Taking to the streets: Who is protesting and why?

Who protests? Why do people protest? Has there been an increase in the number of service delivery demonstrations in the past few years? And who are the people protesting about service delivery? Data collected from seven municipalities was used to answer these four questions. 

The data, which are records of notifications sent by organisations intending to hold gatherings and protests in the past five years, was collected by a team led by Jane Duncan of Rhodes University's school of journalism and media studies.

Who protests the most?

In five of the seven municipalities analysed – Nelson Mandela Metro (Port Elizabeth, Eastern Cape), Johannesburg Metro (Gauteng), eThekwini Metro (Durban, KwaZulu-Natal), Ehlanzeni (Nelspruit, Mpumalanga) and Blue Crane (Somerset East, Eastern Cape) – unions made up the group that submitted the most protest notifications to the local authorities.

In Makana (Grahamstown) and Lukhanji (Queenstown), both municipalities in the Eastern Cape, unions were pushed to second and third place respectively.

Most of the notifications submitted by unions did not give a specific reason for the protests – often it was simply to hand over a memorandum with no other details, which were classified as unspecified. Protest information that included reasons were mainly about wage disputes, protests against labour brokers, and various other labour-related issues.

What are people protesting about?

In all but one of the municipalities, issues related to crime and the justice system were the most common reasons people gave for wanting to protest.

Crime and the justice system were grouped together in one category for the purposes of analysing the data because the stated intention of many of the protests was not just to protest against crime in general, but because people also wanted to hand over memorandums at magistrate's courts asking that bail not be granted to alleged criminals, or they wanted to offer support to victims of crime. Rape, murder and child abuse featured regularly among the reasons for protests.

Nelson Mandela Metro recorded the most notifications of protests about crime and the justice system of the seven municipalities: 26%, or 98 of a total of 384 protests.

Political parties and communities were responsible for 36 and 20 protest notifications respectively. The Democratic Alliance Women's Network was the most active political organisation. Nine of the 20 community protest notifications were to oppose bail to suspected murderers and rapists.

In Johannesburg Metro municipality, of 597 protest notifications submitted, 69 (12%) were about crime and the justice system. Non-governmental (NGO)/non-profit organisations made the most protest notifications – one in particular, Dorcas Women in Action, made 10 notifications about protests against rape and the abuse of women. Communities and political parties made 10 and nine protest notifications respectively. The ANC was the most active party.

Lukhanji had records of 140 protest notifications, 50 (36%) of which were about crime and the justice system – 18 of them were by political parties and 17 were by communities. Twelve of the 18 were by the ANC Women's League.

Twelve (19%) of the 62 protests in Makana were about crime. Seven of them were called by communities; five to oppose bail. In Blue Crane municipality, 12 of the 34 protests (35%) were about crime. Church and community groups accounted for most of them

Crime accounted for 10 of the 65 notifications to protest logged in Ehlanzeni, 50% of which were pickets by communities.

eThekwini Metro is the only municipality where crime is not one of the top two reasons for wanting to protest.

What about service delivery protests?

In 2012, service delivery protests appear to have increased sharply in number in all the municipalities except eThekwini Metro.

A number of factors have to be taken into consideration when looking at the data. Firstly, the records of notifications of protests do not cover the same time period for all seven of the municipalities. The records for Johannesburg Metro and Ehlanzeni start only in 2010, for example; only two metros have data from 2008 – Lukhanji and Nelson Mandela Metro.

Secondly, the data for 2013 does not cover the entire year for any of the municipalities, so it is not possible to determine whether or not people's desire to hold service delivery protests lessened in 2013.

Thirdly, this data comprises notifications made to municipalities of people's intention to protest, it does not include information about any protests that may have happened without somebody notifying a local authority.

Read Jane Duncan's article " The politics of counting protests" for more on this subject.

A pattern does appear to exist in the municipalities analysed, but clearly this data needs to be combined with other datasets before any conclusions can be drawn about wider trends in service delivery protests.

Who is protesting about service delivery?

In Nelson Mandela Metro, 42% of the service delivery protest notifications were submitted by communities and 42% by political parties. The South African Communist Party accounted for most of the political parties' submissions. It is worth noting also that 15 community protests made mention of dissatisfaction with ward councillors.

In Johannesburg, communities made 29 (58%) of the 50 notifications of service delivery protests, another 10 notifications were made by community organisations, mainly by ANC alliance partner, the South African National Civic Organisation.

Eight notifications by communities in Johannesburg mentioned housing, four mentioned dissatisfaction with ward councillors, and five mentioned electricity.

In eThekwini Metro, there were no service delivery protests logged after 2010, but this may be because the people who made the notifications to protest did not give service delivery specifically as the reason. Or it may be because people who take part in service delivery protests in eThekwini do not notify the municipality of their intention to protest.

