Effervescence and Ephemerality: Popular Urban Uprisings in Mozambique

ABSTRACT

This article analyses the large-scale popular urban uprisings that shook Mozambican cities on 1 and 2 September 2010, following the government's announcement of successive rises in the price of public transport fares and basic commodities. Using ethnographic material from the city of Chimoio and the capital Maputo, the following work highlights the organisational character of the ‘strikes’ (greves), as the popular uprisings were called, and explores them as a new form of organising political discontent. Comparing them to other historical and contemporary popular uprisings, this article argues that the strikes violently and rhizomically generated ephemeral and egalitarian forms of political authority and order that simultaneously confronted, replicated and undercut the aspects of Mozambican statehood. Deploying Durkheim's notion of effervescence, the work further argues that the creative fervour, multisemic aspects and festive character of the popular uprisings need to be recognised; thus, this analysis challenges the reductive labelling of these events as ‘riots’ or ‘food riots’.

KEYWORDS:

• Effervescence
• Mozambique
• popular uprising
• rhizome
• sovereignty

When we walked down the streets and tore into the shops and took what we wanted – it was fantastic! We also controlled the roundabouts and bridges – chasing the police off. And we went to the police station, shouted and sang. They [the police] had to hide. They were afraid! It was fantastic – it was popular power [poder popular] all over again and we organised everything ourselves. We could do what we want and everyone was happy!

A month after Chimoio's urban protests on 1 and 2 September 2010, this is how a 22-year-old man excitedly described them to me while I was carrying out fieldwork. Many elements are contained in this typical quotation but worth noting are the emancipatory, collectively egalitarian and festive elements of what its participants referred to as o greve – ‘the strike’. Indeed, the self-organisation, the takeover of state infrastructural space and the fervour of networked mass mobilisation are central aspects of the strikes that have repeatedly challenged Mozambican state sovereignty in recent years. Corresponding to what my interlocutors call os greves (‘the strikes’) and what I, inspired by Badiou's (2012: 90) notion, will term ‘popular uprisings’, the strikes thereby seem to challenge Jean-François Bayart's famous claim (1993) that the African state is rhizomically constituted in that it relies on extensive, dynamic and changing networks that lie beyond and below the reach of its formal domains. Contra Bayart's claim, the 2010 strike was preceded by similar events in 2008, calling into question a reading of the state as incorporating fully powerful rhizomic domains. Gripping cities for several days, the scale of the 2008 and 2010 strikes was such that they encompassed both the urban spaces normally associated with sovereignty and engaged rhizomic networks and capacities beyond the reach of the Mozambican post-colonial state. What is more, the strikes were frequently characterised by a festive atmosphere of upheaval, creativity and collectivity – traits that Durkheim more than a century ago identified as key to dynamics of not only ritual and sociality but also of historical rupture and politics:

There are periods in history when, under the influence of some great collective shock, social interactions have become more frequent and active. Men look for each other and assemble together more than ever. That general effervescence results which is characteristic of revolutionary or creative epochs … Changes are not merely of shades and degrees; men become different. The passions moving them are of such an intensity that they cannot be satisfied except by violent and unrestrained actions, actions of superhuman heroism or bloody barbarism. (Durkheim 2008 [1915]: 210f)¹

Informed by such a perspective on effervescence (also supported by Karen E. Fields’ [1995: xl-xliii] of this), and taking the Mozambican strikes as conforming to his notion of ‘great collective shocks’, this article approaches these events as irreducible to the oft-applied label of ‘food riots’ (cf. Walton & Sneddon 2011). Further, and contrary to much research on Mozambican urban situations and politics, which is generally based on Maputo material (see, for instance, Sumich 2010; 2013 for two incisive analyses), this work also incorporates substantial material from Chimoio, the provincial capital of Manica province. Through mapping and analysing urban sites also beyond Maputo, it has been possible to highlight how post-colonial state contexts entail a networked and, often, incomplete articulation of sovereignty and political authority in urban spaces. Consequently, the following argument presents the sovereign formations of African states neither as necessarily emanating from an imagined or spatially defined centre nor as a formation based on the control of rhizomic domains, as in Bayart's argument. Instead, the following work explores how the Mozambican political order has been involved in recurring confrontations that have produced unsettling effects, which ultimately underscore the contested and partial nature of post-colonial sovereignty (see also Hansen & Stepputat 2005). It analyses the ways in which such sovereignty has been ephemerally appropriated by novel political formations, such as the popular uprisings, as well as how lateral, egalitarian and digital dimensions have been integral to their trajectory and articulation. New political formations, such as those manifested through the strikes, are also characterised by multiple political aims and by multisemic strategies. Finally, such formations are frequently experienced as effervescent, festive and carnivalesque events by their participants.

Interlude: Text Messages, Unrest and Trans-African Political Fears

In an age in which tropes of representative democracy, human rights and civil society are hegemonic and ubiquitous (Fassin 2012), it might be assumed that protests and popular uprisings would disappear. Not so, a long historical trajectory of insurgence and protest across northern Africa and the Middle East (Fosshagen 2014) and, more generally, new global, radical formations of urban protest have again made relevant debates about the potential these have to effect profound political change (see, e.g. Badiou 2012; Harvey 2012). Such potential for change has not gone unnoticed by southern African political leaders. At a Southern African Development Community meeting in April 2011, the Chair of the Organ on Politics, Defence and Security Cooperation, Zambian president Rupiah Banda, warned heads of state: ‘If there is anything that we must learn from the upheavals going on in the northern part of our continent it is that the legitimate expectations of the citizens of our countries cannot be taken for granted’ (quoted in Zhangazha 2011). President Banda's warning came after southern Africa experienced the effects of protests, following a text message that went viral in neighbouring Mozambique during mid-August 2010:

Mozambican – prepare yourself for the great day of strike [greve] 01/09/10. We will reclaim the rise in prices in electricity, water, rice, public transport and bread. Send to other Mozambicans.²

