Why counting service delivery protests is more complicated than it seems

Service delivery protests are reaching an all-time high in South Africa, and the vast majority are violent. That was the claim of a widely reported study released by Municipal IQ, an organisation which monitors local government.

“The second quarter of 2018 [shows] a new record for protests measured by quarter – 101 protests between April and June, against a previous record for the second quarter of 2017 (73),” read the Municipal IQ press statement.

The organisation elaborated to TimesLive that 94% of the service delivery protests recorded so far in 2018 had been violent, which represented a “significant uptick” when compared with previous years.

Municipal IQ states that it gathers its data from media reports of protests, as well as other public domain sources such as police press releases. It classifies service delivery protests as those which “raise issues that are the responsibility or perceived responsibility of local government (such as councillor accountability, the quality and pace of basic service delivery, and in metro areas, housing)”.

It does not include protests on issues “falling outside of local government’s service delivery mandate”, such as the demarcation of municipal boundaries, industrial relation disputes, or “clear party political issues”.

Daily Maverick asked Municipal IQ for more detailed information on its methodology and the location of recorded protests, but we were told by economist Karen Heese that “we do have disaggregated data but this is for our subscribers”.

When it comes to the causes of protests, Heese said: “The most common grievances are housing and electricity but it varies from province to province.”

It is difficult to evaluate the legitimacy of the Municipal IQ findings on the basis of the minimal information provided to media, but it appears that the data provided this week should be treated with a certain amount of caution.

This is due in large part to the lack of agreement over what constitutes a service delivery protest, and what constitutes a violent protest.

Dr Carin Runciman, from the University of Johannesburg’s Centre for Social Change, told Daily Maverick that the picture presented by Municipal IQ appears to be “misleading”, in that it probably under-represents the number of protests happening but over-represents the percentage of protests that are violent.

“Our data at the Centre for Social Change counts at least double, if not more than double, the number of protests,” says Runciman, while acknowledging that the centre’s definition of “community protests” is wider than that used by Municipal IQ.

The fact that the organisation uses media reports on protests as the basis for its data is problematic, suggests Runciman, because it doesn’t take into account the fact that media reporting on protests is quite selective.

The Centre for Social Change noted in a recent paper: “We estimate that the media has reported less than a quarter of the number of protests recorded by the police.”

The centre says that smaller protests tend to go unreported, as do protests without an element of violence or spectacle. Urban protests receive more media attention, as do protests in the two provinces – Gauteng and Western Cape – where most media houses are based.

SAPS records do suggest that the media covers only a fraction of South African protests.

The SAPS annual report, released in August 2017, recorded 14,693 “crowd-related incidents” over the 2016/17 financial year, although not all of these were protests. The police’s habit of classing together everything requiring the presence of public order policing (POP), from a concert to a violent demonstration, has long been a source of frustration among researchers.

Of that figure, the SAPS recorded 10,978 “peaceful incidents, such as assemblies, gatherings and meetings”, and 3,715 “unrest-related incidents, such as #FeesMustFall, labour disputes, as well as dissatisfaction with service delivery by local municipalities and in the transport and education sectors”.

But the police definition of “unrest” is very broad, says Runciman, and would include protest techniques like blocking roads.

Municipal IQ does the same thing in its data: it specifies that it classes as “violent” all protests “impinging on the movement or property of others, including the state”.

Says Runciman: “The implication is that blocking a road is violent. I don’t think that’s an accurate way to characterise violence. It is protest designed to be disruptive of the normal social order. That shouldn’t be confused with violence.”

The Centre for Social Change uses what it says is the internationally accepted definition of violence in such contexts: actions involving injury to people or property.

For the rest, it distinguishes between “orderly” and “disorderly” protests.

Its latest research, which looked at protest figures up till 2017, concludes:

“While there has been an upwards trend in the number of disorderly community protests since 2006, the proportion of all community protests that was actually violent has been relatively stable since 2010.”

This may seem like a semantic dispute of interest only to policy nerds, but Runciman says that the effects can be significant.

“When you can take events that are disruptive and call them violent, you’re mischaracterising the nature of the protest. This is used to delegitimise legitimate concerns.”

She points out that their research suggests that community protests generally happen only after a long period of unsuccessful engagement with authorities.

“The underlying causes of protest remain the same – structural inequality – but protests are also about the quality of post-apartheid democracy.”

It is the absence of effective communication channels with government, she says, that gives rise to protests – but the focus on “violent” elements to protest “plays into a narrative that says that people just want things, rather than wanting to actively participate in development”.

The notion that service delivery protests are overwhelmingly violent also has allowed the police to expand the scale and intensity of its public order policing, as groups like Right2Know have noted with alarm in recent years.

The 2017 annual SAPS report states that in 2015, there were 4,617 POP members. By 2019, this is projected to rise to over 9,000.

The report also identifies as a “critical need” the building of more water cannons, Nyalas and “wire trailers” for use in public order policing.

Service delivery protests are an obvious reality of South African life, but the notion that most of them are violent is questionable. We should be careful of what this narrative could mean, says Runciman.

“We do need to be concerned about that [perception] and its effects on policy-makers,” she says. DM