On 4 September 2002, at the corner of Selvin and Langa Streets in the Nomzamo informal settlement in the Western Cape, Anton Wyngaardt was shot three times and died at the scene from loss of blood, with one bullet piercing his heart. At the same time, John Steven Haggard, who was traveling with the deceased, was shot twice on his right arm and face but managed to drive away. Haggard was a paraffin salesman who had come to Nomzamo to recruit new customers, with the deceased having worked with him for approximately two weeks. Haggard had stopped his vehicle at a Spaza shop on two occasions and noticed the appellant inside the shop on both occasions. He had also seen the appellant on two previous occasions when the latter inquired about paraffin prices. During the shooting incident, which occurred in broad daylight around 13h45, Haggard was first shot in the arm, then saw the shooter move to the passenger side and shoot the deceased who had exited the car. When Haggard tried to help the deceased, the shooter fired again. The shooter then confronted Haggard at a distance of two to three metres, and Haggard knelt and pleaded not to be shot. The shooter took careful aim and shot Haggard in the mouth. Haggard identified the shooter as the appellant. The appellant was arrested five months later on 28 February 2003. He pleaded not guilty and raised an alibi defense, claiming he had accompanied his elder brother to wholesalers at Khayelitsha at the time of the shooting.
The appeal against both convictions and sentences was dismissed. The convictions for murder and attempted murder were upheld, as were the sentences of 15 and 7 years' imprisonment respectively, to run concurrently.
Identification evidence from a single witness in criminal proceedings, while requiring caution, can be sufficient to ground a conviction for serious offences including murder and attempted murder where: (1) the witness had multiple opportunities to observe the accused, both before and during the commission of the offence; (2) at least one of those observations was made at close range (two to three metres) in good lighting conditions (broad daylight) with a direct view of the accused's face; (3) the witness's evidence is credible, honest, and internally consistent; (4) there is independent corroborative evidence supporting the witness's presence and observations, even if that evidence does not directly identify the perpetrator; and (5) the cumulative effect of the observation opportunities renders the identification reliable beyond reasonable doubt. Minor inconsistencies, such as failure to notice superficial facial marks that are not readily visible, do not necessarily undermine the reliability of otherwise sound identification evidence, particularly where the witness was under extreme stress during the critical observations.
The court noted that the judgment was marked as having 'no precedential interest', suggesting it was seen as an application of established principles rather than breaking new legal ground. The court observed that even if the photo identification evidence were excluded entirely, there was sufficient evidence to convict based on Haggard's in-court identification and the other evidence in the case. The court also commented on the physical positioning of spent cartridge shells found at the scene (two inside Haggard's car and three inside the shop), though this aspect was not developed further in the judgment. The court noted that Inspector Koekemoer found the appellant's demeanor suspicious because he was watching every move made by police at the scene and then simply vanished, though at that time Koekemoer did not know who had committed the crimes. While no argument was advanced on appeal regarding sentence, the court noted that there had been an appeal against sentence included in the notice of appeal.
This case provides important guidance on the approach South African courts should take to identification evidence from a single witness in criminal matters. It establishes that while single witness identification evidence must be approached with caution, it can be sufficient to ground a conviction where: (1) the witness had multiple opportunities to observe the accused; (2) some observations were made at close range and in good conditions; (3) the witness had seen the accused on prior occasions; (4) the witness's evidence is internally consistent and appears honest; and (5) there is independent corroborative evidence, even if that corroboration does not directly identify the accused as the perpetrator. The case demonstrates that courts must take a holistic view of identification evidence, considering the cumulative effect of multiple observation opportunities rather than focusing on individual fleeting moments. It also clarifies that minor inconsistencies (such as failure to notice superficial facial marks) in otherwise reliable identification evidence should not detract from the overall reliability of the identification, particularly where the circumstances were traumatic.