Displaying the ubiquitous emoticon, this text message may seem innocent enough, and the Mozambican authorities seem to have regarded it as such. Nevertheless, early on Thursday, 1 September 2010, following a shutdown of public transport, large numbers of people poured onto the streets of the capital Maputo; similar events occurred in the cities of Beira, Chimoio and Matola. The Maputo crowds were engaged in a popular uprising rather than a conventional ‘strike’; however, burning heaps of tyres as barricades on main roads, and overturning vehicles, the participants succeeded in blocking key roads into and around the city centre for two days.³ Despite heavy-handed police intervention, including the use of live ammunition and tear gas (which killed 14 people, including young children), the two-day strike witnessed widespread looting of shops and warehouses, which especially targeted food staples, such as rice and cooking oil (CIP 2010). Such events were not unprecedented in Mozambique, however, protesting against multiple price rises, a popular uprising similarly called a ‘strike’ (greve) was initiated on 5 February 2008 and had also rocked Maputo and eventually spread to other cities including Chimoio. As in 2010, the upheavals of 2008 had involved looting, the takeover of urban spaces and the deaths of protesters.⁴

From Portuguese Rule to Poder Popular and Beyond

The strikes of 2008 and 2010 unfolded in a country that historically has been rocked by various forms of political violence and long-standing contestations over sovereignty and territorial control. Crucially, Portuguese colonialism, as it took hold in what was later to be called Mozambique, was contested from initial struggles for control in the 1500s until the Portuguese state ceded its African colonies in 1975 (Newitt 1995). A decisive blow to Portuguese rule in Mozambique began with the liberation movement Frelimo (Frente de Libertação de Moçambique), which fought against Portuguese rule from 1964 until 1975. With Frelimo's ascendance to power upon independence in 1975, Mozambique embarked on a revolutionary path of development with a particular socialist orientation (Isaacman 1978). Instituted as a one-party regime, Frelimo set out both to radically transform the structural inequalities left by the colonial period and to lead o povo – the people – to a new era of prosperity and equality. This radical politics involved the targeting of what was seen as obscurantism: traditional leaders, chiefs and healers were seen as relics of the colonial regime who were to be substituted by party secretaries and medical doctors (Bertelsen in press).

While initially successful, Frelimo's implementation of a revolutionary politics was severely hampered by a civil war between 1977 and 1992 wherein Renamo (Resistência Nacional de Moçambique), funded externally (initially by Rhodesia and thereafter South Africa), sought to destabilise the regime (Nordstrom 1997). Despite its extreme violence, Renamo paradoxically experienced widespread rural support in certain regions, largely from peasants antagonistic towards Frelimo's radical modernising project (Geffray 1990). The country's first free elections in 1994 reflected such support and Renamo and Frelimo – both conventional political parties at this stage – almost divided the electorate between them, with Frelimo winning by a narrow margin. Since then, Frelimo has remained in office, Renamo remains a key opposition party and political force, and Mozambique has emerged from the ashes of civil war with an impressive annual GDP growth rate averaging 8.3% annually between 1997 and 2008 (Cunguara & Hanlon 2012: 627).

Until recently, most donors (and many researchers) contended that Mozambique's post-civil war development was successful, that economic growth was ‘trickling down’ and that the country was ‘on the right track’ supported by the IMF, the World Bank, multiple bilateral aid agreements and numerous NGOs (Clément & Peiris 2008). In 2009, therefore, the shock among donors was great when reports showed that despite allegedly sustained economic growth for almost two decades, poverty was actually increasing rather than decreasing (Cunguara 2012). Lately, Mozambique has experienced an image problem: the still much-used socialist rhetoric of A luta continua – ‘The struggle continues’ – is now popularly expressed by my interlocutors and others as A pobreza continua – ‘Poverty continues’.⁵

The image of Mozambique as a post-conflict success story is also increasingly questioned by the country's donors (Cunguara 2012) – sometimes due to the mixed effects of a donor-supported multipronged campaign for decentralisation within a range of domains, such as justice, local administration and health (Manning & Marlbrough 2012; McKay 2012). While undoubtedly producing significant structural changes, such developments also need to be framed within a wider political–economic context: after the destruction of state infrastructure and resources during the civil war, in the 1990s, the country suffered the most wide-ranging programme for privatisation in sub-Saharan Africa – a process involving corrupt accumulation and rampant asset stripping as well as what Santos (2006) terms ‘the production of a weak state’. Further, within the domain of justice, the decentralisation of state powers has included the introduction of a community police force (polícia comunitária) that was formed as a result of the low standing and violence of the regular police (Kyed 2009; 2012; Bertelsen 2010; 2011). The establishment of these community police forces has coincided with a dramatic increase in popular lynchings, as well as the emergence of police-led death squads (Bertelsen 2009b). Arguably, therefore, for the urban population in the bairros (urban quarters), an effect of donor-driven decentralisation policies has been the pluralisation of authority structures (Santos 2006).

Taken together, trends such as increased poverty, a multiplication of both forms of violence and of authorities in poor urban contexts and an increasingly business-oriented one-party state system are generally experienced by my Chimoio and Maputo interlocutors as comprising the backdrop for the two recent Mozambican popular uprisings.

Maputo Strikes: Barricading, Looting and Protection

As can be discerned from the text message copied above, an apparently unknown source called for a national ‘strike’ (greve) on 1 September 2010; subsequently, people flooded onto the streets of Mozambican bairros.⁶ A young man whom I talked to in January 2013 recalled these events:

There was singing and dancing everywhere – on the roundabouts, in the streets! Some were drinking. Many were singing ‘Guebuza is a thief’ [Guebuza – ladrão!] Many of my neighbours participated. People I did not know also. Many also took things from the stores that they wanted. It was good!

The alleged thief referenced here was Armando Guebuza – the current president of Mozambique and leader of Frelimo. Soon, however, the ‘street party’ aspects of the strike were supplemented by multiple forms of physical protest and violence primarily directed at two urban dimensions: the spaces and domains of the state and objects of wealth. Underlining the carnivalesque and uncontainable elements of the 2010 strike, a woman in her mid-20s from Bairro Maxaquene C, a low-income, high-density part of Maputo, told me:

The strike [greve] is called xitereka in Shangaan.⁷ Xitereka is better than calling it greve. It means a state of disorganization – where everyone could do what they want. Xitereka was a good thing. It showed Frelimo that we do not like them. That they make things too expensive and that it is dangerous to keep all the nice things for themselves only. So we marched to the places with these things.

