---

IN THE LABOUR COURT OF SOUTH AFRICA

(HELD IN CAPE TOWN)

Case NO: C640-07

In the matter between

FAWU obo Kapesi and 31 others APPLICANTS

and

PREMIER FOODS LIMITED t/a BLUE

RIBBON SALT RIVER RESPONDENT

JUDGEMENT

AC BASSON, J

Nature of the proceedings

1. Most of the workers employed at the Respondent’s Salt River Bakery which trades as “Blue Ribbon Salt River” embarked on a protected strike from 5 March 2007 to 9 May 2007. The individual Applicants, totaling 32 were dismissed pursuant to a procedure in terms of section 189 and 189A of the Labour Relations Act 66 of 1995 (hereinafter referred to as “the LRA”). This Court is called upon to decide on the fairness or otherwise of the dismissal of the individual Applicants.

2. The fact that a strike took place is not controversial as it is accepted by all parties to this dispute that workers have the right to embark on a protected strike and that they are entitled to engage in (lawful) activities such as picketing and peaceful protest. It was accepted by all parties that the right to strike is a right that is firmly entrenched in our law.

The dispute between the parties¹

3. The union (“FAWU”) and its members (the individual Applicants) embarked on a protected national strike on 5 March 2007 in support of the union’s demand for centralised bargaining. The demand for centralised bargaining was aimed at bringing the wages of rural employees up to the levels of employees in the urban areas.

4. Certain of the workers at the Salt River plant chose not to participate in the strike, as was their right. Several of these (non-striking) workers as well as members of management were thereafter subjected to violent acts of a severe criminal nature. The Court heard harrowing evidence from some of these victims who recounted that their homes and that of workers who chose to continue working were firebombed and ransacked. In one incident a firebomb was thrown through a window which landed on the bed of the son of one of the non-striking employees. The court also heard evidence of cars and possessions of employees being set alight. These employees were visited at night by groups of individuals who threatened them with physical harm and death. One female employee was dragged from her home at night and assaulted with pangas and sjamboks. Even after the strike had ended, these acts of intimidation and threats of violence did not cease. Even as late as 30 November 2007 the house and vehicle of Mr. Mdleleni were set alight and shots fired at the house. A neighbour, Mr. T Mdlalo subsequently identified some of the attackers. Shortly thereafter Mdlalo was shot and killed near his home.

5. The court also heard evidence of a conspiracy that was put in place to have the Respondent’s regional director assassinated. Money was in fact collected for that very purpose from some of the striking employees. In summary, this strike was marred with the most atrocious acts of violence on non-striking employees. The individuals who perpetrated these acts clearly had no respect for human life, the property of others or the Rule of Law. What makes matter worse is the fact that it appears from the evidence that the police and the criminal justice system have dismally failed these defenseless non-strikers. Although criminal charges were laid against certain individuals, nothing happened to these charges. The non-strikers were completely at the mercy of vigilante elements who did as they pleased and who had no regard for the life and property of defenseless individuals. It must be pointed out that although a certain measure of rowdiness and boisterousness behaviour are expected or typical to most strike actions, the acts that marred this particular strike were particularly violent and senseless and stretched far beyond the kind of conduct that normally occurs during a strike. The witnesses who gave evidence in Court were visibly traumatized by the acts of these vigilantes.

6. Strikes that are marred by this type of violent and unruly conduct are extremely detrimental to the legal foundations upon which labour relations in this country rest. The aim of a strike is to persuade the employer through the peaceful withholding of work to agree to their demands. As already indicated, although a certain degree of disruptiveness is expected, it is certainly not acceptable to force an employer through violent and criminal conduct to accede to their demands. This type of vigilante conduct not only seriously undermines the fundamental values of our constitution, but only serve to seriously and irreparable undermine future relations between strikers and their employer. Such conduct further completely negates the rights of non-striking workers to continue working, to dignity, safety and security and privacy and peace of mind. In this regard Mr. Oosthuizen for the Respondent also argued that this type of conduct by striking employees will destroy the workplace relationship after the strike is over. In this regard Mr. Lavery (“Lavery” - the Respondent’s regional director for the Western and Eastern Cape) testified that it would constitute a threat to harmonious inter-personal relations between staff if the Applicants were allowed to return to work. I am in agreement that it is difficult to envisage how workplace relationships can be re-established after a particular violent strike marked by intimidation (and especially to the degree in this particular case) comes to an end and how one can expect to resume a workplace relationship with someone who is suspected of having set your car alight or having petrol-bombed the room in which, but for a fortunate coincidence, your child would have been sleeping. I will return to this aspect when I consider an appropriate remedy in the event of a finding that the dismissal was unfair.

7. I must, however, pertinently point out in all fairness to Mr. Kahnowitz (who acted on behalf of the Applicants), that he was at great pains to repeatedly emphasise the point that the Applicants were not in court to defend criminal actions by strikers on non-strikers. Mr. Kahnowitz also did not dispute the fact and employer (and this Respondent in particular) is entitled to take action against individuals who perpetrated these acts of violence. What was, however, in dispute was that these particular Applicants were involved in these acts of violence in light of the common cause fact that their participation in these acts of violence has never been proven.

8. As a result of the violent conduct which marred the strike, the Respondent on 7 March 2007 obtained an urgent interim order against the Applicants to interdict violent conduct during the strike. The order was granted on an unopposed basis and confirmed on 23 April 2007, also on an unopposed basis.

9. The strike was settled on 9 May 2007 and the striking employees returned to work. The individual Applicants were served with notices of suspension on 9 May 2007 which read as follows:

“Please be advised that due to allegations of serious misconduct committed by yourself during the current strike on centralised bargaining, you are hereby suspended on full pay pending disciplinary procedures. You will be notified shortly of the date and time of the disciplinary investigation”.

10. The suspended employees (and in some cases shop stewards) Mncedisi Bonongo, Ntsikelelo Bonongo, Persius Dunjana, Bethwell Nonjola and Sabelo Sijila were served with notices of disciplinary enquiries to be held in the period 12 June 2007 to 14 June 2007. The charges against them related to alleged participation in, inter alia, incidents of criminal misconduct.

11. By letter dated 14 June 2007 the Respondent informed the union that the hearings involving shop stewards and scheduled for that week had been postponed. All affected shop stewards would, however, remain on suspension. The Respondent also indicated that the hearings will be rescheduled.

12. Mrs. Elliott (“Elliot” - an employee of the Respondent) was tasked to transcribe the numerous statements from individuals who came forward with information pertaining to the various acts of violence perpetrated against them (the non-strikers). These statements were submitted into evidence before the Court. Some of the deponents cannot be identified from the statements as their names have been blacked out. Elliot gave detailed evidence on how she recorded the information given to her. She testified that certain of the employees specifically requested that their identities be kept secret. Her evidence in this regard was not seriously challenged. All of these statements formed part of the documents placed before the Court. In some of the statements the deponent would identify the culprits by name. In other statements the victims of violent acts were only able to recount the events but were not able to identify or name the culprit/s. It is significant that the greater majority of statements in which culprits are identified by name it is not possible to identify the deponent.

13. On behalf of the Respondent it was argued that in respect of these statements and in view of the fear of harm and victimization, there exists a clear and compelling reason why hearsay evidence by the person fearful to testify should be admitted. It was also argued that it is in the interests of justice to admit such statements. To do otherwise, so it was submitted by Mr. Oosthuizen, would mean that the threats and intimidation which was the reason why the informants were reluctant to testify would produce precisely the undesirable result sought by the perpetrators. I will return to this submission (and the admissibility of hearsay evidence in the context of disciplinary hearings) hereinbelow where I consider whether or not it was possible, despite the violence and intimidation, to have proceeded with a disciplinary hearing. I have to point out however, that although I accept that, in certain circumstances, it may be warranted (and even desirable) to receive hearsay evidence, it should be borne in mind that a Court is not a substitute for a disciplinary hearing. If the employer decided not to hold disciplinary hearings it cannot come to Court and then request the Court to admit the very same evidence which it (the employer) decided not to use in disciplinary hearing in order to prove the guilt of an employee. I must, however, in all fairness to Mr. Ootshuizen point out that it was not submitted on behalf of the Respondent that the hearsay evidence (the various witness statements) should be admitted in order to prove the guilt of the Applicants, he submitted that it should be admitted merely to establish the existence of an operational requirement and to demonstrate that the Respondent followed a meaningful process of consultation before dismissing the Applicants in terms of sections 189 and 189A.

14. Lavery testified that he accepted the credibility of the deponents because they came to the Respondent to give information about the violent and criminal incidents. When asked whether he was not fearful that some of the persons who supplied information were supplying false information, he said: “It wasn’t, under the circumstances for people to come forward and give evidence of that nature exposes them to risk…”. Lavery even believed what he was told by a certain Mr. Wiseman Xhongo (“Xhongo”), one of the key perpetrators of serious criminal acts during the strike but who subsequently decided to assist the Respondent. Xhongo made statements to the Respondent in which he described some of the violent acts in chilling detail. Lavery conceded that he was not sure whether Xhongo had told them everything and in fact later watered his evidence down by stating that “certainly aspects of his testimony was credible” and “certainly in my opinion he didn’t tell us everything involving his own involvement in some of these issues but there were key parts of this testimony which I believe are credible”. I will return to the involvement of Xhongo in the disciplinary hearings hereinbelow. Suffice to point out that Xhongo was identified as the key witness for the Respondent. He, however, disappeared on the morning of the commencement of the hearings and was therefore not available to give evidence at the disciplinary hearings. It was mainly as a result of Xhongo’s disappearance that the Respondent then decided to abandon the disciplinary hearings and to proceed with a section 189 process.

15. On 16 June 2007 labour advisor Sydney Badenhorst (“Badenhorst”) met with management to give his views on the prospects of success on a case by case basis.

16. By way of a letter dated 9 July 2007, the Respondent gave the union notice of possible dismissals for operational requirements in terms of section 189(3) of the LRA. The employees likely to be affected by the retrenchment were “all employees who committed serious criminal actions during the recent strike”. The reasons for the proposed dismissal were described as follows:

“During the recent strike at Blue Ribbon Bakeries, a number of employees were allegedly involved in serious criminal actions, including but not limited to, assault, arson, intimidation and shootings. Their conduct makes it impossible for the company to continue to employ these employees as there is a significant threat of further violence. We are unable to take disciplinary action against these employees as witnesses are too scared to give evidence.”

17. On 11 July 2007 the Respondent requested the CCMA to facilitate consultations in terms of section 189A of the LRA. Facilitation meetings followed on 1 August 2007, 23 August 2007, 29 August 2007, 7 September 2007, 25 September 2007 and 2 October 2007. I will return to these meetings in more detail hereinbelow.

18. On or about 1 October 2007 the dismissed employees were notified that their services would be terminated on 31 October 2007. Their services were consequently terminated on that date.

Essence of the dispute

19. This brings me to the heart of the dispute between the parties. It was common cause that the Respondent decided not to hold disciplinary hearings against any of the individual Applicants but rather to dismiss the Applicants on the basis of operational requirements. One of the main reasons for this decision was (as already pointed out) the disappearance of one of the key witnesses (“Xhongo”) on the morning when the disciplinary hearings were scheduled to commence. In the circumstances the Respondent then decided to initiate consultations in terms of section 189 and 189A of the LRA relating to the proposed termination of the Applicants on the grounds of operational requirements. It was strongly in dispute whether or not the conduct of the strikers in question did indeed constitute an operational requirement (I will return to this point in some detail hereinbelow). Although some of the individual Applicants were identified as culprits in statements made by various individuals and submitted to the Respondent (to Elliott), the acts of violent misconduct allegedly committed by the individual Applicants have never been proven. In other words, allegations of involvement of the individual Applicants remained mere allegations.

