I walked down to the river to see the body at 5.30 pm on Friday June 20 2014. A young American couple, paying guests on Glenconner farm in the Jonkershoek Valley, found it on their first and only walk around the farm and told me where to look.
I have tried to remember what I noticed about the body as I saw it in the fading evening light. I can only remember three things with certainty. He was wearing new Nike trainers, I could see from his naked torso that he was a young African man, and the opened palm of his right hand had turned white. Those three things stuck in my mind — unreconstructed by photographic memory.
He was lying on some rocks and was only half submerged. His right arm was stretched out and pointing down river towards Stellenbosch. I could not see his face. I did not go into the river to look.
I called some Glenconner workers to see if they knew the dead man. First, 60-year-old Stonie Thundu waded in and said he did not know him. He only told me later that he had seen a number of drownings on Lake Malawi where he had grown up and that if this man had drowned his tummy would have been distended. It wasn’t. Then 56-year-old Laurence Doko, who lived in the Fish Pond settlement near the Jonkershoek Reserve, rushed down in a clapped-out, low-slung white Toyota to take a look. He did not know the dead man in the river either. Nonetheless, he stayed long into the night watching police officials come and go, and acting as the dead man’s guardian angel.
I had rung the police and within half an hour a Detective Appolis, a 30-something coloured detective from the Stellenbosch CID, had arrived in a white bull-barred, unmarked bakkie, followed by colleagues and crime-scene professionals. By then it was dark and the road along the river was logjammed with police vehicles.
The detectives assumed he drowned. When I went to the Stellenbosch pathology lab a few days later to look at the docket, drowning was put down as the cause of death. It was plausible. Only six months earlier a Khayelitsha woman, on a religious outing to a picnic spot higher up the Jonkershoek Valley, had been washed away and drowned in a flash flood after her car had broken down on a river-crossing ford. Her body had come down the winding river, through Glenconner farm, and fetched up opposite the entrance to the Starke-Condé wine estate where penguin-suited emergency and rescue divers had fished her body out of the river.
Yet the drowning bothered me. The Eerste River in flood is terrifying. But the river was not in flood. It was placid on that Friday evening and had not been in flood for a week. Could this stranger have fallen off a bridge in the dark and cracked his skull on a rock? I walked the river on the weekend between the small stone bridge at Glenconner and the broader concrete bridge higher up at Assegaaibosch. I could not see it happening.
I worried what would happen to a stranger dead in the Jonkershoek. What if nobody claimed the body? Would he be given to science as a cadaver for students to dissect? Or would he be used in a fraudulent funeral policy scam, his fingerprints sold by a corrupt home affairs official with access to death certificates to a smart syndicate working the funeral insurance business?
Some of my questions were answered late on Sunday night when I was surprised by a visit from a Lieutenant-Colonel Clark, a 40-something white detective from the Kuils River CID.
“We are looking for a body,” he said as I invited him into the Glenconner kitchen. “We have the perpetrators here and this is where they say they left the body but we can’t find it.”
“We found the body on Friday,” I said. “And the Stellenbosch police took it away.”
He was surprised and then disappointed. I showed him the photo I took of the body and the video of the Eerste River in flood. He was mildly interested. He rang his colleagues, who had formed a phalanx of blue-light flashing cars down along the river, to call off the search.
Then it dawned on me how strange this was. Why did they have the perpetrators without a body? As a crime-fiction aficionado I know that you find the body first and then go looking for the murderer.
“Hang on,” I said. “What’s going on here? How did you catch the murderers?”
“These people talk,” he said vaguely.
“Which people? Where do the murderers come from?” I asked.
“Mfuleni,” he replied. I had no idea where that was, but later discovered it was a township on the other side of the airport road from Khayelitsha.
“You mean this had nothing to do with us here in the Jonkershoek?” I asked.
“Was the murder committed here on the farm?” I asked.
“Yes”, he said. “The perpetrators drove here from Mfuleni and then drowned him in the river.”
“When?” I asked, as the full horror of events began to take shape in my mind.
The murder was committed on Thursday evening, June 19. That evening I had been in the house that is the length of two rugby fields away from the river across a water meadow on which four horses graze. There is a narrow stone bridge across the river at the entrance to Glenconner. That afternoon, from 4pm, cars streamed across the stone bridge, driving along the gravel road past the gate-less entry to Glenconner on the right and onto the Meyers on the left, building-construction industrialists, who were hosting a memorial funeral service for a local resident in their Tuscan fantasy villa they hoped to turn into a wedding-venue-cum-hotel. I hadn’t seen the cars or heard the service. I hadn’t seen a car drive into Glenconner and a murder committed at our picnic spot and a body dumped in our favourite river swimming pool!
Why did the murderers choose Glenconner? The lieutenant-colonel had no idea. He gave me his number, but he never answered my calls over the next three months as I tried to discover what had happened.