As reflected in this extract from an interview in 2013, the strike (or xitereka) involved the targeting of certain state spaces: early on the first day, many main roads leading to Maputo's city centre were filled with large crowds – young and old, men, women and children – who descended on passing vehicles; they overturned a number of minibuses, buses and cars and set them ablaze. Attacks on vehicles soon gave way to the construction of large barricades of burning tyres, which made it perilous to try to pass by car: those who attempted it were pelted with stones, bricks or other hard objects. A vast number of supermarkets, shops and large warehouses were also invaded and ransacked. People were mainly after rice, cooking oil and flour – the goods mentioned in the text message as being subject to price rises – but all kinds of items were taken, including refrigerators, telephones, computers and furniture.

The widespread looting and the state's loss of control of urban spaces did not go unchallenged. The Mozambican police were sent onto the streets to confront the participants, and in many bairros they employed extreme force in their efforts to repress protesters. Using tear gas indiscriminately – even deploying it where they themselves would suffer the effects due to the wind direction – large crowds of people were forcibly moved from location to location. The police also fired both live ammunition and rubber bullets directly into groups of people, killing at least 14 people. Unsurprisingly, reports on police behaviour were damning, emphasising unpreparedness, fatigue, lack of training and the absence of a clear chain of command, as well as the use of disproportionate violence (CIP 2010).

The popular uprising lasted two full days, crippled the Mozambican capital and prevented urbanites from getting to work from outlying bairros using minibuses (called chapa or chapa 100), which were their primary mode of transport. As businesses effectively ground to a halt and almost no vehicles dared travel the roads, the central parts of Maputo became deserted. However, some of the bairros in which the wealthy and powerful lived were exempted from attack. In Maputo, this included the Sommerschild bairro, which comprised foreign development workers, diplomats, politicians and middle- to upper-class Mozambicans (AfricaFocus 2010). This use of police to protect the wealthy and powerful from the melee of the strike that was unfolding in other central as well as more peripheral bairros conforms to an increasingly global uneven distribution of urban security and protection. As Paasche and Sidaway (2010: 1566) assert, the trend is towards a ‘commoditisation and fragmentation of security provision in Maputo combined with the inability and/or withdrawal of the state to monopolise the coercive use of force’ (see also Buur et al. 2006). Moreover, the dramatic events such as those of 2008 and 2010 feed into and sustain urban imaginaries of dividedness, exclusion and abjection that are experienced by urban inhabitants who occupy the spaces adjoining these wealthy and well-securitised areas (Bertelsen et al. 2013).

The Anatomy of the Strike in Chimoio

Chimoio, the provincial capital of Manica, has around 200,000 inhabitants and is located in the Beira corridor some 100 km from the Zimbabwean border. In recent decades, its bairros have grown steadily, attracting migrants fleeing Zimbabwe as well as some rural to urban (partly seasonal) migration; I have observed this process during a number of fieldwork studies in urban and peri-urban Chimoio since 1998 (see also Lubkemann 2008). Visiting Chimoio in October 2010, a month after the popular uprising and, again, in July and August 2011, the following is based mainly on interviews, reports and fieldwork carried out over this period.

In key ways, the trajectory of the popular uprising in Chimoio reflected that in Maputo: roads to the town centre were barricaded in order to block traffic. This takeover of the transport infrastructure included a few crucial roundabouts, such as the symbolically important Praça dos Herois (Heroes’ Square), which enshrines and commemorates Mozambique's liberation struggle and celebrates Frelimo's role in it. A few bridges were also controlled by participants including, notably, a bridge to the north of the city centre, which made access to the populous Bairro Nhamahonha difficult. In addition, central communications, including telephone lines, were cut, as were electricity lines. Police stations and police cars were attacked and, as in Maputo, the Chimoio police applied indiscriminate force, eventually killing some individuals and injuring many more.

Alert to the developments in Maputo, Chimoio and other cities, the Minister for the Interior, José Pacheco, appeared on national television at 1 pm on the first day of the 2010 strike. Intending to calm things down Pacheco firmly rejected the notion that the government would back down on the price hikes and also attempted to blame Renamo for the protests. Later that afternoon, Pacheco was replaced on national television by President Armando Guebuza – the man represented as a thief by many Chimoio protesters. Guebuza denounced the strike as a series of violent acts of vandalism, in an obvious attempt to depoliticise the situation and criminalise its participants. Somewhat paradoxically, Guebuza also repeated the Frelimo government's slogans – ‘Production is the way out of poverty’ and ‘With Frelimo we will win against poverty’. Neither televised announcement had the intended effect and protesters continued to pour onto the streets. The popular uprising was helped, my interlocutors argue, by the private television channel STV: from 9.30 am, STV ran live coverage of the events with camera teams trawling Maputo bairros documenting where and how people were protesting.

By the time the events subsided during the afternoon of 2 September, the Mozambican government had brought in two measures beyond the largely failed attempts of violent crackdown and dissuasion through public announcements. Firstly, it backed down and claimed there would be no price rises, thus awarding the participants a victory in these terms. Secondly, it is widely believed the government managed to coerce the two dominant telecom companies – Vodacom and the national company Mcel – to shut down or seriously impair national text messaging services (AFP 2010). This shutdown effectively meant a halt in the spread of information among participating groups and individuals and has in early 2014 been followed up by a motion to criminalise the spreading of text messages, emails and publications that ‘insult or endanger the security of the state’ (translated and quoted from Club of Mozambique 2014).

This kind of communications blackout is not unique to Mozambique: following their 2005 election, the Ethiopian government shut down the ability to send text messages for two years, fearing their potential use by opposition parties (Eagle 2010: 15). Likewise, responding to urban protests in July 2011 (AFP 2011), the Malawian government blocked access to certain news websites, Facebook and Twitter. In research terms, although Tom Lodge (2013: 151) is too optimistic in claiming that ‘[a]ll over Africa, the spread of communications technology has had an empowering effect for ordinary people’, the potential use of digital technology in governance, security and elections is increasingly an important subject of study in Africa and beyond (see Hellström 2011; Poblet 2011, for example).

This more general trend of a digital component in an African politics of discontent (and its containment) was clearly present in the events outlined here in Mozambique: importantly, the state's (likely) intervention in telecommunications, its use of violence and the relative ease with which participants achieved their immediate goals underline the need to understand both the anatomy and the fervour of the popular uprisings and the organisation of political and social discontent within the African state.