20. The main contention on behalf of the Applicants was that the Respondent was not entitled to substitute the misconduct proceedings (which involve charging the employees with misconduct and requiring them to appear before a disciplinary hearing and proving their guilt) with a section 189 (operational requirement) procedure. I will point out hereinbelow that it was also the Respondent’s case that it was not possible to proceed with disciplinary hearings against the individual Applicants because witnesses disappeared and others were too afraid to testify. In the circumstances the Respondent therefore decided to abandon efforts to proceed with disciplinary hearing and rather to initiate consultations in terms of sections 189 and 189A of the LRA relating to the proposed termination of certain of Respondent’s employees on the grounds of operational requirements. The Respondent therefore submitted that the incidents of criminal violence posed a threat to the running of the Respondent’s business and that it therefore had no option but to resort to the retrenchment route to dismiss the Applicants.

The dispute according to the pleadings

21. The Applicants pleaded as follows:

    1. that the dismissals were automatically unfair in terms of section 187(1) of the LRA in that the Respondent desired to rid itself of union members whom it perceived as militant was the true reason for the dismissals.

    2. alternatively, that the dismissals were unfair in terms of section 188(1) of the LRA in that the reason for the dismissals was misconduct and the Respondent was not entitled to rely on section 189A to dismiss employees for reasons of misconduct. In essence it was argued that the use of a operational requirements procedure - where the dominant reason for the dismissals is in fact misconduct - is not permissible under the LRA and is per se unfair.

    3. further alternatively, and in any event, that the dismissals constituted substantively unfair retrenchments in that:

(a) they were not for a fair reason as contemplated by section 189A(19)(a);

(b) they were not operationally justifiable in terms of section 189A(19)(b);

(c) the selection criteria used were not fair and objective.

22. The Respondent denied that it retrenched employees in order to rid itself of union members who were militant and supportive of strike action. The Respondent further pleaded that the dismissals were for a fair reason based on its operational requirements, the reasons being that: “It had reason to believe that they had committed acts of serious criminal conduct”; “It was unable to take disciplinary action against them due to the fact that the witnesses to the acts of criminal conduct had been too afraid to testify”; “It feared that the individual Applicants would commit further acts of violence”; “It did not wish to continue to spend money on additional security”; “It wished to be able to focus its efforts on managing its business without the fear for further violence’; “The victims of the criminal conduct no longer wished to work with the individual Applicants”; and “Section 189 is a legitimate vehicle for terminating the employment of employees under circumstances where incidents of serious criminal conduct, which had a profound impact on the business of an employer, occurred and where it was impossible to take disciplinary action against such employees”.

Rationale for the dismissals: The Respondent’s version

23. I have already pointed out that it was the case of the Respondent that because the strike was marred by death threats; assaults; arson and intimidation against the persons and property of employees who did not participate in the strike, these incidents of criminal violence posed a threat to the running of the Respondent’s business. It was further submitted that as a result of these criminal acts (especially acts of death treats and intimidation of potential witnesses) the Respondent was unable to take disciplinary action against those suspected of complicity in these acts. The disappearance of their star witness also played an important role in deciding not to go the disciplinary hearing route. It was therefore submitted that the Respondent had no other option but to use the retrenchment procedure to dismiss them. Lavery gave detailed evidence about information received by the Respondent regarding the incidents of criminal violence. He referred to the information received about “attacks on people’s homes, ransacking of the homes, burning of cars, threatening to burn (the Respondent’s) vehicles”. He also testified that after the strike had started he also “got indications” that an assassination had been planned on him. Not surprisingly he was also extremely concerned for his safety. Although Lavery testified that he had “little doubt” that the incidents of criminal violence were related to the strike, he conceded in cross-examination that some were unrelated. He explained that these acts of criminal violence posed a serious threat to the Respondent’s business. He also gave evidence to the effect that the criminal acts would have an impact on future collective bargaining and future labour relations. The gist of his evidence was as follows:

“Absolutely, one of the key issues, and again we got information a little later on in the situation, was that there was further allegations of strike action being planned which would be more powerful than this one and that management were to be taught a lesson, to be frightened so that they wouldn’t try to keep the plant going should there be another strike. It was very clear to me that if there were no consequences to the actions taken during that strike we would have a very, very problematic industrial relations environment at the plant”.

Lavery also explained under cross-examination that these acts would have had an impact on the day-to-day running of the business:

“there was a strong likelihood in my opinion that had those people come back onto site that we would have had further violence, not only in terms of their presence back on site but also in terms of the relationship with the people that had suffered the violence and therefore based on that we could not continue to run the operation effectively and it was an operational requirement that we couldn’t continue the employment relationship with those people”.

These acts would also, according to Lavery, have had a negative impact on the future harmonious and inter-personnel relations between the staff. He explained this with reference to the fact that the victim of a petrol bombing would have to work together with co-workers who were implicated in the incident.

24. The Respondent led evidence to the fact that it had intended taking disciplinary steps against the perpetrators of the incidents of criminal violence. Arrangements were made for a chairperson from an employer’s organisation to chair the disciplinary hearings. For the more sensitive cases it was arranged that an independent labour advisor (Badenhorst) would act as initiator. The first disciplinary hearings would have involved five employees. They were, according to Lavery lumped together because they were the more senior people. In respect of the other people Badenhorst had been asked to manage the process and in doing so he would have regard to the various statements that had been given to the employer. Once Badenhorst had investigated the various statements he would draft and press charges against those which he believed were the perpetrators. During this process, Badenhorst interviewed the deponents of the statements to verify that what they had said was correct and to explain to them the procedures that would be followed. The deponents were also asked if they would be willing to testify at disciplinary hearings. The criteria used in determining provisionally whether a charge should be preferred against someone was that “there must be some evidence to the effect that the person did perpetrate whatever the event might have been”.

25. Badenhorst interviewed a total of twenty people who had given statements. Except for four individuals, all of them were employees of Blue Ribbon or its labour broker Staffgro. Three people who had given statements were not prepared to come and talk to him and a further three spoke to him but told him that they were not going to testify at a disciplinary hearing because “they were afraid for their lives, or being assaulted or whatever the case may be”. On Monday 11 June 2007, Badenhorst also interviewed Xhongo. This interview took place a few days before the first hearing at which Xhongo was scheduled to testify. In respect of Xhongo’s evidence Badenhorst testified that at that stage “having worked through all the witnesses that were prepared to testify, as well as two people who made statements but were not prepared to testify in any event, (he) came to the conclusion that there was only one star witness which I could use in the disciplinary hearing that gave me the (facts) around the total incident”. The following appears from the statements obtained from Xhonga: He confirmed that striking workers destroyed the home of Nothula Makaleni; attacks were planned on Blue Ribbon Trucks; striking employees were responsible for the burning of Zoar Mdlaleni’s car; striking employees wrecked Sage Jooste’s house; striking employees made petrol bombs which were then used in the attacks on the homes of Charmaine Smuts and Vivienne Tywala and that money was collected from employees in order to hire someone to assassinate Lavery. The statement of Xongo is, as already pointed out, a chilling account of what had transpired and what was planned during the strike (assuming that the statements were true). Given therefore the importance of his evidence, arrangements were then made by the Respondent to have Xhongo held incommunicado, and to have him transferred to a non-unionised workplace in the Eastern Cape after he had given evidence. At 05h00 on the morning of the first disciplinary enquiry at which Xhongo was scheduled to testify, Mr. Lambert of the Respondent (“Lambert”) called around at the caravan park where Xhongo was staying only to find that he had disappeared. Staff at the caravan park confirmed that Xhongo had a lengthy telephone conversation on the public payphone the previous day and that he had been visibly upset and shaken thereafter. To this day the Respondent does not know whether Xhongo is alive or what had became of him. The Respondent submitted that it is overwhelmingly probable that someone had threatened the safety of Xhongo and his family in a sufficiently serious manner to cause him to disappear.

26. Badenhorst was asked about the witness statements of the witnesses that he regarded as key witnesses and what the distinguishing factors were in respect of their evidence:

“All (the) evidence, or the statements that were given, most of them, if not all of them, were hearsay evidence. And second-hand information that was passed on. It wasn’t people actually there to observe somebody doing specifically that. There were other people outside of them, like the family members or friends, or whatever, that could physically identify people. And they related that to them, and they used that in their statements”.

27. Badenhorst testified that neither Charmaine Smuts, Zoar Mdlaleni, Vanecia Bowers or Nokutula Mdlaleni were considered to be key witnesses and that of the three witnesses only Xhongo was prepared to talk to him and to testify at the hearing. The witnesses Makeleni, Mdleleni and Smuts, who testified in this Court about the acts of violence against them, indicated that they would have been prepared to testify at the disciplinary enquiry, but that they could not give evidence of any value as none of them were able to identify the persons who had attacked their homes or cars. The Respondent’s case was therefore that absent the evidence of Xhongo, there was virtually no admissible evidence upon which a case could be made out against any of the transgressors save for one or two. According to the Respondent the fact that Xhongo had disappeared had a material impact on the possible success of the disciplinary hearings.

28. Lavery and Badenhorst subsequently had a discussion during which Badenhorst said to Lavery “that his predicament was that although he had a lot of hearsay evidence he did not have people that could actually testify specific issues and specific incidences that had taken place and it was his advice that at the time he didn’t think if those disciplinary hearings were challenged that we would have much chance of succeeding in court.” Lavery also testified that as a result of the disappearance of Xhongo, the Respondent had concerns about the safety of the other witnesses. If the evidence of Badenhorst is perused it would appear that he was more concerned with the success of the disciplinary hearings than with the safety of the witnesses. Badenhorst was asked where the disappearance of Xhongo left him in respect of the enquiries against other employees. He answered as follows: “I had no case, except circumstantial” and “All their evidence, or the statements that were given, most of them, if not all of them, were hearsay evidence. And second-hand information that was passed on. It wasn’t people actually there to observe somebody doing specifically that. There were other people outside of them, like the family members or friends, or whatever, that could physically identify people. And they related that to them, and they used that in their statements”.

29. Badenhorst subsequently explained to the management staff at a meeting that “the role that Wiseman would play in the total hearing and the successful prosecution of the case, and that without him we didn’t have a case in any of the other complaints that were laid”.

30. It was put to both Lavery and Badenhorst that disciplinary hearings could have been held in respect of those individuals who were suspected of complicity in the various incidents of criminal violence. Lavery specifically testified that it was their inability to hold these hearings which resulted in them considering the operational requirements route. He reiterated that the Respondent was faced with a situation in which their witnesses had disappeared as well as with a situation where witnesses were not prepared to give evidence and that the Respondent was concerned about their safety. Badenhorst was specifically asked in cross-examination why he had not proceeded with disciplinary proceedings and why he did not merely rely on the statements which he had. Badehorst replied that in the disciplinary hearing that he had chaired in the past, the principle was that the other side had to be heard. He was also asked what had stopped him from trying to convince the chairperson of the hearing that he should take into account the evidence of Xhongo in the form of a sworn affidavit. Badenhorst responded as follows:

“As I indicated to you, I do not believe that in disciplinary hearings you can use documentary evidence only and no supportive evidence”.

