‘Violent protests’ in South Africa: understanding service delivery protests

ABSTRACT

Violence is a crucial feature associated with service delivery protests in South Africa. The media and scholars have often referred to the rising violence trend in service delivery protests. However, the definition of violent protests is too broad; it fails to paint a correct picture of the violence. Previously, the general tendency was to classify these protests as either peaceful or violent – a simple dichotomy. Therefore, scholars have developed the 3-way formulation of protests as ‘orderly’, ‘disruptive’ and ‘violent’. Although the three-way formulation is the best so far, it conflates damage to property and injury to people as ‘violent protests’. Damage to property, however bad, should not be bracketed together with injury to people. Drawing on qualitative data from low-income communities in Cape Town, South Africa, I consider deliberately vandalising property as ‘vandalistic’ protests and attacks on persons as ‘violent’ protests. Building on the three-fold formulation of service delivery protests, I introduce a new category – the vandalistic protests. I, therefore, argue for a fourfold formulation of protests as ‘orderly’, ‘disruptive’, vandalistic and ‘violent’, which is often the normal order protests evolve. This analysis highlights the need for authorities to swiftly address communities’ grievances to avoid more radical protest tactics.

Introduction and background

South Africa has experienced frequent protests post-1994. Although such protests dotted the South African socio-political landscape from the 1990s, from 2004 (ten years into democracy) South Africa witnessed a steady increase that reached sustained levels in 2009 (Municipal IQ 2017; Runciman et al. 2016). A key feature of these protests has been to characterise them as violent.

However, what is meant by violent protests? In an article titled ‘Violence and democracy in South Africa’s community protests’, Paret (2015) forcefully deconstructed the ambiguous nature of the term violent protest. In the article, he notes the conflation of meanings that characterise the term violence. For example, violence is done by different actors – e.g. the police and protesters – and violence has different meanings ranging from social disruption, damage to property and personal attacks. Drawing from media outlets, he noted the diverse usage of the term. Firstly, the term may be used without any specific meaning or action. In this case, the readers must come up with their meaning of violence. Secondly, violence is sometimes used to refer to actions by protesters – for example, when actors vandalise property. The word violence is even used in reference to ‘social disruption’ (equivalent to Alexander et al.’s 2018 ‘disruptive protests’) where people barricade roads and burn tyres. Thirdly, the phrase violent protests is sometimes used about police brutality; often, such violence is not explicitly attributed to the police which creates an impression that the violence was at the hands of the protesters. Finally, the word violence can refer to protesters hurling stones at motorists, attacking other community members or intimidating foreigners who own shops; arrests for public violence and looting foreign-owned shops. Such ambiguities prompted Paret (2015) to advise scholars to avoid the term violence and search for analytical concepts which are more precise. I take up Paret’s (2015) challenge in the analysis to develop a more accurate term that best captures the dynamics of protests in Gugulethu and Khayelitsha.

In this paper, I rethink the term ‘violence’ and argue that it is conflated and often confusing and needs to be clearly defined and consistently applied. My analysis builds on Alexander et al.’s (2018) 3-way categorisation of protests as ‘orderly’, ‘disruptive’ and ‘violent’. While appreciating the value of the 3-way formulation, attacks on property and injury to people should not be put in the same category (violent protests). Doing so obfuscates the different protest phases and tactics. This reinforces the vilification and demonisation of protests and protesters. I argue that there needs to be a distinction between ‘vandalistic’ and ‘violent’ protests. I regard a deliberate attack on public or private property as ‘vandalistic’ protests and an attack or harm on people as ‘violent’ protests. The term ‘vandalistic’ – an adjective for the noun ‘vandalism’ – is appropriate especially given that vandalism is the action which involves deliberately damaging private or public property. Thus, instead of a three-way categorisation, I propose a four-way categorisation of protests: ‘orderly’, ‘disruptive’, ‘vandalistic’ and ‘violent’, which is often the normal order protests progress.

My approach is guided by a qualitative excavation of marginalised voices, allowing me to give voice to my participants instead of voicing-over their reasons. In other words, having unearthed local realities of service delivery protests in Gugulethu and Khayelitsha (Cape Town), I present these realities from participants’ own contexts and points of view. The paper is organised into five sections. The first has provided the introduction and background of the study; the second briefly considers service delivery protests and understandings of violence in South Africa. The third section provides the methodology that guided this inquiry. The fourth presents results, analysis and discussion. The fifth section concludes the paper.