The Rhizome of the Strikes

Since protesters sent text messages using ‘pay-as-you-go’ SIM cards purchased from informal street vendors, it was almost impossible for the government to identify senders or receivers. Until after the 2010 strike, 95% of mobile phones users in Mozambique used such street-bought SIM cards with no obligation to register their details (Archambault 2011). This situation has now changed, as the Mozambican state has made it compulsory for all users to register their SIM cards (Mabila et al. 2010: 4). At the time of the 2008 and 2010 strikes, however, the lateral spread of text messages reflected the organisational dynamic of the popular uprisings: there was no apparent hierarchical structure in the form of a recognised leadership that endured (even within the context of the popular uprisings), as would have been the case had this been a ‘strike’ in a conventional sense of being organised by a trade union, for instance.

More surprising, perhaps, is that while it might have been expected that Renamo would be involved – it is Mozambique's largest opposition party and regarded as crucial to the country's mostly peaceful post-war political development – it was not, to any significant degree, despite the allegations made by Interior Minister Pacheco and President Guebuza on national television. Indeed, my interlocutors from both Maputo and Chimoio hold that Renamo played no organising role whatsoever and, hence, did not influence the level of popular participation and the extent to which the uprisings spread within and between Mozambican cities. This view is corroborated by the fact that post hoc Renamo failed to claim responsibility for or leadership in the strike or to transform these events into forms of political and symbolic capital. This is especially significant in Chimoio, as there, and throughout the province of Manica, Renamo has historically received the majority of provincial votes in presidential and parliamentary elections. Musing on Renamo's absence in relation to the strike, a young Chimoio man from the afore-mentioned Bairro Nhamahonha commented to me in October 2010:

Ah, Renamo! They used to be big, they used to have power! During the war, they beat Frelimo. Hard! They brought order – they brought the chiefs back. They did all! But now, ah, they are sitting in Maputo eating the way Frelimo has always eaten. Now Renamo has no power. And Frelimo does not want to change anything. So, the people [o povo] need to show them how to solve things.

What is indicated here, and expressed by several other interlocutors, is that recent events such as the spate of lynchings, as well as the 2008 popular uprising, paved the way for a mode of popular direct engagement with the state – in particular with its elite and its accumulation of wealth (Bertelsen 2009a; 2014; see also Serra 2008). In a context in which the two dominant political parties were perceived either to have retreated from politics (Frelimo) or to have lost their traction (Renamo), the strike's speed and lateral egalitarian organisation assumed significance (see also Cahen 2010). At one level, therefore, the dynamic of the popular uprising resembled what Deleuze and Guattari (2002 [1980]: 358), using the example of ‘bands’, call the rhizomic form:

Packs, bands are groups of the rhizome type, as opposed to the arborescent type that centers around the organs of power. That is why bands in general, even those engaged in banditry or high-society life, are metamorphoses of a war machine formally distinct from all State apparatuses or their equivalent, which are instead what structure centralized societies.

In this sense, the form of the band – which resembles key aspects of the lateral and fluctuating form of an insurrectionary popular uprising – exhibits the unsettling potential inherent in rhizomic social and political orders that also reside beyond the reach of the state.

Appropriating the Spaces and Wealth of the State

The strikes’ instant, acephalous, rhizomic and ‘band-like’ features notwithstanding, the protesters also, if ephemerally, occupied key state spaces and infrastructural points, including bridges, roundabouts and main roads. This particularly affected Chimoio's bairro cimento (‘cement city’) – the city centre, which comprise zones of business, wealth and the state. Crucially, this temporary isolation of the bairro cimento had its corollary in the spatial dynamics of lynchings, which have occurred regularly in Chimoio and other Mozambican cities in recent years (Bertelsen 2009a). In the context of the afore-mentioned decentralisation initiative, these events resulted in people perceiving there to be a police presence only in the bairro cimento while some municipal authorities were seen to condone lynchings in the poor bairros. As an effect of these processes, the lynchings have been distributed in a centrifugal manner to urban spaces that are peripheral to the Mozambican state.

The dynamics of the strikes indicate a similar spatial dimension but with a centripetal force: rather than impoverished bairros being abandoned by the state it was the bairro cimento, the very space of the state and wealth, that was unsettled and targeted. In such a way, the infrastructural insignia of state control were disordered and re-inscribed, and sometimes directly attacked – as in the case of Chimoio's main police station and police cars. The participants engaged, then, in a form of spatial re-negotiation of the state apparatus. That the 2010 strike unfolded in particular spaces was expressed in February 2013 by a chefe do quarterão (neighbourhood leader) from Bairro Maxaquene B – partly reflecting the Mozambican government's stance of criminalising the participants:

Good people were silent and did not participate. But the strike [greve] only happened where there were things. Here [in Maxaquene B] we have nothing. So here nothing happened. But in other places, ah, a lot! War comes from the stomach, you know. People need work and things.

Reflecting this critique indirectly, other interlocutors who had been participants commonly held that their actions would provide some form of redress for what they regarded as illicit accumulation: if the state and business elite were perceived to have merged and to be increasingly wealthy, why should people be barred from taking part in these riches? Such re-appropriation of wealth was often explained to me in 2010, 2011, 2013 and 2014 as being the rationale for the looting of supermarkets, shops and other businesses. It was also evident in an interview carried out with a middle-aged woman from Chimoio's Bairro TextAfrica in October 2010. She had participated and saw herself as having been successful: she had returned with a 20 kg sack of rice and explained to me why she had participated:

They have eaten for years and years. Peace has come, democracy has come, elections have come. All has happened – but they are the only ones eating. Our children are hungry: we should eat as well.

Such direct action against those perceived to be wealthy, greedy and powerful underlines the clear political dimension that the popular uprisings have for the participants. However, it also indicates that Mozambicans are neither pacific nor uninterested in politics – despite having experienced gruesome civil war violence, an onslaught of neoliberal reforms dismantling state services, and being continuously under Frelimo rule since 1975 (West 2005). On the contrary, because the higher echelons of state power have become gradually less centred on o povo (the people), and as the vision of socialism retains only rhetorical vestiges of its former self, by engaging directly with state agents such as the police, people address (and actually redress) in practical terms the sources of their marginalisation and poverty (see also Sumich 2010). A young, unemployed man I interviewed in Chimoio in August 2011 revealed the frequently commented upon relations among cost of living, the elite and the police:

I participated because I wanted to destroy the shops that are selling everything at a very high price. We went to the police headquarters [Primeira Esquadra] as well. There we threw great many rocks at the buildings and all the policemen ran to hide inside. From there they fired their guns but more in the air than directed at the population. We went to the police as they are not resolving problems and cases there. If you [are a criminal and] have money, you are released right away. If you do not have money to pay the police, you will stay there and get heavily beaten.