31. Mr. Kahanovitz challenged Badenhorst on this statement and put it to him that Badenhorst did have sufficient evidence which could have corroborated what was in Xhongo’s sworn affidavit. Badenhorst conceded that he had witnesses who could have come and testify about the fact that a house was attacked on a particular day and what happened at the house. This was confirmed by witnesses Charmaine Smuts, Zoar Mdlaleni, Nokuthula Makaleni and Nosamdile Sindelo. All of them were able to testify as to incidents of criminal violence although they could not identify the perpetrators. It was further put to Badenhorst that the sworn affidavit of Xongo gave him “a basis to put up a case that needs to be answered”. Badenhorst respondent by saying: “With very little success, I suppose you’re right, yes”. He also conceded that once the alleged perpetrators were before such an enquiry they would have had to answer to the allegations against them and would have had to give a version. The initiator would then have had an opportunity to cross-examine them. Badenhorst was also specifically asked that “if one says that what Mr Xhongo had to say constitutes evidence which is good enough to retrench people, why should that evidence not be good enough for a disciplinary hearing?” Badenhorst responded as follows: “Because the principle still remains that you’ve got to prove what you’re saying, and if you don’t have the evidence to prove it other than hearsay evidence, or circumstantial evidence, you don’t have much of – success”. Badenhorst was then asked the following question: “But if it’s not fair to dismiss people in disciplinary hearings with evidence of that quality, why is it fair to dismiss them in retrenchment with evidence of that quality? He answered: “I cannot comment on that”. It was also put to Lavery in cross-examination that disciplinary proceedings “...do not have to be particularly sophisticated. All that you have to do is tell the employee what the allegations are against them and give them an opportunity to give a version as to what they have to say about the allegations”. Lavery stated that he understood this but went on to say that the Respondent could not disclose the identity of witnesses to the alleged perpetrators and also could not disclose details of the incident itself or the victims. He said: “I think any divulging of any sort of information relating to (the charge) would then identify the victim”. It was put to him that such details would inevitably have been revealed in the course of polygraph testing but he responded that it would suffice to ask a person “Did you on the 29^th burn somebody’s house”. Lavery then stated the following: “the key thing is that you would not have a scenario where the (victim) who didn’t want to give evidence would have to face…the perpetrators of that violence face to face in a hearing”. Lavery also stated that his major concern at that stage wasn’t only disappearance of witnesses or the refusal of witnesses to give evidence but that the 189 process (because of the facilitation process) was less antagonistic. He also believed that they would have had fewer problems if they had followed the 189 process as opposed to following a disciplinary process. It was also put to Lavery that “the trigger for the decision to go the section 189 route” was the conversation where he had formed the view that the risk attached to going the misconduct hearing route was that “if those people were dismissed there was a strong chance that the CCMA might overturn those findings”. Lavery conceded “[t]hat [this] was one of the factors, yes” but stated that although the situation of having the disciplinary processes overturned was an issue, the safety issues were of more importance. Lavery, however conceded that “[o]ur case was basically built around three key bits of evidence or three key witnesses. Two of them had disappeared and one of them had said he’s refusing to give evidence”. On behalf of the Applicants it was argued that this amounted to a concession that the key factor was the loss of key witnesses rather than the concerns for the safety of the body of witnesses.

32. One witness, even on the Respondent’s own evidence, was willing to testify in respect of some incidents of criminal violence. Mr. Willem Kruger (“Kruger”), a trained security officer hired to provide security services deposed to an affidavit in which he stated that he was able to identify the perpetrators of a particular incident. Lavery was asked why the Respondent did not run a disciplinary hearing with his evidence. He answered: “The circumstances that we faced at that stage was not a single perpetrator of violence. It was a group of people, and usually problematic acts of misconduct that had occurred”. It was put to Lavery that there could not have been any reason why a security officer who witnessed threats to set a company truck alight, could not have been called to a disciplinary hearing. He answered: “Should that have been an individual case on its own, yes, I would have agreed with you”. Badenhorst was asked the same question; He replied that “[t]here was sufficient evidence in this specific case, which is an isolated case, of a person that was a stand-alone witness to certain events”. He couldn’t explain why management did not proceed with this case. He said that he had indicated to management at the meeting of 16 July that there were “one or two” cases where he believed that there were good prospects of success.

33. Lavery was also challenged by Mr. Kahnovitz in respect of the evidence that the Respondent could not proceed with the hearings because of safety concerns. “Mr Kahanovitz: But if that version was true then you would never have announced that there were going to be disciplinary hearings in the first place…The key factor for me which resulted in us changing that approach was the disappearance of the two key witnesses”.

34. Reference was made to the fact that the Respondent brought an urgent application to interdict the strikers. Mr. Kahnowitz took issue with Lavery because he had been prepared to give a mandate to the Respondent’s attorneys to prepare an application to the Labour Court based upon the evidence of a number of witnesses who implicated employees in serious crimes (but who were not prepared to give evidence in a hearing). Lavery confirmed that this was the case. It was also put to Lavery in cross-examination that nothing had been done to test the credibility of the informants on which the Respondent had relied. Lavery simply stated that, in their view, their witnesses were credible. He also stated that once they have decided to follow “a no-fault operational requirements process” they did not deem it appropriate.

Automatically unfair dismissals

35. It was submitted that the dismissals were automatically unfair because the employer, in dismissing the employees, acted contrary to section 5 of the LRA. It was argued that the true reason for the dismissals was the Respondent’s desire to rid itself of members whom it perceived as militant and supportive of strike action. The Respondent disputed this and contended that the dismissed employees were identified for dismissal on the basis of affidavits implicating them in serious criminal conduct. I do not intend to dwell on this point in light of Mr. Kahnowitz submission that the Applicants were not pursuing this point.

Unfair dismissal

36. I will now turn to the more difficult question and that is whether an employer may resort to a section 189 (and 189A) process in circumstances where it is unable to conduct disciplinary hearings in order to prove a charge of misconduct. I have already pointed out that it was the Applicants’ case that, in principle, an employer cannot substitute the misconduct proceedings with a section 189 and 189A procedure.

Was a hearing possible in these circumstances?

37. Before I turn to the general question whether or not an employer may circumvent misconduct proceedings and proceed with a dismissal on operational requirements (and also whether the Respondent established an operational requirement in this instance), it is, in my view, necessary to first deal with the issue whether or not it was possible for the Respondent to have held disciplinary hearings in this particular case.

38. The importance of disciplinary hearings in misconduct cases cannot be understated. The Appellate Division (as it then was) in Administrator, Transvaal & Others v Zenzile & Others² summarized the applicable principles as follows:

“And when, as here, the exercise of the right to dismiss is disciplinary, the requirements of natural justice are clamant. Mureinik (1985) 1 SAJHR 48 points out at 50 that –

'perhaps pre-eminent amongst the qualities of a power that attracts natural justice is its susceptibility to be characterised as "disciplinary" or "punitive’.

The learned author explains that the reasons for this are rooted both in history and in principle; but that the latter are crucial. At 50-1 he summarizes the reasons of principle thus:

‘Where the power is disciplinary, all the usual reasons for importing natural justice generally apply, and generally apply with more than the usual vigour: the gravity of the consequences for the individual, consequences both concrete and such as affect his reputation; the invasion of the individual's rights; that fairness postulates inquiry; and so on. But more than this, there is a reason of principle peculiar to disciplinary or punitive proceedings: that even if the offence cannot be disputed, there is almost always something that can be said about sentence. And if there is something that can be said about it, there is something that should be heard. . . .’

It is trite, furthermore, that the fact that an errant employee may have little or nothing to urge in his own defence is a factor alien to the inquiry whether he is entitled to a prior hearing. Wade Administrative Law 6 ed puts the matter thus at 533-4:

‘Procedural objections are often raised by unmeritorious parties. Judges may then be tempted to refuse relief on the ground that a fair hearing could have made no difference to the result. But in principle it is vital that the procedure and the merits should be kept strictly apart, since otherwise the merits may be prejudged unfairly.’

The learned author goes on to cite the well-known dictum of Megarry J in John v Rees 1970 Ch 345 at 402:

‘As everybody who has anything to do with the law well knows, the path of the law is strewn with examples of open and shut cases which, somehow, were not; of unanswerable charges which, in the event, were completely answered; of inexplicable conduct which was fully explained; of fixed and unalterable determinations that, by discussion, suffered a change.”

39. I am not persuaded that the Respondent was not able to hold the disciplinary hearings. Firstly, the mere fact that an employer cannot prove a charge does not allow it to follow the section 189 route. Put differently, an employer cannot as a matter of principle or as a matter of expedience resort to section 189-procedures in misconduct cases. I will leave out for a moment the next question namely whether or not a section 189 route may be followed where genuine operational requirements so demand. Secondly, I have referred to the evidence of Lavey and Badenhorst in some detail. It is clear from their evidence that they were not convinced that they would be successful in proving the charges without Xhongo. They had according to Badenhorst no case “except circumstantial”. He was also not keen to use the various witness statement without calling the deponents because it would have amounted to hearsay evidence. Although, as I have also indicated, the Respondent did not hesitate to attach these statements to the urgent application.

40. I am therefore not persuaded that the Respondent was not in a position, despite the safety considerations, to hold disciplinary hearings in this case. It is simply not acceptable for an employer to decide that, because I cannot prove the allegations in a disciplinary hearing, I am proceeding to dismiss by using a process where it is not necessary to prove the guilt of the accused employee. I am also not persuaded by the Respondent’s claim that it was not able to proceed with the disciplinary hearings because of safety considerations and because witnesses were too scared to testify. If this was the case, why did the Respondent institute disciplinary hearings in the first place? In other words, if safety considerations were the overriding factor for deciding not to hold disciplinary hearings, the Respondent would have opted for the operational requirement dismissal route from the beginning. The Respondent was aware from the beginning that some witnesses were not prepared to testify.

41. I have already pointed out that it was argued that the various statements from the witnesses should be received by this Court in evidence even though certain hearsay statements are contained therein. Section 3 of the Law of Evidence Amendment Act,³ affords a Court a discretion to admit hearsay evidence if the Court is satisfied that it should do so having regard to the nature of the proceedings; the nature of the evidence; the purpose for which the evidence is tendered; the probative value thereof; the reason why the evidence is not given by the person who has direct knowledge thereof; any prejudice to the other party which the admission thereof might entail and any other factor which is, in the opinion of the Court, relevant. See Hlongwane & Others v Rector: St Francis College & Others.⁴Arbitrators and CCMA Commissioners are in particularly entitled to admit hearsay evidence depending on the particular circumstances in the proceedings before the CCMA (see Swiss South Africa (Pty) Ltd v Louw NO and others⁵). See also Southern Sun Hotels (Pty) Ltd v SA Commercial Catering & Allied Workers Union and Another⁶ where the Court sanctioned reliance on hearsay evidence to prove a serious offence in circumstances where it was alleged that the witness was unwilling to testify due to intimidation:

“If one admits the hearsay evidence, then the prejudice to the (accused employee) is very serious because the evidence goes to the heart of his defence, and, yet, he may well be innocent. However, also, if the evidence is excluded, then the (employer) and (the victim) would suffer serious prejudice because (a) the appellant has no way of proving the guilt of the (accused employee) on what is clearly a very serious offence in circumstances where the (accused employee) may well be guilty, and (b) to exclude the hearsay evidence may well play into the hands of bad elements in the workplace or in society in that it may mean in effect that people can indulge in all kinds of acts of misconduct or criminal conduct with impunity if they ensure that complainants and witnesses to their deeds are either killed or intimidated into not coming to court to testify against them. That is totally unacceptable and is an evil because, if it becomes part of our life, it will destroy the very foundations on which our society is built.”⁷

42. The Court Ngcobo v Durban Transport Management Board⁸ also approved of the decision of the employer to rely on written statements in circumstances where witnesses were too afraid to testify. The Court also acknowledged that “it is not a rare occurrence for … employers to take disciplinary action on the strength of facts conveyed to them in written statements”:

“This absence of viva voce evidence will not necessarily in itself make the enquiry unfair, and may be adequate in the circumstances”.⁹

43. A similar approach was followed in Food & Allied Workers Union & Others v G Smith Sugar Ltd, Noodsberg¹⁰ where employees had been dismissed pursuant to a disciplinary enquiry in which certain of their accusers refused to have their identities disclosed. The accused employees were not given an opportunity to confront or cross-examine the witnesses. The Industrial Court accepted that the “requirements” of pre-dismissal procedure are “not rigidly or inflexibly enforced”:

“The overall test in relation to predismissal procedure is not whether some or other set procedure was followed but what fairness required. The true question is whether the hearing was substantially fair when the proceedings are judged in their broad perspective”.¹¹

The following comments by the Court are of particular relevance to the present case: Absurd consequences might result if an employer could not dismiss guilty employees because they intimidated witnesses:

“An employee’s conduct may be such that no reasonable person could expect an employer to continue employing him. Is he to be obliged to do so merely because guilt cannot be established in a procedure akin to the adversary system followed in a court of law? Must an employer continue to employ someone in charge of a children’s home or an old age home who so bullies and terrifies the residents that they are too afraid to complain openly? The circumstances may be such that guilt can be established only by making use of hearsay evidence.