Understanding service delivery protests

South Africa has witnessed escalating frequency and scale of dramatic service delivery protests (Alexander et al. 2018; Allan and Heese 2011; Booysen 2009; Chigwata, O’Donovan, and Powell 2017; Municipal IQ 2017; 2014). South Africa’s protest tactics have included boycotts, marches, submission of memoranda of grievances, mass community meetings, barricading roads, burning tyres, toyi-toying, street fights with the police, burning of municipal buildings and councillors’ houses, and land occupations (Alexander 2010; Tshishonga 2015; von Holdt et al. 2011). Mainstream media (newspapers, television and radio news) often refer to the rising tendency of service delivery protests (SDPs) to become violent, although the accuracy of such reports is questionable.

The mainstream media cover stories based on the size of the protest, the dramatic nature of protests, including the novelty of tactics, geographical location (main towns are prioritised over local protests) and the presence of ‘violence’ in the protests (Koopmans 2004; Zeurn 2011). This selectivity imbued in media reporting affects the number and nature of reported protests. The effect of the media’s selectivity is that the public often views protests in a negative light. Indeed, the public usually perceives protesters as violent beings without justifiable reasons who threaten stability and democracy (Pithouse 2011; Zeurn 2011). While mainstream media’s hype about violence is understandable, given the desire to sell newsworthy stories, this is often misleading because more orderly and ‘peaceful’ protests do not receive the same coverage as ‘violent’ protests.

Similar to the media’s negative portrayal of SDPs, the government and the police have tended to view protesters negatively, labelling some of the protests as acts by hooligans and criminal elements. Some government officials have even deployed the ‘Third Force’ explanations to the protests. This relates to the African National Congress’ explanations during the apartheid era, in which the organisation explained a spate of violence in townships as acts by undercover apartheid forces. In the post-apartheid era, the ‘Third Force’ explanation attributes the protests as interference by foreign intelligence agencies working in cahoots with white intellectuals and opposition parties to destabilise South Africa’s democracy (Auyero 2012; Butler 2013; Pithouse 2013; wa Azania 2014).

The government’s and police’s view of protests has affected their relationship and interaction with protesters. There is arguably animosity between the police and protesters, with the police sometimes emerging as the perpetrators of violence. Indeed, some commentators have bemoaned brutal police tendencies in some protests (Alexander 2010; Bond 2011). In fact, there is emerging evidence which shows that police aggression has often led to violent protests (Duncan 2014). I argue that scepticism about the alleged existence of a cabal fixated on destabilising South Africa’s democracy should not blur the government’s, police’s, scholars’ and development practitioners’ image of the real and justifiable service delivery problems people demand.

Even scholars have stressed the rising tendency of violence in protests (Chigwata, O’Donovan, and Powell 2017; Jain 2010; Municipal IQ 2014). For example, the Municipal IQ (2014) notes that since 2004, municipalities have been targets of violent protests in South Africa, with a 77% peak in 2012 and 67% lowest in 2011. The Civic Protest Barometer notes that since 2013, 90% of protests involved some element of violence or intimidation from the authorities or protesters (where violence means protests where at least some of the participants engaged in activities that posed a threat and or harmed property or people) (Chigwata, O’Donovan, and Powell 2017). Scholars often define violence loosely, including the destruction of buildings or the blockage of roads, looting, intimidation, personal attack, and arson. This definition of violent protests is too broad; it fails to paint a correct picture of the violence.

The obscurities I have discussed are exacerbated by the sources of data that scholars depend on; the sources have different focuses and objectives, which impact the scale and nature of protests reported and analysed. One of these sources is the Incident Registration Information System (IRIS), which is maintained by the South African Police Service (SAPS) Crime Combating Operations’ Visible Policing Unit (VPU). The problem with this data is that it records ‘crowd incidents’, which include social occasions and sporting events and not just protests. Another data source, the Municipal IQ’s Hotspots Monitor, is helpful but only records ‘major’ service delivery protests and leaves out ‘minor’ ones. This is understandable given that the Municipal IQ relies on media reports. As discussed, mainstream media’s coverage of newsworthy protests results in preferential reporting of dramatic, ‘disruptive’, ‘vandalistic’ and ‘violent’ protests over ‘orderly’ protests. Unsurprisingly, municipal records show a picture that is different from that portrayed by newspapers; municipal records show that most protests unfold peacefully (Duncan 2016; Zeurn 2011). In other words, although ‘violence’ (with its varied meanings) has characterised some protests, the media and scholars have often blown this out of proportion.