The Eerste River winds down through the Jonkershoek Valley to Stellenbosch. For thousands of years Khoi pastoralists relied on the river during their visits to the valley in an annual pattern of transmigration. They came in summer and left in autumn, burning the fynbos before they left so that new grass would grow during the winter rains ready for their next visit. The Khoi pastoralists were displaced by settler farmers soon after Stellenbosch was founded in 1679.
All of the first settler farmers in the valley were freed slaves. Jan Lui, who came as a slave from what is now called Sri Lanka, owned one of the valley’s three great estates, Old Nectar, the one in the middle between Lanzerac at the mouth of the valley and Jonkershoek at the throat where the valley turned into a wilderness of wild honey and fynbos.
Now, three hundred years later, even though I could not see Jan Lui’s fields through the trees lining the river on Glenconner, I could hear machines at work in the field he first planted over the Jonkershoek Road in what is now the Starke-Condé wine estate.
Glenconner farm is a smallholding, one of a patchwork quilt of smallholdings that came into being in the late twentieth century from the breakup of Old Nectar after centuries-old cycles of booms and bankruptcies. The Eerste River ran through all of them from throat to mouth. For centuries the river gave water not only to the farms but also to the town of Stellenbosch as well.
I was thinking about this river giving life to the valley. And now death. I remembered that Andrew Brown’s novel — Cold Sleep Lullaby (2005) — begins with the body of a woman found in the Eerste River near a weir. I wondered where the weir was. So I asked Andrew Brown. It did not exist, he told me. He made it up.
Could this be, I wondered, the first murder committed in the Eerste River?
The bundu court
Two Stellenbosch murders captured the public imagination in the first two decades of democracy and freedom. The first was a racial murder. Pieter Orffer was thirty-five and self employed in 1994 when he and his family were murdered. It was an apparently inexplicable murder by black employees.
When asked why he killed a family he cared for, one of the murderers answered: “The God’s truth is, I don’t know what happened to me. The devil was in me. Pieter gave me everything. I still see his face in my head. I have to live with that. He doesn’t. It’s a scandal, a real scandal.” It’s what many white people feared in the long drawn out birth of democracy: a murderous outbreak of black hatred.
The second was a woman murdered in 2005. Inge Lotz was a 22-year-old Stellenbosch university student and the daughter of a Stellenbosch university professor. She was murdered one evening in her flat in a gated estate. Her 22-year-old boyfriend was tried for and acquitted of her murder. The case failed on the exposure of fraudulent forensic evidence exacerbated by police negligence, both issues that bedevil all homicide cases.
Yet the truth is that neither racial homicides nor female homicides are common in Stellenbosch. Rather the murder of the man from Mfuleni fits the profile of Stellenbosch murders all too well. In 2013, in a census 2011 population estimate of 175 000 people (five in ten of whom are coloured, three in ten African, and two in ten white) only 24 homicides were recorded.
The Stellenbosch police station commander, Brigadier Dirk Mentoor, is reported to have said: “Of the 24 murders, 15 have not been resolved. Most of them were committed in Kayamandi and Enkanini and we suspect most of these had to do with bundu courts.”
Kayamandi is a dormitory town situated close to the centre of Stellenbosch, originally built to house African farmworkers. It’s Stellenbosch’s mini-Khayelitsha. Like Khayelitsha it has a traditional system of justice, the kind that Mentoor had pejoratively called a “bundu court” system.
What’s a bundu court? The Khayelitsha commission of inquiry into policing in the township, which submitted its report a year ago, described the history of alternative legal forms like this. Without a history of state policing, African migrants policed themselves through community-led forums. Order was enforced by taxi associations, which had the muscle to find and punish offenders from thieves to murderers, or street committees that are traditional versions of suburban neighbourhood watches. Order was imposed without “deliberation or evidence”. Punishment was carried out summarily. There appeared to be no chance of an appeal. Mob justice was the result.
One expert witness to the commission said: “… the police are only a small part of the problem that creates ‘mob justice’. The ‘desperation’, ‘anger’, and sense of being ‘cornered’ that residents attribute to life in Khayelitsha says as much about the condition of township life more generally — poverty, inequality, the unchanging urban condition of poor black Capetonians —as it does about the failure of the criminal justice system to protect residents.”
I went to talk to Mentoor, curious to know whether the Glenconner murder was the outcome of a “bundu court”. I found him in his office late on a Friday afternoon. The Stellenbosch CID is opposite the Distell Group in a brick-faced building that looks more like a private school than a government office.
“Do you know about the Glenconner farm murder,” I asked him.
I was surprised when he said yes.
“Do you know why the murderers chose Glenconner farm?” I asked.
“That I can’t say”, he replied.
He was a kindly 50-something-year-old man who must have started his career in the police under apartheid, when Stellenbosch was a coloured town and Kayamandi was in its infancy.
