---

IN THE LABOUR COURT OF SOUTH AFRICA, JOHANNESBURG

Not Reportable

Case No: JR1246/18

In the matter between:

OLD MUTUAL LIFE ASSURANCE SA (PTY) LTD	Applicant
and
NELSON JONGIKHAYA MAKANDA	First Respondent
THE COMMISSION FOR CONCILIATION, MEDIATION, AND ARBITRATION	Second Respondent
COMMISSIONER T HLATSHWAYO N.O.	Third Respondent
Date Heard: 9 July 2019 Delivered: 14 October  2019

 JUDGMENT

SIBANDA, AJ

Introduction

[1]             Old Mutual seeks to review and set aside an arbitration award, in which the Commissioner held that Mr Makanda’s dismissal on allegations of sexual harassment was substantively unfair. This case illustrates two principles.

[2]             The first, is the importance of a Commissioner’s duty to grapple with mutually destructive versions tendered by single witnesses. The correct approach to untangling this knot, is to assess both versions against three factors, namely: credibility, reliability, and the probabilities.

[3]             This approach is tried and tested. However, the importance of its application is particularly pronounced in cases involving sexual harassment.  This is so because sexual harassment often takes place in isolated circumstances, away from the glare of the third parties. In such cases, there is little more than (i) the complainant’s version; (ii) the alleged perpetrator’s version, and (iii) the objective facts.  It behoves a trier of fact to interrogate both versions to determine where the truth lies. The Commissioner did not do so.

[4]             The second principle, is that the Commission for Conciliation, Mediation and Arbitration (CCMA) should be a safe, gender-responsive dispute-resolution forum. This principle is implicit in existing domestic legislation. Apart from this, the International Labour Organisation recently recognised this principle in the Convention Concerning the Elimination of Workplace Violence and Harassment in the World of Work. [1]

[5]             This injunction places several duties on commissioners. Primarily, it requires commissioners (and litigating parties) to be alive to the fraught nature of sexual harassment, and to conduct arbitrations with the requisite degree of sensitivity. More importantly, it requires commissioners to shed patriarchal predispositions in their assessment of the evidence. Failure to do so, leads to outcomes that are simply out of step with society’s fight against gender inequality. Such outcomes cannot be reasonable.

[6]             Regrettably, the underlying arbitration, and resulting award, did not rise to these standards. The Commissioner failed to grapple with both versions presented to him. The award is also replete with factual findings that bear no rational connection to the evidence adduced. Worse still, the inferences drawn from these factual findings are plainly out of step with society’s fight against gender inequality.

[7]             In the circumstances, the award cannot stand.

Background

[8]             The arbitration took place over a number of days, rendering a large record comprising over 1400 pages. In what follows, I shall attempt to distil the material facts before the arbitrator. I begin by identifying the dramatis personae.

[9]             Mr Makanda, who is the first respondent, was employed by Old Mutual as a Provincial General Manager in Gauteng. This was a relatively a senior position. Mr Makanda reported to Mr Clarence Nethengwe (Mr Nethengwe).

[10]          I shall identify the complainant as “Ms S” to preserve her anonymity. During April 2016, Ms S joined Old Mutual as a Human Resources Coordinator. Ms S reported to Mr Joshua Mothlabane (Mr Mothlabane) who, in turn, reported to Mr Maxwell Khubeka (Mr Khubeka).

[11]          Both Ms S and Mr Makanda occupied offices in Old Mutual’s Parktown campus. It is common cause that, on occasion, Ms S’s role required her to assist Mr Makanda with human resources matters. This dispute concerns three incidents, in which Old Mutual contends that Mr Makanda made unwelcome sexual advances towards Ms S. These incidents were alleged to have taken place on 2 August 2016, 18 October 2016, and November 2016. Mr Makanda denied that these incidents took place as alleged or at all.

[12]          As such, each incident is the subject of wildly disparate versions. Given this divergence, I shall set out Ms S’s version and Mr Makanda’s separately.

The Ms S’s Version

[13]          This is Ms S’s story. Ms S testified that, in April 2016, Mr Nethengwe introduced her to Mr Makanda. Over the next few months, Ms S had several work-related encounters with Mr Makanda, which passed without incident.

[14]          However, the first incident took place on 2 August 2016. The office was not busy that day. As Ms S surmised, since 3 August 2016 had been declared a public holiday for Local Government Elections, many employees likely opted to take advantage of a long weekend. 

[15]          At some stage in the afternoon, while Ms S was alone in her office, Mr Makanda entered and struck up a conversation with her, during which he informed her that he was en route to a meeting in Bedfordview. As this conversation drew to a close, Mr Makanda extended his hand to her, for what she thought was a handshake. Ms S reciprocated and extended her hand. Suddenly, Mr Makanda pulled her towards him and hugged her tightly. Ms S described the hug as uncomfortable. She managed to slightly loosen herself from his grip. At that point Mr Makanda, whose cheek was pressed against Ms S’s, tried to kiss her. Ms S managed to push him away, after which, Mr Makanda left her office. 