The three-way categorisation is helpful because it rejects the simplistic dichotomising of protests as ‘peaceful’ and ‘violent’. Instead, it distinguishes between ‘order’ from ‘disorder’ and ‘peaceful’ from ‘disorder’ (Alexander et al. 2018). In this formulation, every violent protest is disorderly but not every disorderly protest is violent. That is, some disruptive protests are peaceful. The three-way categorisation sees violent protests as those which involve damage to other people’s property and injury of people. Orderly protests are contestations tolerated by the government apparatus and often negotiated beforehand such as public meetings, pickets and marches. Disruptive tactics employ campaigns, including barricading roads often with rocks and burning tyres. Alexander et al. (2018) concluded that although community protests have become more frequent, disruptive and violent (what they shorten as disorder), protests reached a high point in 2012. Since then, disorderly protests have somewhat flattened off.

Although the three-way formulation is the best so far, it conflates damage to property and injury to people. This, I argue, obscures the term violent protests. Damage to property, however bad, should not be bracketed together with injury to people. As proponents of the three-way categorisation rightly note, ‘[t]reating barricades as ‘violent’ delegitimises the intentions of the protesters and misconstrues the dynamics of protest’ (Alexander et al. 2018, 31). Similarly, I argue that treating damage to a water metre or electricity metre as ‘violent’ obfuscates the intentions and rationales behind such tactics and reinforces negative connotations attached to protesters.

Research methodology

This inquiry deployed qualitative methodology; this best suited this study whose main concern was on ‘depth’ rather than the ‘breath’ of understanding (Blaxter 2001). This methodology helped excavate participants’ subjective responses and deepen my understanding of people’s experiences (Silverman 2010). This enabled me to appreciate the distinctiveness of the tactics protesters deploy at different times and places in constestations.

The study focused Gugulethu and Khayelitsha, two of the protest prone neighbourhoods in Cape Town. Participants in Gugulethu came from Wards 38, 39, 40, 41 and 44 while informants from Khayelitsha, were drawn from Ward 18 (Site C), 92 (Site B), 95, 96 and 98. I conducted 20 in-depth interviews in Gugulethu and 20 in-depth interviews in Khayelitsha. I augmented the data with 4 focus group discussions (2 in Gugulethu and 2 in Khayelitsha). Data collection commenced upon receipt of ethical clearance; all ethical standards were upheld. This paper uses pseudonyms to protect participants’ identity.

Decisions on protest tactics

The findings from the qualitative data excavated in Gugulethu and Khayelitsha reveal that participants believe that, among other things, the level of anger, frustration, the ineffectiveness of prior tactics and the urgency with which a grievance must be addressed determine the tactic protesters employ. For example, in describing the choice of a protest tactic, Unati, a man who has helped burn tyres on the day of protests in Khayelitsha, stated:

There is a belief that when things are being destroyed the government will quickly answer. It’s an idea of fast-tracking the process, believing that it will do something for the government to respond. So, either burning tyres or putting a container on the road will fast-track the process.

Unati’s comments are consistent with the prevalent belief that disruptive and violent protests receive quicker responses (Alexander 2010; Paret 2018). This belief incentivises residents to engage in collective action by affecting expectations of success or failure (Tarrow 1994). Given this view, radical activism is meant to fast-tract development in the community. Moreover, this view suggests that the government must be coerced to deliver to poor communities. Understandably, the government has limited resources at its disposal; at the same time, impoverished communities have pressing needs which would have been raised repeatedly but to no avail.

