On 10 September 2021, South Africa’s Parliament passed three amendment bills, known as the Gender-Based Violence Bills, in response to public outrage over soaring levels of gender-based violence and femicide across the country. But in a nation where strong legislation is poorly implemented and rarely changes front-line practice, can we expect the new bills to make a real impact? 

While each bill promises to strengthen protections for survivors, the Domestic Violence Amendment Bill (Amendment Bill) is particularly vital. Domestic violence is one of the most prevalent forms of gender-based violence, and intimate femicide is the leading cause of murder for women in South Africa.

The Amendment Bill removes the ‘imminence’ requirement under Section 8(4) the Domestic Violence Act 116 (1998). In practice, this means that a survivor can report a violation of their protection order without being expected to prove that they will be killed if the police do not make an arrest immediately. 

‘Imminence’ is a murky legal concept. It requires the police to ascertain whether the threat of harm is ‘so likely to happen’ that the only way to prevent it is by making an arrest, even when there is a protection order — with an attached warrant of arrest — already in place.

From the South African Police Service’s (SAPS) perspective, this amendment has the potential — at least in theory — to offer one of the most critical changes to front-line practice. One police member who requested anonymity told the Institute for Security Studies (ISS): “So, you come to the station, and you say, ‘I was assaulted, and I still have a fear that he will assault me even further.’ We will take the charge here, but we won’t go with the imminent harm because most of the time the imminent harm cannot be proved.”

It would therefore make sense to think that removing the ‘imminence’ requirement from Section 8(4) would make it easier for the SAPS to arrest for protection order violations. However, this assumes that other provisions of the Domestic Violence Act are being implemented correctly — and as previous research studies have shown, this is simply not the case.

Historically, the SAPS has been the primary target of public scrutiny over poor compliance with the act. Magistrates’ courts have skirted much of the criticism, despite having the power to grant, deny, or set aside certain conditions of protection orders, and to issue arrest warrants. 

Research conducted from 2020 to 2021 by Mosaic, a Cape Town-based non-governmental organisation, in magistrates’ courts across the city discovered systemic levels of non-compliance by magistrates. In particular, some magistrates refused to attach ‘suspended’ warrants of arrest with interim protection orders, in direct violation of Section 5(7) of the act.

Interim protection orders are temporary, and intended to provide immediate protection to survivors while they wait for their protection orders to be finalised, which can take up to six months. Under Section 5 of the act, magistrates must attach suspended warrants of arrest to the interim protection order so it’s easier for police to make arrests if the order is violated before it is finalised. 

When Mosaic researchers asked about this practice, one magistrate explained: “We do not issue the warrant on the same day that the complainant comes to court. It must be served first to the respondent and then when it is served, they come back for the warrant of arrest. 

“Most times they will approach the police station, and after that, what we’ve implemented is that the police will serve it because we’ve designated the police officers who are serving these domestic violence orders who deal with domestic violence matters. Then they will inform them and they will come back.”

The Mosaic research further found that some magistrates felt this would protect respondents from being arrested before they had been served the interim protection order. This practice would seem fair except the warrant of arrest is ‘suspended’ — meaning it does not take effect — until it is served on the respondent, as is the case with interim protection orders.

Under the Amendment Bill, the interim protection order and suspended warrant of arrest will take effect immediately once it has “come to the attention of the respondent,” rather than at the time it is served.

Magistrates also told Mosaic researchers they believed evidence of an interim protection order was sufficient grounds to make a warrantless arrest if the protection order was violated and the victim was in danger. This rationale ignores the operational realities of policing and the risks in making warrantless arrests, never mind the undue hardship it places on survivors to meet a burden of proof they have already satisfied. 

As one SAPS member explained to the ISS: “We are much more successful with the final protection orders. Much more. So we go on the first charge of contempt of court. But if there is nothing, then the act says that imminent harm must still exist for us to make an arrest. So if the [protection order] has a warrant attached, from what I gather, it is much easier than when someone comes in with an interim protection order.” 

So, if magistrates and police have been following their intuition rather than the prescripts of the law for the past two and a half decades, why would the Amendment Bill be any different? 

The Domestic Violence Act certainly has its weaknesses, but law reform will not change patterns of non-compliance that have slid under the radar for years. In the rollout of the Amendment Bill, attention now needs to be turned to overhauling the current act training curriculum for both the police and magistrates, and strengthening accountability systems for non-compliance. DM

This article was written prior to the bills being signed into law by President Cyril Ramaphosa on 28 January 2022.

Dr Kelly E Stone, Senior Consultant, ISS Pretoria.

First published by ISS Today.

[hearken id=”daily-maverick/9041″]