In this particular case the individual applicants knew the details of the factual allegations against them. They were able to deal with the facts. The fact that use was made at the enquiry of statements from unidentified witnesses and that no opportunity was given to put questions to or to cross-examine such witnesses did not prevent the applicants from producing their evidence nor of correcting or contradicting prejudicial statements”.¹²

The Court then set out the test that should be applied when deciding whether or not written statements should be admitted into evidence:

“That the rule of thumb by which an adjudicator decides whether to admit or exclude an oral or written statement tended as evidence should be whether the statement is relevant, reliable and logically probative and of such a nature that responsible people would rely upon it in serious affairs”.¹³

44. See also Marutha v Sember CC t/a Review Printers, Pietersburg¹⁴where the Industrial Court found that the employer was justified in not calling witnesses in circumstances where they feared for their safety.¹⁵ See also Shishonga v Minister of Justice & Constitutional Development & Another¹⁶ where the Labour Court, with reference to numerous case law, gave a detailed exposition of what constitutes hearsay evidence and under which circumstances should it be admitted and why hearsay evidence should generally not be admitted:

“[149] The constitutional right to a fair trial is at the heart of the question as to whether hearsay evidence is admissible. Hearsay is inadmissible because it cannot be tested by cross-examination and is therefore unreliable.”

The Court then explains under which circumstances hearsay evidence will be admitted:

“[150] The singular consideration for the admissibility of hearsay under s 3(1)(c) Law of Evidence Amendment Act is the interests of justice. The interests of justice are not dependent on whether the declarant testifies. Nor is the disavowal or non-confirmation of a statement enough to prevent it from being admitted, if it is in the interests of justice to do so. Its reliability can be weakened if it is disavowed or not confirmed. Prejudice which is always present when hearsay is admitted against a party will not usually outweigh the interests of justice. To outweigh the interests of justice it will have to be prejudice of the kind suffered by accused if the prosecution is allowed to reopen its case to lead hearsay evidence after the accused have closed their case.

[151]Safeguards must be applied to ensure a fair trial whenever hearsay is tendered. What would be appropriate safeguards could be different for criminal and civil trials. In criminal cases where an accused against whom the statement is sought to be used is unrepresented, the court must exercise greater caution.

[152]These are civil proceedings in which the parties are legally represented and are themselves legally trained.”

45. I am in agreement with the approach followed in the aforementioned cases. It is possible to proceed with a disciplinary hearing on the basis of written statements in circumstances where witnesses are too scared to testify. To exclude hearsay evidence in circumstances such as those which prevailed in this case will only play into the hands of people who have the attitude that they can do as they please without impunity. Allowing individuals to get away with their acts of misconduct simply because they intimidate potential witnesses “will destroy the very foundations on which our society is built”. In the present case, having had the benefit of hearing all the evidence I have little doubt that the chairperson would have allowed the hearsay evidence. In this regard the following factors would have been persuasive in allowing the hearsay evidence: (i) It was not suggested that Elliott was untruthful in her evidence as to how she had compiled the statements or that she had misunderstood the information that was conveyed to her. It was also not disputed that she was specifically requested to keep the identify of the employees secret. The applicants therefore would have had to attack the credibility of Elliot. It had not done so in Court and I am not persuaded that it would have been able to do so at the hearing. (ii) The circumstances under which the statements were received are crucial. It was not disputed that this was a particularly violent strike and that there was a fear of harm and victimization. These circumstances are, in my view, compelling in admitting the hearsay evidence. (iii) Lastly, and most importantly, it would have been in the interest of justice for the chairperson to have admitted these statements into evidence. It cannot be disputed or questioned that an employer (the Respondent) is expected to act on the serious allegations contained in the statements because failing to act on these horrific acts would mean that threats of violence will in the future provide an opportunity to violent strikers to avoid being called to justice before a disciplinary hearing.

46. It is important to also stress that, in the present case, the hearsay evidence would have been admitted in the context of a disciplinary hearing. A disciplinary hearing is not a criminal trail. It is also not a civil trail. A disciplinary hearing is an opportunity afforded to the employee to state a case in response to the charges leveled against him her by the employer. See in this regard Avril Elizabeth Home for the Mentally Handicapped v CCMA & Others¹⁷ where the Labour Court, in great detail explained what the purpose of a disciplinary hearing is. The Court also emphasised that a disciplinary hearing should not be equated to a criminal trail:

“It follows that the conception of procedural fairness incorporated into the LRA is one that requires an investigation into any alleged misconduct by the employer, an opportunity by any employee against whom any allegation of misconduct is made, to respond after a reasonable period with the assistance of a representative, a decision by the employer, and notice of that decision.

This approach represents a significant and fundamental departure from what might be termed the 'criminal justice' model that was developed by the industrial court and applied under the unfair labour practice jurisdiction that evolved under the 1956 Labour Relations Act. That model likened a workplace disciplinary enquiry to a criminal trial, and developed rules and procedures, including rules relating to bias and any apprehension of bias, that were appropriate in that context.

The rules relating to procedural fairness introduced in 1995 do not replicate the criminal justice model of procedural fairness. They recognise that for workers, true justice lies in a right to an expeditious and independent review of the employer’s decision to dismiss, with reinstatement as the primary remedy when the substance of employer decisions are found wanting. For employers, this right of resort to expeditious and independent arbitration was intended not only to promote rational decision making about workplace discipline, it was also an acknowledgement that the elaborate procedural requirements that had been developed prior to the new Act were inefficient and inappropriate, and that if a dismissal for misconduct was disputed, arbitration was the primary forum for determination of the dispute by the application of a more formal process.”¹⁸

The Court continues to say the following about the quality of evidence that may be admitted in the process:

“On this approach, there is clearly no place for formal disciplinary procedures that incorporate all of the accoutrements of a criminal trial, including the leading of witnesses, technical and complex ‘charge sheets’, requests for particulars, the application of the rules of evidence, legal arguments, and the like.¹⁹

The nature and extent of the fair procedure requirements established by the Labour Relations Act and the Code is supported by international labour standards. International Labour Organisation Convention 158 requires procedures to promote compliance with the obligation to ensure that dismissals are based on valid reasons.”²⁰

47. In light of all the circumstances and the evidence before the Court I am of the view that the Respondent was able to conduct the hearings and that it would have been in the interests of justice to allow the hearsay evidence. In the event, the dismissal of the individual Applicants is found to be substantively and procedurally unfair. I will return to the remedy hereinbelow.

48. I will now proceed to decide the dispute also in the alternative in the event that I am wrong in my finding that the Respondent ought to have followed through with the misconduct (disciplinary hearings) proceedings.

Section 189 procedure

49. The question that must now be considered is whether or not the conduct of the non-strikers constituted an operational requirement which permitted the Respondent to proceed with dismissing the Applicants on the basis of operational requirements

50. The LRA provides for three categories of dismissals (where the dismissal is not automatically unfair). It would appear from a reading of section 188(1) of the LRA that the three categories of dismissal are premised on the fact that it is recognised that there are different reasons for a dismissal. Section 188(1) of the LRA provides that a dismissal will be unfair if the employer fails to prove –

“(a) that the reason for dismissal is a fair reason-

(i) related to the employee’s conduct or capacity; or

(ii) based on the employer’s operational requirements; and

2. that the dismissal was effected in accordance with a fair procedure.”

51. As already pointed out, the Respondent pleaded that the dismissals were for a fair reason based on its operational requirements. The reason specifically pleaded referred to the operational impact of strike-related misconduct on the workplace in circumstances where it was, according to the Respondent, impossible to take disciplinary action against the suspected perpetrators.

52. In terms of section 189A(19) an employee will be found to have been dismissed for a fair reason if-

“(a) the dismissal was to give effect to a requirement based on the employer's economic, technological, structural or similar needs;

(b) the dismissal was operationally justifiable on rational grounds;

(c) there was a proper consideration of alternatives; and

(d) selection criteria were fair and objective”.

It is trite that in terms of section 192(2) of the LRA, the onus of proving that these requirements have been met will lie with the Respondent.

53. The question that this Court is called upon the consider is whether or not section 189 (the so-called retrenchment procedure) may be used where misconduct is the reason. Put differently: Can an employer resort to section 189 procedures where it is the employer’s case that it is unable to hold disciplinary enquiries as a result of violence, disappearance of crucial witnesses and where witnesses are being intimidated and to afraid to testify. I pose the question in this way in light of the pleadings of the Respondent and the evidence obo the Respondent as to why it decided not to proceed with disciplinary hearings. This question in my view, appears to stand on three legs:

(i) Firstly, in principle, can an employer decide to substitute misconduct proceedings with section 189 operational requirement proceedings? And if so, under what circumstances?

(ii) Secondly, if the answer is yes, was the employer in this case entitled to circumvent or abandon misconduct proceedings and proceed with operational requirements?

(iii) Thirdly, if the answer to the previous question is yes, was the dismissal on the basis of operational requirements fair.

54. In terms of section 189A(19), an employee will be found to have been dismissed for a fair reason if-

“(a) the dismissal was to give effect to a requirement based on the employer's economic, technological, structural or similar needs;

(b) the dismissal was operationally justifiable on rational grounds;

(c) there was a proper consideration of alternatives; and

(d) selection criteria were fair and objective”.

The Respondent (employer) bears the onus of proving that these requirements have been complied with.²¹

55. Section 213 of the LRA gives a definition of the term “operational requirements” in the context of section 189 of the LRA. It refers to the “requirements based on the economic, technological, structural or similar needs of an employer”. The LRA does not, however, provide for a definition of the terms “economic”, “technological” or “structural”. The Code of Good Practice: Dismissal Based on Operational Requirements²² (the “Operational Requirements Code”) defines economic reasons as “those that relate to the financial management of the enterprise” and structural reasons as those that “relate to the redundancy of posts consequent to a restructuring of the employer’s enterprise”. On behalf of the Respondent it was argued that although section 203(3) requires the Court when interpreting or applying the LRA to take into account any relevant Code of Good Practice, section 1 of the Operational Requirements Code cannot be regarded as introducing an all-encompassing definition of the terms economic, technological and structural, as used in the definition of “operational requirements” contained in section 213 of the Act. The Respondent therefore submitted that in light of the fact that the Operational Requirements Code does not comprehensively define the terms economic, technology or structural, the Court should interpret the LRA so as to give effect to the objects of the LRA (including those set out in section 1) of the LRA. More in particular, this Court should give content to the phrase “or similar” and in doing so should apply a purposive approach. A Court will, so it was submitted, not give a judgment in terms of which strikers are allowed, without impunity, to terrorise and harm non-striking co-workers. The Respondent argued that the fact that a particular operational requirement involves elements that could be resolved by recourse to misconduct or incapacity proceedings does not necessarily mean that the employer is precluded from initiating the consultation process envisaged by sections 189 and 189A.

56. I have already indicated that I am in agreement that strikers should not be allowed, without impunity, to terrorise and harm non-striking workers. I have also indicated that I am of the view that in the present case the employer (Respondent) could have proceeded with the misconduct route. The question here is whether or not the Respondent was able to circumvent the misconduct route by resorting to the operational requirements route. That will, of course, depend on whether or not the Respondent was able to persuade this Court that it had an “operational requirement”.