“Am I in danger living on the farm then?” I asked.
“No I don’t think so,” he replied. “You see, it was a contract killing and had nothing to do with you.”
It was at the time that the Shrien Dewani case was being heard in the Cape Town high court and the issue of a contract killing was at its heart. It was the state’s case that Dewani had paid contract killers R15 000 to murder his just-married wife. He was accused of arranging the contract on his arrival at the airport with the taxi driver who drove him to the Cape Grace hotel at the Waterfront where he was to stay and where much of the circumstantial evidence used against him was captured on CCTV footage.
There was more surprise on social media over the ease with which a contract killing could be arranged than over the badly prepared case the state had put together. It was a case that had involved a lengthy extradition process at considerable state expense, only to find in court that Dewani had a reasonable alternative explanation to a contract killing: it was a failed ransom attempt gone wrong.
“Why was the victim murdered then?” I asked Mentoor.
“He sold a house he did not own in Mfuleni. When the buyer came to occupy, he found there were people living in the house who claimed to own it,” he said.
“Why do you have to kill a man for that?” I said.
“That I can’t say,” he replied.
I had heard about RDP housing scams. I remember that when Thuli Madonsela, the public protector, investigated maladministration in the sector in 2012 she found the most common complaint was “illegal occupation, the lengthy process of removing illegal occupants once they had been identified, and public officials either selling houses or allocating them to their friends or families”.
The fraud lay in bribing local officials to allocate houses or simply allowing friends to move in. The ease with which this scam worked had to do with fictional waiting lists and limited deeds registration for ownership of township property. In 2010 the South African Institute of Race Relations revealed that only half of the three million RDP homes were registered at deeds offices. This meant that their owners were unable to use them as collateral for loans and other financial services.
“How do you know all this?” I asked Mentoor.
“They confessed,” he replied.
“Is that all you need? Did your detectives go to Mfuleni to find out what the victim had done and how he had been caught,” I asked.
“No need. Open and shut case,” he replied.
I had read Jonny Steinberg’s written evidence to the Khayelitsha commission about police inefficiency. He had noticed how difficult senior police officers found working in Khayelitsha, how many looked down on migrants who had come from the Eastern Cape, and where it was difficult to find their way around, let alone to find anyone to explain who had committed crimes.
“Yet we know from the meticulous studies of policing in many other urban settings in contemporary South Africa,” he wrote, “that there is in every case a complicated filigree of personal connections between various informal settlement residents and individual police officers and that these personal relationships play no small part in shaping how informal settlements are policed.”
That’s what I thought, having read James McClure’s celebrated novels — Steam Pig (1971), The Sunday Hangman (1977), and The Song Dog (1991) — about the Pietermaritzburg townships in the 1960s. Subtly subversive of the racial order, Afrikaans-speaking Lieutenant Tromp Kramer was able to solve township crimes because Zulu detective sergeant Mickey Zondi had inside access to township hustlers and a cultural understanding of how the traditional justice system worked.
“Can I read the confession?” I asked the brigadier.
“No, the case has been sent to the NPA [National Prosecuting Authority] in Cape Town for a decision on prosecution, but you can visit the accused in jail and ask them yourself if you like. One is being held at Pollsmoor, but the other one is an awaiting trial prisoner in Stellenbosch prison,” he replied.
I went to talk to one of the accused in the Stellenbosch prison. Or rather my Xhosa-speaking interpreter, Inga Dyantyi, a Langa-born Stellenbosch law student, talked to him. His name was Siphe Hlalukana and his age on the charge sheet was 21, but he must have been older because he was married with four children. He worked at the airport as a security guard and lived in Mfuleni. He said he was a friend to the 41-year-old victim, Luyanda Michael Matsolo. He said he was driving him into the Jonkershoek Valley that evening so that he could look for work. He denied killing his friend.
Over an hour or more Inga tried to find out why he had been arrested, why he had confessed, and why he had “pointed out the body” in the Eerste River on Glenconner farm. She made no impact. He made no admissions. He denied murder. The more he talked, the more she believed him. Looking at him, sitting in that bare and barren prison cubicle for visitors, I found it difficult to see a killer inside him.
What must have happened? Here, again, the Khayelitsha commission is revealing. Traditional justice is a complex set of practices and police play a variety of roles. They turn up after a crowd has punished an accused. They protect some people who inflict punishment on suspects. They turn a blind eye to traditional executions on the local sports field. They also arrest victims if he and his family are known and there is community pressure for this solution.
This is what must have happened to Hlalukana and his accomplice. They were fingered not by the police but by a street committee (one of Mentoor’s “bundu courts”) in Mfuleni. Some people in Mfuleni knew what he had been contracted to do and had turned him over to the police.