As the young man indicated, the strikes unfolded in particular urban spaces and had specific targets (see also Bertelsen 2009a). In Chimoio, this included recent symbolic and material domains of exclusion and immoral accumulation: the Chinese and their shops. These were targeted alongside stores owned by Mozambicans of Indian descent (see also Serra 2010). An important backdrop for such targeting is that Mozambican newspapers have repeatedly exposed the fact that members of the Frelimo elite part own large companies otherwise controlled by Chinese people, people of Indian descent and other non-Mozambicans (Hanlon 2009; Ribeiro 2010). For my interlocutors, such affinities and connections between Frelimo and ‘big business’ made these shops natural targets during the strikes: attacks on them were construed as appropriating the visible riches of the party and on what was perceived to be its betrayal of o povo for its own benefit and that of foreign interests.

A simple ‘eat the rich’ interpretation is, however, too crude in this context; in Maputo, stalls and small shops owned by Nigerians were also targeted to some extent, while Chimoio market stalls (banca fixas) owned by Somalis were looted and razed to the ground. These people – Chinese, Nigerians and Somalis – are not only regarded as successful businessmen at the expense of Mozambicans, they are also widely believed to control the drugs trade, and to be involved in vehicle theft and robberies. In Chimoio, Somali stalls in large informal sprawling markets can hardly be seen as representing the party in power. In addition, and reflecting local politics and civil war memories, the stalls of people from the northern Mozambican city of Quelimane were looted and burnt. Thus, it would seem that the participants’ understanding of their marginalisation was in part shaped by the kinds of xenophobic political dynamics to which Mozambican migrants have been violently subjected at various times – in South Africa in 2008 (Friebel et al. 2012), for example.

‘Now They Are Afraid of Us’: The Strikes as Exuberant and Unsettling

Above, it has been observed that the strikes relate to recent Mozambican politics in general and to the state in particular. However, we should be cautious about seeing these Mozambican popular uprisings – and those of North Africa and the Middle East in the spring of 2011 onwards – as entirely new phenomena. Hobsbawm (1965 [1959]), for instance, examines pre-industrial urbanites (‘the mob’) involved in ‘primitive rebellion’ in southern Europe: ‘[i]n such cities the popolino [wage-earners, small property-owners and urban poor] lived in an odd relationship with its rulers, equally compounded of parasitism and riot’ (Hobsbawm 1965 [1959]: 115). The description Hobsbawm provides of this as a political order that cyclically oscillated between modalities of re-distributive rioting and modes of appropriation by the elite seems to resemble the recurrence of strikes in Mozambique in 2008 and 2010. Hobsbawm goes on to note three salient features of ‘the mob’: that it ‘did not merely riot as a protest, but because it expected to achieve something’; that it was ‘always directed against the rich and powerful (though not necessarily the official head of the state or city)’; and that it was infused with hostility towards ‘non-townsmen’ (Hobsbawm 1965 [1959]: 111f). While some of these aspects conform to characteristics of the Mozambican popular uprisings – the instrumentality of looting and the targeting of conceptual strangers (Indians, Chinese, Nigerians, Somalis) – the directionality of the strikes ‘against the rich and powerful’ does not. As shown, Maputo's and Chimoio's elite urban spaces were cordoned off by security, and a number of non-elite targets (Somali stalls, for example) were also targeted.

Hobsbawm's orderly account of rebellions resembles analyses that see them as ‘safety valves’ for venting popular frustration; such analyses often employ the term ‘food riot’ (see Walton and Sneddon 2011). This view smuggles in instrumentality, neatly contained in the term ‘disaffection’, and glosses over the fervour of creativity and the experience of exuberance characterising many popular uprisings such as those in Mozambique. Moreover, the festive experience my interlocutors recounted corresponds with the atmosphere reported elsewhere: an insider account of the 2001 anti-globalisation demonstrations in Genoa, for example, observes that the running battles with the police and the smashing up of cars and specific stores constituted a carnivalesque occasion (One Off Press 2001); such traits are also often invoked to describe more recent protests across urban Britain (Easton 2011). A similar reflection is found in anthropologist Sherry Turkle's (1975) analysis of the symbolic action of festivity characterising the student uprisings in Paris in May 1968. Crucially, Turkle argues that ‘[t]hese groups never aim, it seems, to implement a program, but to express the self instead, or the spontaneity of a new-formed group, freeing themselves by their protests from structural roles and necessities’ (Turkle 1975: 68).

Similar features are also identified in several contemporary and historical studies of Africa. Glassman (1995), for example, underlines the important relationship between religious fervour and large-scale uprisings in Swahili Coast cities in the late 1800s; he shows that the potential of the festive occasion was underestimated by the German colonial masters. Glassman concludes that such miscalculation in relation to the ensuing violent events led to a temporary defeat and the ousting of colonial rulers, arguing that the rituals created an atmosphere in which uprising and looting was made possible through idioms of feasting. In the same vein, a study by Villers and Tshonda (2004; see also Boeck 2007) delves into the large-scale rioting that has characterised urban Kinshasa from time to time from the 1950s onwards. The 1991 and 1993 Kinshasa protests, for instance, resulted in subsequent waves of looting: ordinary people followed the example, first, of looting soldiers and then of police officers. Many Kinois refer to these riots as la grande fête (the great party) and their dynamic is embedded in historically produced understandings of political culture and authority, as well as in modalities of appropriation and redistribution.