What does the case law say?

57. There is authority for the proposition that an employer may resort to dismissals on the basis of operational requirements in circumstances where the operational requirements of the business so requires but where misconduct prompted or underlies the dismissals. This will usually be in cases where a company suffers stock losses as a result of pilferage and obviously suffers financial loss. In other words, although misconduct underlies or prompted the operational requirements to dismiss, the actual or dominant rationale for the dismissal is the operational requirements or survival of the business. In SA Commercial Catering and Allied Workers Union and others v Pep Stores (1998) ILJ 1226 (LC) the Labour Appeal Court accepted as a valid operational requirement warranting retrenchment the fact that a retail store was suffering massive stock losses due to pilferage and that the employees appeared unable to protect the goods in their custody. In this case the employer notified the union and employees that two of its branches would be closed for an indefinite period during which the employees would receive their full pay. It was common cause that the reason for the closures was the severe stock losses the company was suffering at these branches. The Court accepted that the company had shown good cause to shut down the two branches because they were not profitable and that the reason for that was the unexplained stock losses (shrinkage). The Court accepted that that was a sufficient reason to close the branches for operational requirements.

58. Mr. Oosthuizen submitted with reference to this case that this was clearly a case where the operational requirements that existed could have been addressed by way of a collective misconduct.

59. Is the decision in Pep distinguishable from the present matter? I am of the view that it is. In the Pep matter the “dominant purpose” for the consultations was to search for a solution to the shrinkage problem: “If no solution was found, it followed that the branches would close and there would be retrenchment, unless there a suitable alternative to retrenchment”.²³ In other words, the purpose was to save the business from financial failure. I was Mr. Kahnowitz’s argument that the fact that the financial problems were created by misconduct (someone obviously was stealing the company into bankruptcy but could not be identified) does not detract from the fact that the purpose of the retrenchment was to prevent the closure of the business (which is an operational requirement of the business). I am in agreement with this submission. This is also clear from the following words of the LAC (see the decision in Chauke – infra):

“It posits a justification on operational grounds, namely that action is necessary to save the life of the enterprise.”

The operational requirements in the Pep-case clearly were the financial losses that the company was suffering. A process of consultation was then embarked upon in order to find a solution to the stock losses. The present case is different. The employer postulates mainly its justification on misconduct and because it cannot successfully prosecute the culprits therefore it proceeded with the operational requirement route. The reason for the dismissal in the present matter is therefore clearly misconduct and not financial.

60. The Court was also referred to the decision in Food Brands Limited t/a Albany Bakeries v Levy N.O. & Others.²⁴ In that case, a case substantially on all fours with the present matter the Labour Court held that a section 189A facilitation process could be used to initiate a consultative process where the employer sought to terminate the employment of employees suspected of misconduct in the form of involvement in knowledge of an assassination attempt. In that case the Commissioner declined to facilitate under section 189A, on the basis that section 189A was intended for use in the case of no fault dismissals. The employees in that case were also involved in a particularly violent strike. The striking employees attacked replacement labourers with knobkieries and an assassination attempt was made on the life of one of Albany’s senior management. It was also not possible for the employer to identify the persons involved in the assassination attempt and hence it was submitted that it was impossible to have resorted to disciplinary action. The employer in the Albany-matter formed the opinion that it was no longer possible to manage the bakery due to the incidents. Management also feared for their safety. The employer formed the opinion that its operational requirements required that the employment relationship between it and the group of employees who may have been involved or have known of the assignation attempts or threats be terminated hence the application in terms of section 189A(3) of the LRA for the appointment of a facilitator. At the commencement of the facilitation process the point was raised by the union that the proposed dismissals did not fall within the definition of operational requirements as set out in the LRA. The Commissioner upheld the point. It was this ruling that was the subject of the review application. The union (FAWU – the same union involved in this case) raised the point that the problem could have been dealt with by way of disciplinary or criminal proceedings. The Respondent submitted that it was unable to proceed with individual disciplinary proceedings as it was unable to identify the culprits.

61. The Court in Tiger Foods referred to section 213 of the LRA which defines “operational requirements” as meaning requirements based on the economic, technological, structural or similar needs of an employer. The Court then referred to the Code of Good Practice where an attempt is made to clarify the meaning of these terms:

“[14] Section 213 of the LRA defines the operational requirements as meaning requirements based on the economic, technological, structural or similar needs of an employer. The Code of Good Practice attempts to clarify the meaning of economic, technological or structural needs of the employer. The code acknowledges that it is difficult to define all the circumstances that might legitimately form the basis of a dismissal for that reason. The code defines economic reasons as those that relate to the financial management of the enterprise. Technological reasons are defined as those that refer to the introduction of a new technology which affects work relationships either by making jobs redundant or by requiring employees to adapt to the new technology or a consequential restructuring of the workplace. Structural reasons are those that relate to the redundancy of posts consequent on restructuring of the employer's enterprise. What the code does not attempt to define is the all encompassing term being 'similar needs of an employer'.”

62. The Labour Court set aside the Commissioner’s ruling holding that even when the employer did not require to downsize, the use of section 189A was still permissible because section 189A(19)(a) referred to “requirements based on the employer’s economic, technological, structural or similar needs”. The Code of Good Practice “does not attempt to define the all encompassing term being ‘similar needs of employer’”.

63. It was also argued in that case that the employer needed to be able to manage it business and that it was not able to do so where managers were being assassinated and direct threats were made against them. The Court concluded as follows:

“[28] The applicant needs to be able to manage its business in order to be able to turn it around. It has to deal with the safety of the managers and be able to control access to the bakery. It is not able to do this if the managers are being assassinated and direct threats are made against them. How does the employer protect its own management team and also be in control of the business faced with violent resistance? Management needs the workforce, which will be able to work on public holidays, and be able to finish unfinished work before knocking off. With all these problems in mind, does the CCMA have jurisdiction to facilitate the dispute? The answer lies in answering whether the problems the applicant is facing constitute the employer's operational requirements.”

….

“[38] The need to get the business of the applicant going again on a permanent basis and in a stable environment is the prime consideration. When managers are being threatened with death, the applicant cannot operate its business. It has a duty to protect its managers. At the same time, the employees have to be fairly treated. The need for stability cannot be dismissed as not being an operational reason or economic reason for retrenchment.”²⁵

64. The Court accepted that the inability of the employer to manage its business does affect the “economic viability of the enterprise” The Court then proceeded to interpret “similar needs of an employer” as set out in the Retrenchment Code and concluded that it “relates to the needs of the employer that have some resemblance of economic, technological or structural”.

65. The Court in the Tiger Foods-case thus clearly recognized that the reason for the dismissal must ultimately be the economic viability of the enterprise and that the reasons must relate or have some resemblance to the economic, technological or structural needs of the business. In this sense this judgment does not differ much from what Landman, J concluded in Pep-stores. As long as the employer can prove that the dominant purpose of the retrenchment route is the economic viability of the enterprise, the employer may well be entitled to go the section 189 route depending, of course, on the particular circumstances of the case. This was, in my view, confirmed by the Court in Tiger Foods as follows:

“[39] This does not mean that for any misconduct, the employer may decide not to have the employee dismissed for operational reasons. It will depend on the facts of the case. In the present case, the employer is faced with problem of having to turn around its business because of losses. It is met with violent resistance in which the managers are at a risk of being killed and the perpetrators cannot be identified.²⁶

[40] I am satisfied that it was proper and legitimate for the applicant to request the facilitation. There is an economic reason or reason similar to that for the anticipated retrenchment. If there is a solution or suggestion that can assist in the avoidance of the dismissal, that is an issue to be dealt with at the facilitation hearing. The disclosure of the perpetrators may assist the applicant in stopping the retrenchments and commencing disciplinary proceedings. The ultimate result required by the employer is the protection of its business and its management from criminal actions.”

66. I am therefore not persuaded that an employer can never opt for the section 189 –route where misconduct triggered the operational rationale. The section 189 route will, however, in my view, not be available to an employer simply because he cannot prove the charges against the employees. However, where the employer can prove that the misconduct affects the economic viability of a business (such as massive shrinkage – as in the Pep- matter) or where the misconduct prevents an employer from turning around its business because of the losses (Tiger Foods-case), the employer may well be able to proceed via a section 189-process. The section 189 process, although prompted or caused by misconduct, is about the continued economic viability or survival of the business. The reason for the dismissal ultimately is not about the misconduct; the reason for the dismissal is the operational requirements of the business. Whether or not an employer will be able to opt for the section 189 route will, however as already pointed out, depend on the particular circumstances of each case. I must also point out that it is envisaged that the Court will not readily accept that an employer is entitled to opt for the section 189 procedures where misconduct prompted the process. This is clear from both the Pep- case and the Tiger Foods- case. In both of these cases the Court was at great pains to point out what the operational needs of the employer were and why it was therefore entitled to proceed via the section 189 process. Misconduct per se cannot be said to constitute an economic rationale for a dismissal. The employer will have to persuade the Court that the misconduct has caused an economic rationale for dismissal in the sense that the company’s economic viability or economic stability is under threat to such an extent that dismissal on the basis of operational requirements is the measure of last resort. A Court is also, of course, not precluded from investigating the real reason for the dismissal. If the real reason is misconduct, the Court will not be inclined to allow the employer to dismiss via the section 189 route. See in this regard: In SA Mutual Life Assurance v IBSA²⁷ where the Court held that where the evidence showed that the employer was actually dissatisfied with performance of certain members of the department and chose not to initiate proper disciplinary inquiries but rather to restructure as a means of dismissing those employees with whom it was dissatisfied,²⁸ does not constitute an operational requirement under section 213. The employer also did not show that their jobs were redundant.²⁹

67. In Chauke & Others v Lee Service Centre CC t/a Leeson Motors³⁰ the Labour Appeal Court considered (although obiter) the possibility of dismissing employees where misconduct necessitating disciplinary action is proved, but management is unable to pinpoint the perpetrator or perpetrators. The Court identified two different kinds of justification for such a dismissal. The first is where one of only two workers is known to be planning major and irreversible destructive action, but management is unable to pinpoint which one it is. If all avenues of investigation have been exhausted, the employer may then be entitled to dismiss both. This involves the dismissal of an indisputably innocent worker. It posits a justification on operational grounds, namely that action is necessary to save the life of the enterprise. Clearly the dominant reason for the dismissal is the operational requirement of the dismissal. I have already indicated that I am of the view that section 189 is an option in such circumstances because the justification for the dismissal is the operational requirements of the business. The second category is distinguishable in that the justification advanced is not operational - it is misconduct. No innocent workers are involved. Management's rationale is that it has sufficient grounds for inferring that the whole group is responsible for or involved in the misconduct. Here a fair dismissal may be postulated. (i) The first is that a worker in the group which includes the perpetrators may be under a duty to assist management in bringing the guilty to book. Failure to assist an employer in bringing the guilty to book violates the duty of trust and confidence, essentials in the employer- employee relationship, and may itself justify dismissal. Though the dismissal is designed to target the perpetrators of the original misconduct, the justification is wide enough to encompass those innocent of it, but who through their silence make themselves guilty of a derivative violation of trust and confidence. (ii) The second line of justification is the inference of involvement, namely that the evidence justifies the inference that all the employees either participated in the misconduct or lent their support to it.³¹

68. It is important to point out that in the Chauke-matter the employer did not advance an operational rationale for the dismissal. Twenty employees were charged with misconduct – malicious damage to properly and concluded that they had all been guilty of it. The Court held that the facts and circumstances offered a strong evidentiary justification for this conclusion. In that case the sabotage was defined to a particular group of workers and they all worked together. Although the employer in that case concluded that there might have been some people that were not entirely involved, all of them must have been aware what was going on. The workers decided to keep silent. The Court concluded that in that case that the Court was justified in drawing a primary inference of culpable participation. The Court also referred to the decision in FAWU v ABI³² where the court held that on an application of evidentiary principles, the failure by any of the workers concerned to give evidence either in the workplace hearings or in the Industrial Court justified the inference that those present at the workplace on that day “either participated in the assault or lent it their support”. In the ABI matter the Court rejected the suggestion that the Appellants may have declined to come foreward because of intimidation or from a sense of “collegiality”. The Court therefore concluded in effect from the absence of “evidentiary self-absolution” that the Appellants were indeed present when the assault took place and either participated therein or lent their support. The Court in the ABI –case therefore decided the case on the basis of derivative misconduct.