Four out of the five service delivery protests recorded by the Ehlanzeni district municipality were by communities.

In Lukhanji, the Anti-Privatisation Forum organised four service delivery protests in 2008; since then, however, most of the notifications of protests have been made on behalf of communities.

In Makana, the Unemployed People's Movement and the Rural People's Movement appear to have been active at organising service delivery protests.

No notifications for service delivery protests were recorded at the Blue Crane municipality between 2008 and 2014.

What does all this mean?

The media regularly reports on protests and South Africa has a reputation for having a large number of violent protests. But, as Duncan has pointed out, we should be asking whether the perception that violence has become inherent to these protests is true, and if commentators are too quick to assume that protests are about service delivery.

Lizette Lancaster of the Institute of Security Studies  wrote in February that "according to the South African Police Service Incident Registration Information System (IRIS), police officers were deployed to monitor a total of 12 399 crowd-related events [34 incidents a day, on average] between April 2012 and 2013".

Most of these were public gatherings that had been given permission in terms of the Regulation of Gatherings Act, she added.

The information collected by Duncan and analysed by the Mail & Guardian is about those public gatherings. Not all of them are protests and not all of them turn violent.

Protests are defined by Duncan as: "Gatherings that are directed towards state institutions or other power holders, and that seek to influence or contest decisions made by them".

Non protests are non-political events that include, for example, open-air concerts, church meetings and school fun runs. Often they are social or religious events.

The municipal data analysed includes information about protests that were cancelled or were denied permission but it does not tell us which protests turned violent or which went ahead without notifying the authorities in accordance with the Act. But how municipalities implement the Regulation of Gatherings Act can influence whether a protest turns violent. For example, Duncan wrote about her research into the way the Act has been implemented in Rustenburg municipality.

To gain more insight into which protests turned violent, this municipal data will have to be combined with other datasets, of which there are many, some maintained by the police, others by researchers.

The Institute for Security Studies, for example, has made data it has collected about public violence available on an interactive map.

You can explore the municipal data in the interactive graphics below.

The data is also available to download as Google spreadsheets under a Creative Commons Attribution Share-alike licence here. If you think you can do something interesting with it, please email us and let us know.

More graphics will be added as data for additional municipalities is obtained.

If you are interested in the definitions used in this analysis, you will find more information below the graphics.

Learn About Tableau

Learn About Tableau

Learn About Tableau

Learn About Tableau


Each notification included in this dataset was classified as a protest or non protest by Duncan's team.

The "reason for gathering" and "type of organisation" categories were added to the datasets by Laura Grant to simplify the data analysis and visualisation. Some commonly used categories are:

  • Unions;
  • Workers groups, which are groups of workers that are not part of a union;
  • Political parties, which comprise the parties and all their leagues and ward branches;
  • Communities, which are those community groups that did not state an affiliation to a political party or any other organisation; and
  • Community organisations, which include groups such as the South African National Civic Organisation, the Anti-Privatisation Forum or the Soweto Electricity Crisis Committee, as well as community policing forums or groups that claim to represent more than one community. 

The stated reasons for gathering were also grouped into broad categories, for example:

  • Service delivery, if those words are stated in the reason on the notification to protest;
  • labour, which covers just about all work-related issues stated by unions and workers groups;
  • crime and the justice system, which, as stated earlier, covers not just protests about crime in general, but also picketing outside magistrate's courts to oppose the granting of bail to a suspected criminal, or offering support to an affected family; and
  • unspecified, which covers reasons such as "to hand over a memorandum to ... ", or "to picket outside a magistrate's court" that contain no specific information as to why people would want to do that, as well as instances when no details at all were offered.

More on gatherings that are not protests:
In Lukhanji, for example, churches made the most notifications in the non-protest group to gather, and most of these gatherings were classified for the purposes of visualising the data as social events. These include school cadet parades, a procession to mark the start of a rugby tournament, and Youth Day and Women's Day celebrations.

In the much larger Nelson Mandela Bay Metro, which has a population of around 1.5-million people, celebrations to commemorate Youth Day, Women's Day, World Aids Day, Nelson Mandela Day and many others were separated from the social category into their own commemoration category because they were so frequent. Commemorations were by the far the main reason for notifications to gather.

In Makana, the biggest group of non protest notifications were classified as private, mostly because they were made by individuals who intended to erect a tent near their homes without specifying the reason. Rhodes University social events also feature prominently.

In Blue Crane municipality, schools made the most notifications for non protest events. In Ehlanzeni, fun walks are one of the most popular non protest events organised by government departments, churches and NGOs.

In Johannesburg, most non protest notifications were called by religious groups. The second-most frequent are commemorations.

eThekwini is the only municipality of the groups where non protests outnumbered protests, and the bulk of the non protest notifications were for fun runs or walks.