public functionaries] are starting to understand that this is not the tomato production chain,’ he said laughingly, implying the need for a different, higher level of technical and economic knowledge to manage the extractives sector efficiently.

Conclusion

The idea that mining was ‘socially thick’ in the 1950s–1960s and is now ‘socially thin’ with few to no linkages to society outside the mining and oil enclaves remains a popular trope. Shifting the analytical perspective to the co-production of stateness nuances this conception. We can observe the desires of state, corporation, and civil society actors to make resource extraction (seemingly) socially thick, as opposed to the ‘mutual abdication of responsibility’ described by Hannah Appel (2012) in Equatorial Guinea. This is clearly a perspective gained by ‘studying up’ – evidence from ‘the ground’ in Tete shows that for surrounding communities there are few positive benefits from the activity of mining corporations; instead they are facing displacement and serious environmental and health problems from the coal dust resulting from blasting of open-pit mines (Büscher, 2015; Kirshner and Power, 2015). In Mozambique, ‘most large state-driven development projects […] have not only failed to alleviate poverty and promote sustainable livelihoods but also often imperiled the lives of the poor’ (Isaacman and Isaacman, 2013: 7). And yet, the wilful entanglements in which state, corporate, and civil society actors engage to develop the sector in Mozambique challenge the idea that extractive industries in Africa lead to enclaving, the fragmentation of sovereignty, states of exception, bifurcated governance (Kirshner and Power, 2015: 70), or the mutual abdication of responsibility (Appel, 2012). While evidence from the mining sites suggests that all of these can be observed in Mozambique (Mosca and Selemane, 2011), there are parallel processes taking place in office suites and conference hotels in Maputo. In these entanglements and interlinkages new forms of ‘articulated sovereignty’ (Lunstrum, 2013) are co-produced. And these entanglements have a historicity that goes back to before the current boom-and-bust cycle, revealing how some ideals of efficiency and the development of the country have been reworked in the country’s economic liberalisation.

This is not a matter of uncritically singing the praises of ‘what extractives can do for a poor country’, in the words of an informant at Anadarko. The development of extractive industry has effects in unexpected places, recombining a cast of diverse actors in constellations that occasionally reverse common assumptions about their unequal power relations. As my informants told me, the pressure to manage the sector and its revenues for the benefit of the Mozambican state led to what they perceived as improved efficiency in the public service; new fiscal regimes; strengthened capacities in the audit court; and more assertive NGOs. At the same time it provided justification for expert missions for donors to continue their work, and common ground upon which mining companies could petition the government for concessions. Beyond challenging the imagined power gradient between the mighty multinational and the weak mining inspectorate, the example of the safety managers at Vale indicates the continued need to disaggregate ‘the corporation’ as a social actor (cf. Welker, 2014, e.g.). We could dismiss these interactions as pure window-dressing, corporate actors pretending to uphold international standards of environmental and social responsibility (as their reputation is on the line on the stock markets), and state functionaries and NGO activists skimming off individual benefits. Yet these entanglements reveal the multiple, unexpected outcomes of interactions between state, corporate, and civil society actors.

Critical scholarship has noted how such ‘good governance’ interventions serve primarily to make countries ‘fit for foreign investment’ (Sabaratnam, 2017), while allowing foreign companies and governments to shift the responsibility for potential negative developments onto the respective national governments (Donner, 2009). Extractive industries in Mozambique have so far had limited local beneficial impact, leading instead to state violence, dispossession, and environmental degradation (though advocates of these ventures will point to the provision of electricity, housing, and training schemes). For those who work the state and work for the state, the development of extractives provides ample opportunities to collude (Fairbairn, 2013), or jointly pursue the visions and promises of national development and modernisation through a big leap forward (cf. Engel, 2018), triggered by investment in extractive industries.

It is not for us to judge how much of this is self-interest, false consciousness, or ‘real’ commitment to the greater good. We can observe complicity and a co-production, where it is profitable for all concerned to uphold an idealised vision of development through extractives. These wilful entanglements result not so much in the abdication of responsibility (Appel, 2012) but reveal, in their gestures towards making resource extraction responsible, orderly, and socially thick (cf. Rajak, 2011), the co-production of articulated sovereignty.