In both Gugulethu and Khayelitsha, some sub-groups in the protests decided to employ disruptive and other dramatic protests. While radical forms of protests usually start owing to their resonance with some of the movement’s actors (della Porta 1995), in Khayelitsha, similar tactics were employed by protestors who do not necessarily support ‘disruptive’, ‘vandalistic’ and ‘violent’ protests. Several particularly elderly people did not support ‘disruptive’ and ‘violent’ protests but took exception when it came to speeding up the development process. For example, after Mfundisi (pastor) Zama distinguished between peaceful and violent protests, I asked him whether he thought there were times violence was important in a protest. He responded:

I will not say violence is important, but we have seen that when we do violence, [the] government responds, but when we just don’t [protest violently], then they don’t [respond]. You know I think, according to that context, I will nod my head two times to violence.

Asked whether the destruction of public goods does not deprive the community of the services they already have, Thembeka a female activist who illegally reconnected her electricity in Khayelitsha stated:

Some people think about it but there are dominant voices. And sometimes, the guys who are thieves and tsotsis in the community have some influence of some sort [take advantage of the situation] and suddenly the direction takes another tone which is not a good one which ends up damaging the same things that we need in the future.

There is no uniformity in protests; some protesters are against some protest tactics but (i) dominant voices win – which is consistent with the view that radical forms of protests often start due to their resonance with at least some of the movement’s actors (della Porta 1995); and at other times, (ii) thieves and tsotsis take advantage of the protests to engage in criminal activities. A similar sentiment was advanced by an elderly man – Tata Xolani – in Gugulethu, who stated that at times protests are ‘hijacked’. There is a clear understanding that damaging the things the community needs in the future is counterproductive. Still, such vandalism is sometimes deemed necessary to achieve a greater good as the pastor above explained.

Some tactics and strategies may work well to achieve one aim but may conflict with the necessary behaviour desired to achieve another (McCarthy and Zald 1977). Understanding this is important because even protests are used as a last resort, just as violent protests are used as a last resort after, ‘orderly’, ‘disruptive’, and ‘vandalistic’ protests have failed. Moreover, Thembeka’s idea that at times protests take a different turn or are hijacked suggests that protests sometimes deviate from repertoires planned before the protests, which shows that protests are not static; different repertoires combine in other places or moments in time.

Disruptive and vandalistic tactics

A key feature of South Africa’s protests is erecting barricades on the road and attacking public and private property. Many protesters believe disruptive tactics (barricading roads and burning tyres) and – what I call – ‘vandalistic protests’ (destroying properties) are the only way to attract the government’s attention (Alexander 2010; Paret 2015; von Holdt et al. 2011). Traditionally, tactics that include road blockages, tyre burning, property destruction and attack on people were grouped under ‘violent’ protests, especially in media. Paret (2015) has warned analysts of using the concept of violence uncritically. He aptly highlights the conflation of meanings imbued in the term violence, noting that violence is carried out by different actors – such as the police and the protesters – and violence refers to diverse actions, namely social disruption, destruction of property and personal attacks.

In Khayelitsha, a male protester, Uuka, who sees the value of protests, described the disruptive protests they engage in on the day of protests:

We also take rubbish bins and scatter them on the roads. There are green containers that we have here. There is an area here where they don’t have a formal rubbish system - that container is used to burn rubbish. So, we will take such containers and close the roads. You see that 6X3 it’s a container. We will take it and close the road and burn tyres on the road.

Public goods such as roads, and rubbish bins are strategically utilised in the fight for development in the impoverished community. Barricading roads is meant to disturb traffic flow and temporarily hold ‘progress’ for the greater good of airing grievances. Barricaded streets bring delays, inconveniences, confusion, frustration, and chaos to commuters. Similarly, in Cato Manor, Durban, Mottiar (2014) found that obstructing main roads and burning tyres was meant to create chaos, thereby publicising grievances and sending a message. Arguably, this tactic is meant to say that our lives in impoverished townships are stagnant and inconvenienced daily; this tactic is intended to inconvenience the commuters and thereby exert pressure on the government to deliver.

Another crucial characteristic of South Africa’s community protests is the deliberate destruction of property, what I have called ‘vandalistic’ protests – to distinguish them from ‘disruptive’ and ‘violent’ protests. For example, Siya, a protest leader in Gugulethu, lamented the slow pace of housing delivery as ‘a slow-moving train’. His words best capture the vandalistic protests: ‘Development in Gugulethu is like a slow-moving train; it is a train that delays, and when it delays – people get angry. They end up destroying the railway line’. Here, Siya equates development to a scheduled train, which, when delayed, infuriates people, as anyone would. To strategically bring their grievances to authorities and the public, activists destroy elements of that development or any other public building or property. For example, participants in Gugulethu recall how they deliberately vandalised water metres as an act of defiance and to send a message about water restrictions (I will revisit the contestations around water metres in the next section). Unathi’s earlier quote bears repeating because it underlines the belief behind vandalistic protests:

There is a belief that the government will quickly answer when things are being destroyed. It’s an idea of fast-tracking the process, believing it will do something for the government to respond.