Arguably, then, the fervour of the Mozambican strikes and their ludic and festive elements may not be explained fully either by invoking political instrumentality or by arguing for them as blindly reflecting crude economic necessity. Although the ethnographic and historical contexts described above differ greatly from Chimoio and Maputo in 2010, similarities do exist in terms of how the popular uprisings involve dynamic forms of egalitarian collectivity exhibiting a politics of exuberance. Arguably, these elements are contained in Durkheim's notion of effervescence, quoted in the introduction of this article, as integral to great revolutionary or creative periods. Parkin (2007: 246) understands Durkheim's notion to be necessary if we are to appreciate the force, volatility and dynamics of the (insurrectionary) crowd:

In other words, effervescence, like the crowd, is inherently ambivalent emotionally, able to switch moods through a combination of internal and external dynamism. Inscribed within the bodily constitution of the crowd, then, we have in effervescence the potential for schismogenesis, expressed as their bodily embrace or violence and always sheer physical energy.

This notion of effervescence aids our understanding of the sense of thrill and empowerment that is important to my interlocutors’ experiences of the riots (Groes-Green 2010; see also Verkaaik 2004: 111ff for a related analysis). Further, my interlocutors strongly emphasise this excitement, which occurred as they engaged in running battles with police, participated in looting and as they attacked or appropriated spaces associated with the state. Such multiplicity in terms of relations between the state and its population has also been argued by Mbembe (2001), who emphasises that the post-colonial state cannot be fully grasped in familiar dichotomies such as repressed/repressors or resistance/power. Instead, Mbembe asks us to analyse the playful and the carnivalesque in epistemologies and practices of power, in which ‘those who command and those who are supposed to obey are so entangled as to render both powerless’ (Mbembe 2001: 133).

However, the enmeshed nature of the two (often indiscernible) categories does not necessarily end in spirals of simulacra or mutual disempowerment and, again, the strikes as they unfolded are cases in point: beyond the violence perpetrated by the Mozambican state in trying to quell (rather than contain) the popular uprising (similar to the Kinshasa case quoted above), many policemen actively participated in barricading the streets and, in particular, in the looting of warehouses and shops. Further, interlocutors divulged that not only did ordinary police officers take advantage of the popular uprisings to run off with rice and cooking oil, they also shot tear gas canisters into crowds, drawing in bystanders. This dynamic of carnivalesque, effervescent and transgressive upheaval was also expressed by a male interlocutor whom I interviewed in October 2010. He is typical of those with whom I spoke: he lives in a Chimoio bairro that only gained access to electricity in 2010; he is also part of a household in which most adults are under salaried or work as informal traders. However, he is atypical in being a police officer with the national force who joined the participants during the first day of the 2010 strike:

The strike [greve] was like a big party [festa grande]. You know, we are poor. And we have been for a long time – the end of the war changed nothing. The greves, they are like we are saying ‘Now they are afraid of us!’ It is a good feeling – they can now feel some of our fear. And we get to take something back from them – from those that are connected to the party, to the criminals, to the business. They fear us now.

There are several fascinating aspects to this account, the most striking being that as a police officer – far too often regarded as a mere state agent – he participated in the strike. The second significant aspect (in line with Mbembe's vision) is that the form of popular uprising can be seen as one of the only ways in which people can interact with what they identify as the post-colonial state: its agencies (the police, for instance), its wealthy elite and the objects and domains of its power (roads, infrastructure, telecommunications). Crucially, the strikes harboured and actualised the potential for confronting and appropriating aspects of Mozambican statehood and the riches of its domains, spaces and agents. A new form of egalitarian political collectivity emerged, informing (and informed by) memories and experiences of emancipation that were contained in what became a general expression: ‘Now they are afraid of us!’

In Chimoio, I first recorded this expression following the 2008 strike, which, as mentioned above, was successful in that it forced the Mozambican government to withdraw (if momentarily) from the announced price rises. Fuelled by the successes of 2008 and 2010, many of Chimoio's poorer inhabitants regard uprisings as a way of sending messages to the elite, which is seen as controlling the state and its mercantile and political affairs. The form and modality of the popular uprisings, however, transcend this dimension: they are informed by the Mozambican past and occur in a context of globalised connectivities and emerging lateral and anti-hierarchical forms of political protest.

Scripting the Strikes: Reappropriating Poder Popular and Global Inspiration

The directionality and symbolic action of the strikes as ‘sending messages to the elite’ are reflected in the ways in which the participants’ chants changed from 2008 to 2010. In both events, Frelimo was attacked by protesters using an inverted form of a song of praise that characterises past and present Mozambican political culture (Vail & White 1991). In 2008 in Chimoio, according to interlocutors interviewed, two chants were common: ‘Wanhanha kudurissa Frelimo’, which translates as ‘You are selling at prices that are too high, Frelimo’ and ‘Kudurissa noinda Frelimo’, which means, ‘Even the smallest things are expensive, Frelimo’.⁸ Significantly, both chants directly related experiences of escalating costs to the ruling party. Similar reasoning was evident in September 2010 but now the chant indicated historical depth: ‘Wadjaira kudurissa Frelimo’, which means, ‘You are now used to selling at prices that are too high, Frelimo’. In addition to the chants’ consistent addressee (Frelimo), the 2010 version contains a more pronounced threat in emphasising that Frelimo had still not learnt from the past.

There are two important aspects to these chants that require closer inspection here. Firstly, the rhetoric of ‘taking back’ and ‘sending messages’ is frequently associated with the notion of poder popular (popular power). Poder popular used to animate and fuel Mozambique's radical post-independence socialist politics under Samora Machel's presidency (1975–1986) but it fell into disuse by officials following the end of the civil war (1976–1992) and the subsequent neoliberal reforms of the country's political and economic system (Sumich 2013). It is worth noting, therefore, that the revolutionary rhetoric of the past is now regaining momentum outside officialdom in these contexts of political discontent and violence – as has also been noted, perhaps unsurprisingly, in relation to Nicaraguan and South African gangs (Jensen & Rodgers 2009).

Secondly, the chants during what is predominantly called o greve – ‘the strike’ – point to broader economic relations involving labour – and its current absence. During the early post-colonial period under socialist policies, labour was to be provided by the state through state-run enterprises and a state system of, for instance, seed shops, agricultural collectives and factories (Dinerman 2006). In the current post-socialist era of economic growth, continued poverty and rising prices, labour remains related to the state and its elite but it no longer involves o povo. The idiom of the strike is thus also a way of shutting down the business of the state, and the chants underline the conflict between the (relative and experienced) absence of labour and rising prices.