69. Turning to the present matter. Were there operational requirements present in that case that permitted the employer to resort to the section 189 procedure? I have already referred to the fact that it was the Respondent’s case that the numerous incidents of violence it was not possible to proceed with the disciplinary hearings. The disappearance of Xhongo only compounded the problem. Although I have already pointed out that I am not persuaded that the Respondent was unable to proceed with the hearings, the question the Court now has to consider is whether or not the Respondent has succeeded in establishing that work-related violence posed a serious risk to the management, life and/or sustainability of its business. It is evident from the chronology of the incidents relied upon by the Respondent that the overwhelming bulk of these incidents occurred between 5 March 2007 and 30 March 2007. Lavery was also, in my view, not entirely clear in his evidence as to which of several “risks” underlay the retrenchments. He mainly testified that the Respondent was concerned about violence in the workplace if the alleged perpetrators were to return and also that it would have been untenable for victims of violence to be required to work alongside the alleged perpetrators (although there was no evidence of the reluctance of the victims in this regard). Lavery was also concerned that industrial relations disputes in the future would be marked by violence if no steps were taken against the perpetrators of violence in this strike. I am not persuaded that the conduct of the strikers threatened or affected the economic viability of the Respondent. In any event not to the extent that the Respondent had an economic rationale to implement section 189 procedures and therefore able to circumvent the disciplinary route. In the event I am of the view that the Respondent did not establish an economic rationale. The dismissal of the Applicants on the basis of operational requirements was therefore unfair.

Operational requirement route

70. If I am wrong in my conclusion namely that the Respondent has failed to establish that it had an operational requirement to proceed against the Applicants in terms of section 189 of the LRA, it then becomes necessary to consider whether or not the Respondent followed a proper consultation process; whether or not in the course of the consultation process the Respondent properly considered alternatives which would avoid or minimise the retrenchment and lastly whether or not the selection criteria applied were in accordance with section 189(7) of the LRA.

71. Section 189 of the LRA places a duty on the employer to consult when retrenchments are contemplated. Our Courts have given content to this duty in numerous judgments, most notably the matter in Atlantis Diesel Engines (Pty) Ltd v National Union of Metalworkers of South Africa:³³

“Consultation provides an opportunity, inter alia, to explain the reasons for the proposed retrenchment, to hear representations on possible ways and means of avoiding retrenchment (or softening its effect) and to discuss and consider alternative measures. … Consultation provides employees or their union(s) with a fair opportunity to make meaningful and effective proposals relating to the need for retrenchment or, if such need is accepted, the extent and implementation of the retrenchment process. It satisfies principle because it gives effect to the desire of employees who may be affected to be heard, and helps serve the underlying policy of the Act, to avoid or at least minimize industrial conflict. Where retrenchment looms employees face the daunting prospect of losing their employment through no fault of their own. Proper consultation minimises resentment and promotes greater harmony in the workplace. …. [T]he endeavour to avoid retrenchment, or minimise its consequences, should amount to a joint problemsolving exercise with the parties striving for consensus where possible. I agree that consultation, if circumstances permit, should be geared to achieve that purpose (bearing in mind that problemsolving is something distinct from bargaining and that the final decision, where consensus cannot be achieved, always remains that of management). Such a course would best serve the objectives of the Act and be conducive to industrial peace”.

72. Section 189(2)(b) of the LRA requires the parties to consult and attempt to reach consensus on the method for selecting the employees to be dismissed. Section 189(7) requires the employer to select the employees to be retrenched according to selection criteria that have been agreed to by the consulting parties; or if no criteria have been agreed, criteria that are fair and objective.

73. Six facilitation meetings were held during the consultation process. The Applicants do not complain that the consultations were inadequate, that they were not afforded an opportunity of tabling any suggestions at the consultation process, or that such suggestions as were put forward were not properly considered. Some of the alternatives raised by the union were: That the only fair course of action was for the Respondent to hold disciplinary hearings on an individual basis. I have already pointed out in detail why the Respondent was of the view that this course of action was not viable. It was also suggested that the parties consider instituting lawful and reasonable security measures at the Respondent’s premises and place the witnesses in so-called witness protection programmes. The Respondent testified that this was not practical and was also very expensive. It also did not provide any effective security to any individual employees after hours in their homes.

The selection criteria

74. During the consultation process the Respondent suggested as a further alternative that the persons identified as potential retrenchees submit themselves to a polygraph test.

75. The Court was not asked to provide a final and definite answer on the much debated issue of whether in a misconduct situation, an employee can be dismissed solely on the basis of an adverse polygraph test, where there is no other evidence, either direct or circumstantial, linking the employee to the transgression. I interpose here to point out that after having listened to the expert evidence, it appears that the experts were, in any event, ad idem that the polygraph cannot on its own be used as sufficient evidence to dismiss or to point to a person’s guilt. The case now under consideration involves a different question, namely whether, in the particular circumstances of the case, the proposal put forward by the Respondent was a reasonable suggestion aimed at minimising the number of dismissals, as envisaged by section 189(2)(a)(ii) of the LRA, or whether the raising of such a proposal was so unreasonable that it vitiated or fundamentally affected the fairness of the entire process. The question before this Court is not one of admissibility but of the weight which a Court should attach to polygraph test results.

76. The Respondent’s case was that the Applicants were selected for retrenchment “on the basis of affidavits linking them to acts of serious criminal conduct”. In this regard Lavery testified that he had discussions with Badenhorst, who had identified particular individuals who were going to be subjected to disciplinary hearings. During this process of selection the statements of all the persons who made them were reviewed in consultation with Elliott, Rodney Lambert and John May. The available video evidence were also considered in compiling the list. Lavery made the final decision in respect of who would be selected after he had consulted with various other people. Lavery also stated that he had accepted that the persons to whom the statements were made were considered to be credible. He, however, conceded that it was the credibility of the person who made the statement and not the person who took the statements down that was important. He did not, however, deem it necessary to send the witnesses of the respondent for a polygraph test.

77. It is common cause that in an undated letter between 5 and 26 September 2007 (at a late stage in the consultation process), the Respondent recorded that:

“Our client does not accept that enquiries are the only manner in which the impact of the retrenchment exercise may be mitigated. Bearing in mind that it is not possible to conduct enquiries, for the reasons already given, our client proposes that polygraph testing be utilized to establish whether anybody could be excluded from the proposed group of retrenches”.³⁴

It was thus proposed that the Applicants subject themselves to a polygraph test and the outcome would then determine whether or not they should remain on the list as potential retrenchees. It comes as no surprise that this selection proposal was rejected by the union. The Respondent refers to this rejection as follows in a letter dated 28 September 2007:

“[the union] rejected the company’s proposal that all potentially affected employees undergo a polygraph test and that those who pass, be removed from the list of potentially affected employees”.

78. Lavery was in favour of using the polygraph test because he was of the view that it was a “sound practical instrument” and that it was a viable option in taking people of the list. According to Lavery, if people had passed the polygraph, the Respondent would have gone back to the statements and review why they were then on the list. It is thus clear that even if the Applicants had passed the polygraph, that still would not necessarily have resulted in them being removed from the list. He stated as follows:

“It would be people that wouldn’t, I think the jargon is have a significant reaction to the question asked that if that sort of reaction had been obtained in the polygraph we would certainly have gone back to our statements and re-looked and where we could get hold of witnesses discuss those issues with the witnesses to see whether in fact we could take them off the list.”

……

“…..had those people agreed to do polygraphs and they had not had significant reaction to the polygraphs, we would then have had to have gone back to the statements to review why they were on the list”.

Polygraphs

79. Both parties led extensive evidence on the issue of polygraph testing as a valid, reliable and fair means of proving involvement in a crime. It was the Applicants’ case that the selective criteria used by the Respondent were not fair and objective. The Respondent on the other had pleaded that:

“the individual Applicants were selected on the basis of affidavits linking them to acts of serious criminal conduct”

and

“In any event, the individual Applicants were invited to undergo polygraph testing in order to show that the evidence against them was not true, and they had declined to do so. Had they not declined to do so, they would have been excluded from the group if they had passed the tests”.

80. In the pre-trial minute the legal issues arising from the papers are defined as follows:

“Whether the selection criteria employed by the Respondent was fair and objective. In particular, whether the employees were selected on the basis of their union affiliation and/or race. Further, and in the alternative, whether the Respondent acted fairly in selecting employees for retrenchment based on untested allegations of misconduct under the relevant circumstances and where they had allegedly been linked to serious misconduct by affidavits and had declined to undergo polygraph tests”.

81. What the Respondent therefore set out to do was to use the polygraph test, conducted by a trained and competent examiner who will be using the so-called control question technique, to determine who from the group of Applicants were involved in the violent incidents referred to above. Those who passed the test might then be excluded from the list of retrenchees. In essence, the method of selecting the retrenchees will be the results from the polygraph.

82. It was clear from the outset that the union was vehemently opposed to the polygraph test exercise. Not only was it argued that polygraphs were not reliable, it was also argued that a polygraph test cannot be held to be a fair and objective method for selecting employees to be retrenched. Furthermore, what was effective being required by the Respondent in the context of a retrenchment was that the employee (on the list of retrenches) be given an opportunity to reverse a decision already made by passing a polygraph test. In the context of misconduct (dismissal) proceedings the polygraph test will generally be used to determine a person’s guilty.

83. Because the union rejected the polygraph test as a selection criterion, it therefore became necessary for the Respondent to prove that the selection criteria were fair and reasonable. See CWIU & Others v Latex Surgical Products (Pty) Ltd³⁵ where the Labour Appeal Court held that where parties have not agreed upon selection criteria it is not permissible for an employer to use any selection criteria unless they were “fair and objective” as required by section 189(7) of the LRA. The use of selection criteria that are not fair and objective renders a dismissal substantively unfair.

84. This brings me to the next point. Is the polygraph a reasonable suggestion aimed at minimising the number of dismissals as envisaged by section 189(2)(a)(ii) of the LRA? In order to answer this question it is necessary to consider the value and reliability of polygraphs in general.

The approach of our courts in respect of polygraph testing

85. Two points must be made at the outset: Firstly, polygraph testing, although frequently used in the context of workplace discipline, is by no means uncontroversial. In fact, the extensive expert evidence led in this trail confirms this point overwhelmingly. Secondly, polygraph testing is usually used as a method of determining the guilt of an employee (in the context of misconduct investigations). In the present case the polygraph test would have been used as a selection criterion in determining who should be dismissed and to determine who is not guilty.