An analysis of what would usually be called violence in Unathi’s explanation reveals what Alexander et al. (2018) called ‘disruptive’ protests (burning tyres and blockading the road), and the deliberate destruction of property is what I have called ‘vandalistic’ protests. Conflating deliberate destruction of property with injury to people obscures the message protesters seek to send. The strength of the four-fold categorisation of protests that I advance in this paper over the three-way categorisation of protests by Alexander et al. (2018) can be found in Unathi’s deliberate omission of injury to people. In such cases, to regard destruction of property as ‘violent’ is as misleading as relating to barricades of roads as violent.

Vandalistic protests are also evident in protesters’ distinction between protests/picketing and striking. Participants stated that a protest/picket involves peaceful demonstration where a memorandum [of demands] is handed to officials and a strike involves a forceful demonstration. Letlhogonolo, a university student who stays in Khayelitsha explained the difference:

During a protest we have marshals – from the community – who make sure that people are not doing illegal things. The marshals will be there if it is a picketing but if it is a strike, we never have marshals because it will be chaos. And during the strike there is nothing else that we do except all vandalising. We don’t talk. There is no time to talk.

… We are looking forward, that if we are going to the police station or the councillor, we are going to vandalise his office, whatever we meet along the way. There is nothing that is happening. The only thing that is happening is that we are going to that guy [councillor] and we are going to vandalise so that they see that we are really angry.

Remarkably, Letlhogonolo does not mention personal attack on individuals. His reference to ‘going to that guy’ (councillor) is not to assault him up but ‘going to vandalise’ the councillor’s property. Vandalistic protests as described in this extract, are informed by history. Asked why protesters sometimes destroy or burn buildings, Lerumo, who recently completed university and stays in Gugulethu stated that vandalistic protests are informed by history:

History has taught us and continues to teach us that from the apartheid until today, the government will never listen to you and respond very well according to what we want until we stand up and show them that we are outraged now. Eh … if you see a strike, you must know that vandalism will start. There are no second thoughts. In strike we say we will vandalise at 3:00; as soon as a strike begins, expect vandalism to start soon.

These two extracts show that there are two different rules of engagement in ‘protests’ and ‘strikes’. In protests, activists have ‘marshals’ who ensure that protests unfold as planned and no one engages in illegal activities. Sticking to the agreed-upon repertoire helps avoid clouding the message of the protest with other unplanned activities. Marshalls, however, are only necessary when it is an orderly protest. If it is a vandalistic protest, protesters seek to cause ‘chaos’ to show the extent of their anger.

What is apparent in these extracts is that there is a degree of organisation in assigning marshalls who help protesters adhere to the ‘agreed upon’ repertoires. When repeated ‘orderly’ efforts to air a view fail to yield results, protesters plan a ‘strike’ to express the extent of their anger. Such transition from ‘orderly’ and ‘disruptive’ protests to ‘vandalistic’ protests is informed by history lessons that people reflect and act upon. Likewise, Paret (2015, 115) found that his participants distinguished a ‘march’ from a ‘strike’; a march was viewed as ‘a relatively benign and entirely legal affair’ and a strike as ‘illegal and involves the destruction of property’.

Protesters have at times torched ‘symbolically significant buildings (e.g. council offices and councillors’ houses)’ (Alexander 2010, 32). While this is sometimes the case, I found that in Gugulethu and Khayelitsha, as elsewhere, there are times people do not target symbolically significant buildings. For example, when people burn libraries, it is often not because the activists find fault with the library and demand a better library. Rather, it is merely a call to be heard. This is similar to instances where protesters vandalise private properties, e.g. cars. Future analyses should consider what informs the decisions to attack either private and/or public property and the implications of such decisions. As I have argued, it would be wrong to regard a deliberate attack on a library or cars and injury to persons under one term, violent protests. In explaining the need for disruptive and vandalistic tactics, a pastor in Khayelitsha, Mfundisi Akhona used a compelling metaphor:

Actually, … to make someone to respond, pain must be there, I believe. You know, so that is the issue with our government. Our government knows that there is a pain somewhere. If there is a sore in your arm then you are turning a blind eye to it but if someone starts to pinch that sore arm you will feel the pain and you will want to do something, so I think that’s what happens. They [government officials] have to respond (emphasis added).