Hlalukana, a man without a matric, migrated with his family four years ago and found a place to live in Mfuleni and work as a security guard at the airport. Others called him lucky to find employment in an area where housing policy reinforced apartheid segregation, while they were left to hustle over the theft and sale of state resources such as copper cable and RDP houses.
But he didn’t make enough on his night job as a security guard to support his family and when, like the state witnesses in the Dewani trial, he was offered a slice of a murder contract he found it difficult to resist. He may have heard on the grapevine that these murders were easy to commit and unlikely to be punished.
I was unable to discover anything about the victim, Luyanda Michael Matsolo. Is it likely that he acted alone and that this was the only sale of an RDP house that he had made? Or was he part of a broader housing fraud involving better-connected township worthies? I surmise that friends or family missed him and sent the local justice system in search of his killer. They found him and handed him over to the police. Yet Hlalukana, awaiting trial in Stellenbosch jail is, according to Mentoor, only a hired killer settling a claim for R13 000 with taking a life.
All the state has is a confession that is unlikely to stand up to a proper interrogation in court. Yet the NPA thinks it has a winnable case. The last I heard was that Hlalukana was in Pollsmoor and is due in the high court this month, a year and two months after the murder. He applied for bail but it was refused. If he had been given bail, he would surely have disappeared and the case would have become yet another statistic in the high number of unsolved murders in Stellenbosch.
Until I can read the confession and discover how Hlalukana knew where to look for the body if he was not the murderer, consider a few concluding thoughts on how detectives solve murders.
Are unsolved murders part of police policy or a result of police incapacity or incompetence? RW Johnson certainly thinks it’s a policy by omission and sees the Khayelitsha commission as proof.
“Inevitably when the commission finally met,” he writes in How Long Will South Africa Survive (2015), “it heard repeated testimony to the fact that the police had largely given up bothering about this vast settlement where, as a result, vigilante justice rules and, on average, one person a week is burnt to death in a necklacing. No one had ordered the police to neglect Khayelitsha — they have just decided on their own to ignore the settlement of over half a million people”.
Zackie Achmat in his Groundup article, “Apartheid geography and murder in Cape Town” (2014), also thinks it’s a policy by omission but he takes a different political tack. He quotes a conclusion of the commission that the absence of policing was evidence of a systemic bias against poor black communities and “evidence of a failure of governance and oversight in every sphere of government”. He argues a case against spatial inequality and against leaving poor black people in urban ghettoes. There are many murders in ghettoes. There aren’t many murders in socially integrated communities. That’s his point: compare Rondebosch in Cape Town with Khayelitsha.
Look at what happened in Bogotá in Colombia, he says, when a progressive municipality shuffled up its spatial living arrangements. It’s worth a long quote:
“There is a relationship between gangsterism, drugs, alcoholism, our violent cultures, poverty and murder in our city. But one of the most important factors, often ignored, is spatial injustice: the fact that many people essentially live in ghettoes.
“We should be asking the question: Where are people murdered in Cape Town, and why there? Where people live — CBD, suburb, township or informal settlement — directly affects their right to life and chances of being murdered. Where you are murdered also increases the chances of an investigation.
“Class and race inequality (meaning disparities in wealth, income and education) are vital to understanding murder. Related to this is geographic apartheid, meaning historical (and today’s) ‘dumping’ of the poorest and most vulnerable people furthest away from CBDs and traditionally white suburbs.”
I don’t know what police policy in Stellenbosch is on solving murders. I do know that police anywhere have always found it easier to solve domestic murders than predatory murders, because domestic murderers often confess and admit to their crimes through guilt or remorse. I do know that policy can change the way murders are solved. Police have long chosen which homicides to investigate as murder and which to shuffle into the magistrate’s court as culpable homicide — and not only on legal grounds. It’s expensive to investigate a murder.
If you look at policing practice in the United States there were periods when more murders were solved than now. Currently it’s a two in three chance of a solution. Fifty years ago it was close to a full house. The argument is that police are now more focused on prevention than solution. It depends on policy. Choose your policy and then reorganise your resources to satisfy your aims.
Can it be acceptable that Stellenbosch leaves an urban ghetto to traditional justice? Is it acceptable to have no Kayamandi murder solved, even in the light of the astonishing number of high profile suburban murder cases that have gone unsolved in the recent past? Or is it acceptable to explain away this failure, as Appolis said to me, by asserting: “Stellenbosch is the dumping ground for Khayelitsha murders.” If the Kuils River CID has a presence in Mfuleni, surely Stellenbosch needs to work harder in Kyamandi. McClure did not romanticise a golden age of township policing. It’s the policy that’s to blame.
Rob Turrell is the author of White Mercy, A Study of the Death Penalty in South Africa (2005).
All other visuals and video by Rob Turrell.
Layout by Ines Schumacher.
Published by the Mail & Guardian © 2015