This second dimension – of labour and its absence – also points to an international dimension of the strike. It suggests that there has been a process of learning involved – of format, organisation and success – from 2008 to 2010. Many of those I spoke to in Chimoio in July–August 2011 and in Maputo in January 2013 pointed to large-scale transnational protest movements against the effects of marginalisation, unemployment or corruption. The events in England (see also Žižek 2011) mentioned above were often invoked by my Maputo and Chimoio interlocutors, for instance. Often with considerable glee, they underlined rhetorically that violent protest was not merely the prerogative of os negros (black people) but also carried out by white people. Put differently, for my interlocutors, the urban disturbances in England legitimised their participation in and the format of the Mozambican popular uprisings.

With recent movements, such as Occupy Wall Street or Los indignados, an argument may be made for a more comprehensive global ‘script’ of protest that is laterally/horizontally and rhizomatically organised against a range of ills and facilitated through various forms of digital interface, social media and viral messaging (see Juris 2012). However, while the new technologies’ insurrectionary potential emerged in Mozambique in the guise of viral messaging and live media coverage, the popular uprisings here entailed a far more direct confrontation with the state, its apparatus and its domains than many current North American and European protests. Through blocking the state's arteries (roads and roundabouts), attacking police stations and police officers, chanting slogans against the party in power and by engaging in looting that was regarded as a form of re-distribution, the strikes unsettled, at least momentarily, the apparatus and elite of the post-colonial state. Crucially, many interlocutors also made a direct connection to recent events in other African countries, and specifically to the violent Malawian protests of July 2011 (AFP 2011). However, although the protests in neighbouring Malawi were seen as inspiring, a young Chimoio friend working as a black market money changer pointed out a significant difference in an interview in August 2011:

In Malawi, they had leaders. These organized the protests. For us, there are no leaders. And we do not want leaders! For us … we are just o povo [the people]. We do it ourselves!

He was not alone: many I spoke to took pride in such self-organisation as being inspired by Samora Machel's poder popular (popular power), now fuelled by text messaging and mobile phone technology. Interestingly, the discarded politics of populist socialist rhetoric – such as o povo and poder popular – re-surfaced as a politico-cosmological framing for the protests and their organisation. Spontaneity, creativity and the immanent threat of the mass were celebrated here as a particular resource that may be mobilised (see also Buur 2009). Also historically (mid-August to mid-September 1991), the mobilisation of crowds – ostensibly in the service of poder popular – had escalated into an uncontainable force in Maputo when groups of people took over policing and lynched around 20 alleged criminals (Mozambiquefile 1991; see also Granjo 2008).

Rather than merely reflecting international protests or showing the potential of new digital technology, the recurring strikes highlight the nature of state authority in Mozambique as confronted by forces not only external to it but also by those internal to it through, for instance, the complicity of police officers and other security agents. From such a perspective, the very order of the Mozambican state is set against a range of alternative and shifting entities of which the rapidly gathering and dissolving urban crowd of the greves is one. Such confrontations between processes of state ordering and formations of political and social discontent beyond state control produce impermanent urban zones of state and non-state presence.

This unsettledness does not just reflect a long-term distinction between domains where state authority and citizenship rights apply and where they do not – as reflected in established debates on rural and urban forms of citizenship in Mozambique (O'Laughlin 2000) and in Africa more widely (Englund & Nyamnjoh 2004). Rather, the sheer number of protests across many urban African contexts over the last few years indicates that there is increasing discontent with both formulaic neoliberal reform and formalist forms of (urban) citizenship. Bond and Mottiar (2013: 289), for example, note that South Africa's ‘high level of social protest known as “popcorn protest” for its tendency to flare up and settle down immediately’ is increasing rapidly and that in 2011–2012 there were 1091 incidents reported as ‘unrest’. In a similar vein, Chance (2013) analyses the role of fire in recurring revolts and protests in South African townships, which signals a resurgence of ‘ungovernable collectivities [that] threaten sovereign power’.

In line with Comaroff and Comaroff's (2012; see also Simone 2004; Simone & Abouhani 2005; Berg 2010) argument about African metropolises forecasting global urban futures, I see such confrontations, uprisings and ‘unrest’ as manifesting global cities’ increasingly polycentric, nodal or hybrid form of governance impacting spatial ordering and security arrangements (Goldstein 2010; Paasche et al. 2013). Arguably, these developments entail growing urban domains of unsettledness, where police officers or private security guards, for instance, do not necessarily exercise state authority but actively produce confrontations and themselves engage in the popular uprisings.

At an analytical level, that various Mozambican agents of law and violence participated in these events adds another dimension to the existing spectral character of policing – as Benjamin (1986) and, later, Agamben (2000) have asserted – in which police officers are commonly believed to participate in bribery, assassinations, kidnappings or organised crime. However, for Chimoio (as noted above), the strikes and their participants assume additional significance because state representatives recently de facto allocated a sovereign authority to people in certain urban areas to exercise summary justice (Bertelsen 2009a). While such allocation may be increasingly common alongside a growing conjunction of development and security (Baker 2008; McNeish & Lie 2010), it nonetheless entails forms of mobilisation and political agency that generate other egalitarian collectivities, such as the 2008 and 2010 strikes.

Conclusion

Depending on one's perspective, the 2008 and 2010 popular uprisings may be seen as having been ‘successful’ in the sense of achieving immediate goals, such as barring increased prices for basic commodities and services. However, on an analytical level, the uprisings generated novel collectivities and an ephemeral political authority that was neither imagined nor desired by Mozambican state agents. It is difficult to ascertain what the long-term effects of such recurring encounters between rioters and state agents will be in terms of state stability, police accountability, Frelimo's future and the potential for the political opposition to capitalise on popular discontent.

However, events in 2013 and 2014 seem to indicate that memories of the large-scale strikes in 2008 and 2010 are significant, as a number of further protests have emerged across the country: for instance, there were spontaneous violent protests in Beira after it was rumoured that the Mozambican military were forcibly drafting young people into its armed ranks to be deployed in increasing military confrontations with Renamo in 2013 and 2014 (AIM 2013a; Mediafax 2014). In addition, numerous other events – such as a violent clash between a group of opposition supporters and the Mozambican riot police and Frelimo supporters (AIM 2013b) during the recent election campaign – indicate an on-going effect (both in terms of memory and as a form of mobilisation) of the 2008 and 2010 strikes.