86. In SATAWU & Others v Protea Security Services³⁶ Protea Security Services sought to place guards with a new client who insisted that security guards used by it had to pass a polygraph test. When the guards could not be persuaded to undergo polygraph testing the employer embarked upon a section189 consultative process. The Court held that before it could be fair to force employees to subject themselves to testing, the test or assessment used would need to be: (i) shown to be scientifically valid and reliable; (ii) shown to be capable of being fairly applied to employees; (iii) shown to not be biased against any employee or group. The court held that the employer had the onus to show that a dismissal was fair and therefore had to show that the test used is valid and reliable and not biased against any employee or group of employees. In Truworths v CCMA³⁷ the Court expressed doubts as to the probative value of the polygraph on its own:

“[37] What appears from the aforegoing is that a polygraph test on its own cannot be used to determine the guilt of an employee (see also John Grogan Workplace Law 9^th edition page 160.) However, a polygraph certainly may be taken into account where other supporting evidence is available provided also that there is clear evidence on the qualifications of the polygraphist and provided that it is clear from the evidence that the test was done according to acceptable and recognizable standards. At the very least, the result of a properly conducted polygraph is evidence in corroboration of the employer’s evidence and may be taken into account as a factor in assessing the credibility of a witness and in assessing the probabilities. The mere fact that an employee, however, refuses to undergo a polygraph is not in itself sufficient to substantiate an employee’s guilt.”³⁸

87. The use of polygraphs has also received considerable attention in foreign jurisprudence. In United States v Scheffer³⁹ the Supreme Court held that a per se exclusion of polygraph evidence in court martial proceedings did not violate an accused’s constitutional right to present a defense. The Court held as follows in respect of the scientific status of polygraph testing:

“..there is simply no consensus that polygraph evidence is reliable. To this day the scientific community remains extremely polarized about the reliability of polygraph techniques”… Some studies have concluded that polygraph tests overall are accurate and reliable…Others have found that polygraph tests assess truthfulness significantly less accurately- that scientific field studies suggest the accuracy rate of the “control question technique” polygraph is “little better than could be obtained by the toss of a coin,” that is, 50 percent…This lack of scientific consensus is reflected in the disagreement among state and federal courts concerning both the admissibility and the reliability of polygraph evidence (4).”

“Whatever their approach (to the question whether polygraph evidence should be banned per se or admitted or excluded at the discretion of district courts) state and federal courts continue to express doubt about whether such evidence is reliable”;

“The approach taken by the President (of the Military Court) in adopting Rule 707 – excluding polygraph evidence in all military trials – is a rational and proportional means of advancing the legitimate interest in barring unreliable evidence..”

88. Pretorius et al, in Employment Equity Law⁴⁰ points out that because of the concerns in the United States regarding the scientific validity of polygraphs the United States Federal Government passed the Employee Polygraph Protection Act of 1994 making it unlawful for private employers to require any employee to take or submit to a polygraph test, to discharge, discipline or to discriminate against any employee who refuses to submit to such a test, or to take any action against an employee on the basis of the results of such a test. There are certain narrow exceptions and when testing is permissible there are strong safeguards in place. The employee has the right to counsel before each phase of the test and may secure in advance a copy of the questions to be asked. All tests must be administered by operators who possess minimum required qualifications.

89. The Supreme Court of Canada has found the results of polygraph tests to be inadmissible as evidence of credibility. (See R v Beland.⁴¹)

90. I am in agreement with the submission that while there may be a serious debate about the reliability of the outcome of the test (see the discussion hereinbelow) it has never been argued that the outcome of a polygraph test can serve as a substitute for a hearing. In fact, as is clear from the decisions, a polygraph on its own can never be conclusive proof of the guilt of an employee. At best the polygraph could be used as part of the investigation process to determine whether or not a further investigation into the conduct of a particular individual is warranted.

91. Extensive evidence was led by two highly competent and respected experts. I do not intend repeating the extensive and impressive qualifications and degrees of both experts. Both are clearly experts in the field of polygraph testing. The Respondent’s expert Dr Gordon Barland (“Barland”) is qualified in the field of forensic psychophysiology and is a consultant in psychophysiology in the United States. Both his master’s dissertation and his doctoral thesis were on the accuracy of the polygraph. Professor Tredoux (“Tredoux”) is a professor of Psychology at the University in Cape Town and holds a master’s degree and a PhD. I do not intend for purposes of this judgment to give a detailed discussion of the scientific theory underlying the use of polygraphs nor do I intend to refer to the evidence and comprehensive literature on this subject in great detail.

92. A polygraph is a devise for the measuring of the emotional or cognitive responses to various questions. Electrical signals are transduced from the body through conducting electrodes attached to the examinee and are then filtered and amplified through electronic circuitry so that an accurate measurement and recording of physiological activity may be made. In essence it measures respiration, electrodermal activity and cardio activity (blood pressure and heart rate). The polygraph test records momentary changes in a person’s level of physiological arousal. It is generally accepted that the physiological measurements are accurate in the sense that the polygraph apparatus will accurately measure physiological activity.

93. What should, however, be stressed is the fact that the polygraph test or devise does not measure deception or lying. The polygraph devise merely records physiological activity in the examinee. It is the results from this recordal that are used by the examiner to attempt to detect a deception. This is done by drawing certain inferences from the physiological activities which were recorded by the polygraph devise.

94. No controversy or dispute exists about the use or accuracy of the polygraph to measure certain physiological activities. The controversy surrounding polygraph testing lies in the theory and method underlying the manner in which inferences are drawn from the physiological recordings and, more in particular, the accuracy of the inferences drawn from the physiological activities. At the heart of the controversy is the fact that it cannot be stated unequivocally that there is a unique lie response in the sense that when a person lies one can see a unique pattern (referring to the physiological measurements or responses) that looks like nothing else and which occurs only when a person lies. In other words, it cannot unequivocally be stated that certain psychological responses or measurements only occur when a person lies and never under any other circumstance. The experts were ad idem that many things or factors can cause a body to respond similarly such as a noise, random thoughts and obviously when telling a lie. Barland’s evidence was that the polygraph test can be structure in such a way as to eliminate or control extraneous sources of reaction such as those referred to. Random thoughts may, for example be controlled by asking the same question several times.

95. Tredoux referred to the two main theories underlying the use of the polygraph test with extensive reference to research articles.⁴² The one version, referred to as the strong version, claims that there are certain specific and characteristic physiological response patterns that are associated with deceptive behaviour such as for example an increase in heart rate and a decrease in electrodermal activity. By merely observing that pattern, one can determine whether the examinee is attempting to deceive.⁴³ Tredoux, however, also refers to another study which claims that no research has demonstrated the existence of such a pattern and that there is no empirical evidence supporting this theoretical position.⁴⁴ The second theory, also referred to as the weak version, claims that the general rate or pattern of physiological activity increases or changes when a person is lying. There is, however, no specific pattern to this increase. In order to measure deception, the examiner will have to compare the rate of physiological activity at one point with the activity at another point of time. This will be done by asking a critical or relevant question and observing the response and then compare that activity with the activity that results when the examinee is asked a non-relevant question. It can therefore be said that the examiner’s diagnosis of deception is an inference based upon the elimination of alternative causes of a response. Tredoux referred to the German authors Fiedler, Schmid and Stahl⁴⁵ who argue that there is no necessary reason why physiological activation should increase when a person tells a lie. According to them this theory is assumed rather than demonstrated. This, according to the authors, is a fundamental and extraordinary failure of the polygraph.

96. As already pointed out, the examiner may ask various different questions during the interview. Because detecting deception relies on observing changes in physiological activity and since a question regarding a particular event might be more arousing than a neutral comparison question, it is clear that the comparison question needs to be formulated very carefully. The matter of finding the right kind of comparison question is often formulated as an issue of scientific control. In order to be able to draw an inference that the increased psychological arousal is due to an act of attempted deception, the examiner must ensure that the only possible difference between the responses to the two types of question is the presence of attempted deception:

    1. The first category of questions is the so-called relevant question (“RQ”) and will usually be directly relevant to the matter under investigation. A typical question would be: “Did you not commit the crime?” The truthfulness of this question is to be diagnosed by using the polygraph.

    2. The second category of questions would be the so-called irrelevant question which will usually be unconnected to the matter under investigation. For example, “Is today Wednesday”. This question is verifiably truthful.

    3. The third category of questions is a question which is only indirectly related to the investigation and aimed to elicit a lie or create a concern about whether or not the examinee is concealing information. This type of question is referred to as the comparison question (“CQ”). For example: “Before the age of 20, have you deliberately lied to somebody who believed you.” The CQ is designed to cause the innocent and guilty examinees to react to different categories of test questions in order to determine whether they are lying to the relevant question. The innocent person knows he is answering the relevant question correctly or truthfully. He is lying to or doubtful whether he can answer the comparison questions completely honestly with a simply “No” and would therefore react more to the comparison questions than to the relevant questions. The guilty person is lying to both the comparison question and the relevant question but is more concerned about his lie to the relevant question than he is to the comparison question because they are more direct and have greater consequences. The guilty subject will therefore react more to the relevant questions than to the comparison question.

97. Tredoux explained the different tests used in the detection of deception. He referred to the widely used test which is referred to as the relevant-irrelevant test (also referred to as the “RIT”). In terms of this test the greater physiological activation for a relevant question than for an irrelevant question may conclude that these questions have greater significance for the examinee or that the examinee is being deceptive. He referred to numerous scientific literature in which the RIT is criticised as being a weak form of control. He also referred to studies in terms of which the results bear out the contention that innocent subjects are likely to be falsely labeled as deceptive when the RIT is used.⁴⁶ Tredoux also referred to the so-called Control Question test (“CQT”) in terms of which the physiological activation made by examinees in response to critical questions is compared to activation made by the same examinee when telling an unrelated lie. According to this test it will be taken as evidence of deceptive behavior if the level of activation recorded by the polygraph is greater for the critical question than the control question. Where the level of activation is greater for the control question than the critical question, the examiner will take this as evidence against a conclusion of deceptive behavior. The examinee must, however, be induced to tell a lie, falsely believing that the examiner is not aware that a lie is being told. In other words, the examiner must deceive the examinee into believing that the polygraph accurately measures deception. Because the examinee must be intentionally misled for the test to be conducted, it comes at no surprise that this test has been criticized on ethical grounds.

98. Tredoux also referred to numerous other reasons why the polygraph test is intrinsically susceptible to producing erroneous result. He also referred to the fact that a guilty person or deceptive person is able to beat the polygraph by adopting countermeasures such as counting from 7 to 1 backwards to raise physiological activation levels when answering the control questions. By adopting such techniques the examiner may well be led into concluding that the examinee is being truthful.

99. Barland explained the procedures that must be adopted in conducting a polygraph. The examiner must first confer with the investigators and read any relevant reports and will then plan the test. The examiner must inform the examinee of his legal rights and must assess the examinee’s ability to take the test. The examiner must then review the examinee’s knowledge of the incident. The questions must then be prepared in a manner that would exclude or prevent any extraneous reactions that may influence the outcome of the test. The questions must then be asked in such a way to prevent the examinee to be surprised because this reaction may influence the accuracy of the test. Any cognitive activities to a particular question must be prevented. The examinee must therefore know in advance how he will answer each question. The examinee must understand the question to eliminate any ambiguities from the relevant questions. However, in the case of a control question, some ambiguity is desirable in order to force the examinee to reflect and to be concerned about the fact that he cannot answer the question completely truthful with a simple yes or no. This anxiety will generate a reaction which will help the examinee to clear him on the relevant question. The psychological data is scored on three charts. The results on the relevant and control questions are compared. The different channels (respiration, electrodermal and cardio activity) area each scored. Each score can range from -3 to 0 to +3. If the relevant question’s reaction is the larger it is scored as a minus. If the control question reaction is greater it is scored as a plus.

100. Both experts agreed that the basic theoretical assumption underlying the use of polygraphs to detect deception is that when an untruthful examinee fears detection that fear will produce a measurable physiological reaction. Accordingly, the polygraph measures the fear of deception rather than deception per se and the examiner will infer deception when the physiological response to questions about crime or unauthorized activity is greater than the response to other questions.