It was surprising to hear a pastor openly admitting to supporting disruptive and vandalistic tactics in the presence of his colleagues. Yet Mfundisi Akhona’s support of disruptive and vandalistic protests should be seen as a last resort to make the government listen and address the inconveniences in impoverished townships, which they believe the government knows but ignores. As Larana (1994) notes, activists blame certain people or institutions they believe are responsible for the problems they face, in this case – the government. Activists’ actions in their ‘interpretative work’ (Snow and Benford 1988) are well calculated hence the ‘pinching’ to inflict pain (disruptive and vandalistic tactics) to solicit a quick response from the state, that is, to address the otherwise ignored ‘sore arm’ (poor conditions). What seems to be a mindless spontaneous activity has a level of thought and plan, to determine the sore and feasible arm they can find to inflict pain proportionate to the objectives of a protest event, which activists can explain well after the event. It is interesting that activists believe the government knows the problems exist but ignores them. Indeed, activist-scholar, Trevor Ngwane, explained that before communities protest, they usually exhaust other peaceful avenues, and only resort to more radical tactics when these processes fail; he remarked that ‘[w]hen people start hitting the streets, they should have a banner saying ‘All protocols observed’, because they’ve gone through all the channels’, (van Schie 2014, np). Thus, pinching, by means of disruptive, vandalistic and even violent protests, helps remind authorities and onlookers of the ignored sole arm/service delivery problems. This calls for swiftness in addressing communities’ grievances to discourage more radical protest tactics.

Violent tactics

While it is true that the categorisation of protests as being violent is ambiguous, and that critical scholars should opt for more precise concepts that capture the dynamics of protests (Paret 2015), this does not mean people have not been harmed or targeted in protests. There have been incidents where unpopular people have been targeted and hounded out of communities. At other times, this has been acts by a few individuals engaged in criminal elements who ‘hijack’ the protest for their ends. However, media outlets have blown the ‘violent’ protest descriptor out of proportion.

In their desire to sell newsworthy stories, media outlets have over-reported so-called violent protests and under-reported orderly protests. In so doing, ‘journalists continue to repeat the chime ‘violent service delivery protests’ even in cases where it is not warranted’ (Duncan 2014, np). Media outlets’ lumping up burning tyres, destruction of property and throwing stones at motorists, and police brutality as violence ‘enables the state to raise alarm, demonise protesters and justify repressive responses’ (Paret 2015, 121). For example, the then Secretary General of the ANC, Gwede Mantashe, once said, ‘by participating in violent protests, you are actively destroying your right to protest because you are inviting the police to be there, while it is not their place’ and went on to direct the ‘state to find ways and implementable means as a matter of urgency to deal with the twin phenomenon of violent strikes and violent community protests’ (Burger 2013, np). Following Mantashe’s speech, Ngoako Ramathlodi, the then head of the ANC elections committees, stated that the government would deploy an ‘iron fist’ to quell what he called the ‘seas of anarchy’ in violent strikes and service delivery protests (Burger 2013). Here, the police are viewed as the solution and never the perpetrators of violence.

Evidence from a hostage situation in Gugulethu reveals that not all protesters are bent on engaging in violent protests (harming people). In this hostage, a White man and his contracted company’s employees were installing water metres. Important here is that some protesters, as Bulelani, a protest leader recounted, wanted to ‘beat him up a little … to make an example out of him’, but ‘we [the leaders] said no, let’s just hold him here’. Indeed, the lives of the contractor and his team were in danger, but leaders protected them to avoid physical harm to the contractor. Interestingly, the decision to only hold the contractor hostage occurred after the earlier tactic to ‘vandalise’ the water metres did not yield fruit. Solly, a well-connected leader in and outside Gugulethu who fought in the anti-apartheid struggle, narrated what transpired:

They came [Contractors] and they implemented it [the new water meters]. We just saw private company cars coming to our yard [and] [d]igging holes and put[ting] up this blue thing. Now what is this here, we asked [in an astonished voice] … maybe you are planting a time bomb into my yard … Fine, it went on.