Beyond its effect on Mozambican city streets, it is clear that these two large-scale popular uprisings have destabilised the image of a powerful sovereign state under the control of Frelimo, which has at times lost control of urban spaces in both the capital Maputo and in other large cities such as Chimoio. Therefore, as anthropologists concerned with notions of sovereignty in Africa, we should neither confine our research to the centric contexts of capitals nor should we only analyse its elites – whether these be government officials or development bureaucrats. Rather, anthropology should also recognise both the popular uprisings’ dimensions of effervescence, creativity and feast and the participants’ exertion of a directed and highly egalitarian political form of action through the occupation of state spaces and attacks on police stations. The notions of effervescence and ephemerality that characterised the popular uprisings are key aspects of a form of post-colonial politics as it unfolds in contexts such as Mozambique – a form that is irreducible to the orderly world of representation (through NGOs, trade unions and political parties, for example) or to well-orchestrated participation in public discourse (through mass media in its various forms).

This means that we can neither uphold an image of the post-colonial subject in Mozambique as solely interested in his/her wellbeing nor claim that the composite nature of a one-party state embroiled in development politics and elite accumulation has somehow subdued Mozambicans – or embroiled the rulers and the ruled in a mutually disempowering relationship. This is not to say that poor people in the bairros are generally optimistic: quite the opposite is the case: cynicism reigns and they express little hope for themselves in terms of economic development. Despite this outlook, however, politics is followed closely, commented upon and critiqued, and the fact that people also engage in risky direct action – such as the strikes – testifies to such political engagement. More broadly, the case of the popular uprisings also signals that anthropology should be careful about reproducing the official rhetoric of ‘multiparty democracy’ or about analytically conforming to the technicalities of procedural approaches to citizenship, politics and democracy. The strikes are multisemic, festive and carnivalesque events that ephemerally unsettle and appropriate elements of post-colonial state sovereignty; they are instances of the kinds of non-conforming political engagements with which anthropology should be concerned (see also Obadare & Willems 2014).

To return to this article's starting point, Bayart's notion of the post-colonial ‘rhizome state’ does not seem to have universal application to post-colonial states such as Mozambique. Here, rhizomic capacities inherent in the popular uprisings present themselves as unmoored and have the capacity to disorder and confront state spaces, state agents and domains of wealth. In these contexts, rhizomic networks do not, as they do in Bayart's analysis, link the low and high in multiple symbolic and material ways to comprise a ‘mesh of networks’ that is controlled by and constitute the post-colonial state.

Bayart himself notes that ‘[w]e are also “tired of the tree”, of this arboreal metaphor of the state which, in truth, has exhausted the theoreticians. Our time would be better spent trying to understand mysteries of the rhizome’ (Bayart 1993: 221). Indeed, but the ‘mysteries of the rhizome’ involve wholly novel forms of engagement in politics and democracy in Mozambique and beyond, and these forms have a horizon beyond that of a predatory state resting on a ‘mesh of networks’. That is to say, the politics of the Mozambican strikes are of open and experimental kinds – replete with egalitarian aspirations and exuberance that are not contained in the bleak imagery of the post-colonial political order often circulated by, for instance, Bayart. While the present reading is optimistic – and sustained by more recent new forms of protest in Mozambique in 2012, 2013 and 2014 – it also corresponds to what is said by those who took part in or sympathised with the popular uprisings in Chimoio: ‘Now they are afraid of us’.

Acknowledgements

This text has been long in the making and I would like, first of all, to thank Nils Ole Bubandt and the two anonymous readers for important and highly useful critical comments. I also would like to thank Jean and John Comaroff for letting me present an earlier draft of this article at the Africa Workshop at Harvard University in October 2013. I am grateful to both of you as well as Achille Mbembe, Kerry Chance, Lauren Coyle, Claudia Gastrow, Pauline Peters, Lucie White, Jorge Kmpox and all the others who took the time to read, comment and discuss the text while at Harvard. I am grateful also to comments I received when presenting earlier incarnations of the article at the Department of Social Anthropology, University of Oslo (invited by Rune Flikke), the Department of Anthropology and Archaeology, University of Tromsø (invited by Ole-Bjørn Fossbakk) and, finally, the Department of Anthropology, Stockholm University (invited by Staffan Löfving). My gratitude is also extended to Carmeliza Rosário, Inge Tvedten, Hege Toje, Ingrid Samset, Jason Sumich, Ørnulf Gulbrandsen, Kjetil Fosshagen and Bruce Kapferer who have all, at some points, read and commented this work. All flaws and faults are, of course, my own.

Notes

1. A near similar translation by Karen E. Fields from the French original is provided in Durkheim (1995 [1912]: 212f).

2. Forwarded to me by a Mozambican friend, the original (including typos and misspellings) reads: ‘Mocambicano prepara-te no grande dia de greve 01/09/10. Revendicamos sobre a subida de precos J e energia, àgua, arroz, xapa e päo. Envia pra outros Mocambicanos.’ While various text messages were disseminated, they all indicated 1 September 2010 as a strike day against price hikes.

3. Greve participants generally conform to Rudé’s (2005 [1964]: 4) notion of ‘crowd’, which includes ‘political demonstrations and what sociologists have termed “the aggressive mob” or the “hostile outburst” – to such activities as strikes, riots, rebellions, insurrections, and revolutions’. I follow Mazzarella's (2010) work in seeing the notion of the crowd not in the pathologising sense famously expressed by Gustave LeBon but by viewing these as harbouring great generative potential and being ‘anarchically unruly’.

4. A term such as ‘riot’ is problematic in that it is commonly employed in imprecise and/or state-centric ways (see also Ginty 2004). Reading this material, it becomes clear that the terms ‘strike’ and ‘popular uprising’ better capture the empirical complexity and the politics of the events in question that are in Mozambique often called greves.

5. Given the politically sensitive issues around the strikes, all persons quoted in this article are un-named to protect their confidentiality.

6. While media speculation has been rife about the circumstances of the first text message, neither a ‘first sender’ nor allegations of mass distributions have been verified.

7. Shangaan is the dominant language in Southern Mozambique.

8. The dominant language in the Chimoio area is chiTewe, and key works from the Manica region have been consulted in order to form the spellings given here.