101. The greatest controversy in respect of polygraph testing lies in the different views about the accuracy of the test. The experts did, however agree that the crucial variable is the examiner and that the composition and formulation of the control questions is crucial. Barland was of the view that although the accuracy of polygraph testing remains controversial, it does not mean that it should not be used. What it does mean is that one must exercise caution in drawing conclusions from polygraph testing. It was further his evidence that although the scientific basis underlying the use of polygraph testing is weak; polygraph testing (in particular the use of the so-called comparison question test (“CQT”)) has demonstrated accuracy and accordingly is useful in certain contexts. In respect of the argument that the test’s accuracy is unknown and probably unknowable and therefore not accurate enough to use as evidence, Barland stated that although he agrees that the research does not justify a pinpoint accuracy figure, there is research to show that the test is likely to be accurate within certain ranges. Barland estimated that, if the inconclusive results are set aside the tests amongst examinees who by other evidence have been found guilty of the incident under investigation are roughly 90% accurate. Amongst examinees whom by other evidence have been found innocent of the transgression under investigation, the tests have been shown to be accurate in about 85%. Barland, therefore, pointed out that studies show that polygraphs are more accurate with guilty or deceptive subjects than with innocent or truthful subjects. In other words, it is easier to detect deception than it is to clear the truthful person. Barland is therefore of the view that the polygraph compares favourably with many other types of evidence which are routinely admitted as evidence such as eyewitness testimony, handwriting analysis and various medical and psychological diagnostic tests. Barland conceded that there is an inadequate theoretical basis for the polygraph as it has not been established that there is a causal relationship between a lie and a reaction. Even if there was such a relationship, there is as yet no adequate theory to explain that relationship. Barland agreed that the research on the theoretical basis for lie detection lags behind the application of the theory.

102. Tredoux’s view on the accuracy of the polygraph test was more critical. Apart from the fact that there is, in his view, no satisfactory scientific basis for the use of the polygraph, he was of the view that the test is not accurate and is dependent on a range of variables. With reference to a comprehensive 2003 review of polygraph testing of deception, the National Academy of Sciences concluded that most of the empirical research evaluating the accuracy of the polygraph was “…far short of what is desirable”. He referred to the research in respect of the RIT where the results have been overwhelmingly negative in the sense that the test is unable to correctly identify truthful responses at an accuracy level greater than 50%. In respect of the CQT test he concluded that studies provide diverging results regarding the accuracy of the CQT and that scientists are divided on how to interpret the results. He attributed many of these problems to the difficulties experienced in setting up acceptable field studies.

103. It appears therefore that researchers are divided on the accuracy of the CQT. The researchers are also divided on how to interpret the results and on what weight to attach to the different studies. One group of researchers Raskin, Honts, Amato and Kircher argue that the accuracy rates from suitable conducted studies are sufficiently high to justify the CQT as a form of polygraph testing whereas another group of researchers is of the view that these figures are inflated due to the unrealistic nature of laboratory experiments. Tredoux also pointed out that even scientists are devided about the results of laboratory-based studies. Scientists are, however in agreement that the so-called field or real-life studies of the polygraph would reveal more accurate statisctics. There are, however, as pointed out by Tredoux, numerous difficulties in doing field research, I do not intend dwelling on these problems. Suffice to point out that even David Raskin, Charles Honts and John Kircher, who are all, according to Tredoux, well known proponents of polygraph testing, place a rather low score on the accuracy of the polygraph. They conclude that the percentage of deceptive subjects which the test correctly classifies as deceptive ranges from approximately 73% to 100% with a weighted average percentage of approximately 86%. This is 36% more than could be expected by mere random guessing. The percentage of truthful subjects which the test correctly classifies as truthful ranges from approximately 30% - 83% with a weighted average percentage of approximately 50%. This is no more than could be expected by mere random guessing. The accuracy rates of these statistics may also be deflated by what the experts call an “outlier point”.

104. It was clear from the evidence that researchers in the field of polygraph testing are not in unanimous agreement about the scientific status of the polygraph particularly as far as the overall accuracy and usefulness of the test in detecting truthfulness or deception. On the one hand David Raskin, Charles Honts, Susan Amoto and John Kircher appears to be of the view that the accuracy rates from suitably conducted studies are sufficiently high to justify using the CQT as a method to detect deceptiveness. Other researchers such as David Lykken and William Iacono argue that the accuracy rates from suitably conducted studies show that the CQT has a higher probability of classifying truthful people as deceptive and should therefore not be used. The Court was also referred to a study of the British Psychological Society (2004) wherein it was concluded as follows:

“Our conclusion is that while polygraph-based techniques have some limited application in forensic investigations, they are unlikely to be acceptable in the British context of employment and staff screening. Even in the context of criminal investigation, there is controversy over the theoretical rationale behind lie-detection procedures and their accuracy and efficacy.”

105. Barland and Tredoux agreed that there are a range of factors which may at least theoretically affect physiological responses and that responses taken as indicating deception in a polygraph testing situation may actually have other causes. Barland also agreed that there is an error rate to the extent that there are a lot of variables that could affect the error rate. Tredoux was in particular critical of some of the key assumptions that underlie the methodology of the CQT. In fact, Barland also conceded that many proponents of the polygraph test (including himself) are of the view that although they don’t know why a polygraph test works, they are of the view that it does in practice accurately detect deception. He also conceded that it is very difficult to establish with any degree of precision how accurate the polygraph is. Barland pointed out that one of the crucial variables of the test is the quality and skill of the person conducting the examination (the examiner). In this regard he pointed out that it is important that the examiner has a degree of experience and knowledge in order to be able to develop good comparison questions. The examiner must also be sufficiently skilled to be able to structure the testing environment in such a way that can eliminate extraneous factors. The examiner must establish a rapport with the examinee to reduce the examinee’s overall level of anxiety and the examiner must choose vocabulary according to the background and education of the examinee. In order for the examiner to be able to be able to achieve the aforegoing, it is also required that the examiner have certain skills in interpersonal relations.

106. In respect of the South African context, Tredoux testified that one of the factors which may affect physiological responses in a polygraph test for deception is that of racial stigmatization. In this regard he referred to a scientific review in which it was pointed out that if either the examiner or the examinee is a member of a stigmatised group, the examinee may show heightened physiological responses particularly during difficult aspects of the testing situation. In this review the point was then raised that the responses received in these circumstances might increase the rate of false positive results among members of stigmatised groups when the CQT is used. Barland agreed that in a country like South Africa, with a history of systematic racial discrimination, racial stigmatisation would be widely prevalent and agreed that according to studies, racial stigmatisation may result in false positive results, although he indicated that he could not see why this would be the case.

107. The Applicant thus submitted that this is a weak and inadequate foundation for the respondent’s claim that polygraph testing has demonstrated a sufficient degree of accuracy in the detection of deception to allow one to conclude, based on the outcome of a test, that the person tested is guilty (or not guilty) of deception, and to then draw the further inference that because deception is indicated in response to a relevant question it follows from the existence of deception that the employee is guilty of misconduct .

108. In the context of this case where there were incidents of fire-bombing of houses and burning of vehicles, Barland was, however, of the view that the value of polygraph testing would be valuable to determine who were innocent of the incident:

“Well, I think that it would be very appropriate for the investigators to consider using the polygraph. I think it can be very valuable in eliminating the innocent and in focusing the investigation on those people who do not pass the polygraphs”

109. It would therefore appear that Barland himself accorded a limited role to polygraph testing in the present situation. In other words, the polygraph would be useful or valuable to eliminate the innocent and then to focus the investigation on those people who did not pass the polygraph.

110. It was also submitted on behalf of the Applicants that because the fundamental assumptions underlying the use of polygraph testing for this purpose have not been established, the polygraph test cannot be regarded as a fair and objective method of identifying wrongdoers in the employment context. It was further submitted that the following problems stand in the way of accepting the use of polygraph testing as a fair and objective means of identifying wrongdoers in the context of this case: (i) Firstly, “significant error rate, both with the guilty and the innocent”; (ii) Secondly the relatively higher error rate of polygraph testing in detecting truthfulness (rather than deception) and thus exonerating the “innocent”; (iii) Thirdly, the concerns (recognised by major reviewers) arising from the variations in methodology, sample selection and human variables in the studies relied upon by polygraph proponents to demonstrate its accuracy in detecting deception; (iv) Fourthly, the range of variables which may impact on the accuracy of the test, the most significant being the skill and experience of the examiner in formulating appropriate questions, conducting the test and scoring the results; (vi) Sixthly, the hypothesis (referred to in reviews) that heightened physiological responses of members of racially stigmatised groups during key aspects of the testing situation may increase the rate of false positive results among them when the CQT is used; (vii) Lastly, the absence especially in the South African context, of a regulatory framework which might provide a safeguard against examiner error.

111. I am in agreement that polygraph testing, as they presently stand, can do no more than show the existence of non-existence of deception. Even on this score, scientists are divided. Moreover, it is an accepted principle in our law that the mere fact that a person lie (in a criminal case) cannot in it self prove that the accursed is guilty of a crime. By no means can it be used as conclusive proof of guilt of a crime or misconduct. At best the polygraph test can prove that a person lied, not that he is necessarily guilty of a crime or misconduct. See in this regard R v Du Plessis:⁴⁷

“…. much of his evidence is hypothesis, wrung from him in cross-examination or given in answer to the court, at the time when he might have genuinely forgotten all about it. Under such circumstances the temptation, even to an innocent man, would be great to venture any explanation which might occur to him in the course of his evidence. And I would point to the danger, in a case such as this, of allowing what is at most a makeweight, such as the untruthfulness of the accused, to loom too large and to take the place of other essential evidence: Cf. Rex v Nel (1937), CPD at p 330).”

112. In light of the aforegoing and in light of the controversy that surrounds the accuracy and reliability of polygraph tests, I am not persuaded that the polygraph is a reasonable or fair alternative to minimise retrenchment. If a proven lie in a courtroom cannot by itself prove that the accused is guilty of a crime, then deception purportedly identified by a polygraph examiner cannot provide anything more than proof of a lie. In the context of a disciplinary process the polygraph can be a useful tool in the investigation process but can never substitute the need for a disciplinary hearing. A polygraph test on its own cannot be used to determine the guilt of an employee. In the context of an arbitration, the results of a polygraph may be taken into account where other supporting evidence is available provided also that there is clear evidence on the qualifications of the polygraphist and provided that it is clear from the evidence that the test was done according to acceptable and recognizable standards. I am, as already pointed out, not persuaded that it constitutes a fair and objective selection criteria or a fair an objective method alternative to mnimise retrenchment in the context of section 189 and section 189A of the LRA.

113. In the event the dismissal of the Applicants was substantively and procedurally unfair. The Applicant seeks the retrospective reinstatement of all of its members to the date of the dismissal. I do not intend to dwell on this aspect in much detail suffice to point out that enough evidence was placed before this Court by the Respondent to show that an employment relationship will never be able to exist between the Applicants and the Respondent. I therefore decide against reinstating the Applicants. I do, however, award each of them compensation equal to twelve months’ salary.

114. In respect of costs, I am of the view that the present case warrants a cost order in favour of the Applicants. In the event the Respondent is ordered to pay the costs including the costs of two counsel as well as the qualifying expenses of the expert witness Professor Tredoux.

Order:

1. The dismissal of the Applicants was substantively and procedurally unfair.

2. The Respondent must pay each of the Individual Applicants compensation equal to twelve month’s salary.

3. The Respondent to pay the costs, including the costs of two counsel as well as the qualifying expenses of the expert witness Professor Tredoux.

AC BASSON, J

Date of Judgment: 4 May 2010

Counsel on behalf of the Applicants:

C.S. Kahanovitz, SC and Adv. M.L. Norton.

Instructed by: Cheadle Thompson & Haysom

Counsel on behalf of the Respondent:

A C Oosthuizen SC.

Instructed by Deneys Reitz Inc.

---