Later, the installed water metres created problems. Then, in a focus group, Chipa – a well-respected leader in the community – remembered that:

Before we knew it, people started to come to me, saying … we do not have water. I asked, you don’t have water? What do you mean? People responded ‘no, there is no water and yet some of our neighbours have … ’ Then we [protesters] said, okay, we called the officials saying, so and so doesn’t have water. They [the municipal officials] said, ‘no it’s because you have a new system, she used up all the water that she was provided with.’ ‘What do you mean, used up all the water’, we asked. ‘Well, there is a new system’ [officials responded]. Then we said, okay fine. Hold it right there. Is this the new system that the mayor was talking about? […] Then we said [amongst ourselves] we need to do something here [in a serious voice]. We said to them, come here. They [did] not want to come. We need to go to them. But who did they come to, when they installed [water meters]? We took action [in a stern voice]. We started a protest. We vandalised that [the water meters] […]. We even took some [of] the water meters. Now it’s going to cost the city millions of Rands.

In the same focus group, Zakora explained the transition from the vandalistic protests to the hostage-taking:

They didn’t feel the pain [when we vandalised water meters] but when we caught up with the contractor who was contracted to install these things we took them in for hostage. We said to him come here. We told him: Now, you call the person that gave you the tender. And the person is the City [the City of Cape Town]. Call the person that gave you the tender. That person came. We had said to him, if he or she does not come, you can just as well call your family and tell your family that you are not coming home tonight until those people [affected people in Gugulethu] have their water back! They came! [Following the call, the municipal officials came].

If one is to think of the hostage situation as a ‘violent’ protest given the White contractor and his team's danger, then the protests moved from ‘negotiation’ to ‘vandalistic’ to ‘violent’ protest for the water to be restored. The move to vandalistic protests only happened when officials did not come to resolve the water dilemma through ‘negotiation’. When the municipality did not feel the pain of vandalistic protests and restored water, residents, in their bid to have their water back, took in the contractor hostage. Yet even in the hostage situation, leaders barred individuals who wanted to ‘beat up’ the contractors and make an example. Thus, conflating the destruction of water metres and hostage or beating up the contactor as all violent protests would be incorrect. Indeed, the concept of violence is empty and ambiguous and does a disservice to protesters.

Conclusion

The general consensus has been to classify protests as peaceful or violent with research showing a tendency of protests to progress from peaceful to violent protests to meet community’s demands. Violence has to be understood as a means of engagement, the last resort by the poor who seek to be understood and served (von Holdt et al. 2011). Building on Alexander et al. (2018) who provide a three-way categorisation of protests as ‘orderly’, ‘disruptive’ and ‘violent’, I find the violent categorisation of protests as misleading. This is because it converges damage to property and injury to people as violent protests. I propose a fourth categorisation – ‘vandalistic’ – to refer to deliberate damage to private or public property. This addition helps see the progression of protests from (i) ‘orderly’ – to refer to contestations that are tolerated by the government such as public meetings, to (ii) ‘disruptive’ where the normal day-to-day business is affected but there is no damage to property and injury to people, e.g. barricading roads and burning tyres, to (iii) ‘vandalistic’ where private and public property is deliberately attacked, and to (iv) ‘violent’ protests where people are attacked. This allows me to represent the rationale behind the different kinds of protests rightly and argue that as long as there is no significant urban development in peripheral areas, dramatic protests will likely continue in Gugulethu and Khayelitsha.

Violence in protests is complex and multifaceted and should be treated as such. However, the unbridled focus on ‘disruptive’, ‘vandalistic’ and ‘violent’ protests, chiefly by mainstream media, often deflects people from the real justifiable reasons for protest such as service delivery-related problems. This is vital in media selectivity where ‘disruptive’, ‘vandalistic’ and ‘violent’ protests are over-reported and even exaggerated while smaller-sized, less-dramatic ‘orderly’ protests in impoverished and peripheral areas such as Gugulethu and Khayelitsha are marginalised, underreported and ignored. Although the four-fold typology is not mutually exclusive, it can help reveal trends and how the different repertoires combine in different places or moments in time.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.