[16]          Ms S testified that this incident shocked her. She initially sat in her office for a while, trying to unravel what had happened. Over the next few weeks, she mulled over the incident and tried to avoid Mr Makanda. 

[17]          It is common cause that Ms S did not immediately report the incident. Ms S gave two reasons for failing to do so. The first, was that she tried to deal with the immediate trauma of the incident. Once she had sufficiently recovered from her immediate trauma, Ms S remained reluctant to report the incident because she feared being disbelieved. In Ms S’s analysis, she had insufficient evidence to lodge a complaint.  

[18]          Ms S testified that, in the interim, Mr Makanda telephoned her several times to invite her to late lunches and dinners outside the office. Ms S diplomatically declined each of these invitations. 

[19]          On 28 September 2016, Ms S had a discussion with Mr Mothlabane. In this conversation, Ms S told Mr Mothlabane about the incident of 2 August 2016. Mr Mothlabane invited Ms S to lodge a formal complaint in terms of Old Mutual’s Sexual Harassment Policy. Ms S declined to do so, as she did not feel emotionally ready. 

[20]          In his evidence, Mr Motlhabane confirmed that this meeting took place and that Ms S reported the incident of 2 August 2018 to him. Mr Mothlabane testified that, in an attempt to manage the situation, he instructed Ms S to avoid working with Mr Makanda.

[21]          After his meeting with Ms S, Mr Motlhabane decided to informally escalate the issue to his line manager, Mr Khubeka. It is unclear from the record whether Ms S asked him to do so. Nevertheless, Mr Khubeka wanted to hear about the incident from Ms S personally and to ascertain what support he could offer.

[22]          On 18 October 2016, Ms S had a meeting with Mr Khubeka and Mr Motlhabane. In this meeting, Ms S recounted her version of the 2 August 2016 incident. What is clear, is that Mr Khubeka advised Ms S to confront Mr Makanda and that there was a broad discussion of how best to do so. 

[23]          There was, however, dissensus during evidence, about the exact advice given to Ms S at this meeting. It appears the trio sought a diplomatic way for Ms S to confront Mr Makanda. According to Ms S, Mr Khubeka suggested informing Mr Makanda that her managers reprimanded her over a rumour that someone had observed her being intimate with Mr Makanda in his office. This rumour was, of course, concocted. In her evidence, Ms S admitted this upfront. Mr Khubeka denied dispensing this advice to Ms S. However, he admitted that a rumour was likely part of the conversation

[24]          Very little turns on this. Whether on Mr Khubeka’s advice or not, the fact is that Ms S took this exact approach in her later conversation with Mr Makanda. 

[25]          At any rate, immediately after this meeting Ms S telephoned Mr Makanda and arranged to meet with him. It is common cause that this meeting took place at approximately 17h00 that evening. This meeting gave rise to the second allegation of sexual harassment. 

[26]          It is apposite to set out Ms S’s version verbatim. During her evidence in chief, Ms S’s testified as follows:

“MS S: I told Mr Makanda that I had just been to a meeting with my seniors wherein I was reprimanded about unbecoming behaviour from myself because there are allegations that had surfaced about an inappropriate relationship between him and I.

Mr PATEL: Was what you told him factually correct?

MS S: It was not.

MR PATEL: So why did you tell him that?

MS S: Because that is the advice that I got from my managers in the meeting earlier that day.

MR PATEL: Good.  Now let us go on.  When you told him what did he say?

MS S: He was shocked.  That was his initial reaction then he then (sic) immediately said he has no idea where that would come from considering that there is such limited contact between him and I and most of the time when we do have contact at all it is in a public setting.  So he is shocked on where those rumours would have come up or what would have informed such rumours and then he then (sic) immediately then just took over the conversation.  I did not get to say much after how I positioned my being there with him.  He then took over the conversation and went on to say a lot of things which also included him stating the incidents that I had mentioned from the 2^nd August as well as his own calls and he spoke to them in the sense that well I do know – I do remember on such a day I was in your office.

…

MS S: On such a day I was in your office and as I was about to leave your office I hugged you. Now I know my hugs might have been a bit tight and uncomfortable but the reason for that which is something that I never intended on telling you but because we are now having this conversation today I will let you know that the reason for that is because I have had a crush on you ever since I met you. So every now and again my feelings will cause me to do things that I would – I did not necessarily want to let out or let show and then he went on then and spoke about his phone calls…

MS S: For lunch and late dinners and he would also justify them by it is feelings that I have for you that I am struggling and finding it really, really hard to contain and then he went into lashing out about MAX [Mr Khubeka]…

So he was lashing out about MAX. MAX always being the instigator or in the middle of such rumours because it is not the first time where he finds himself in a situation where is being – where there is alleged relationships, inappropriate relationships between him and women and all that coming from MAX…

So he was just letting me know that MAX is always in the middle of such issues where people are being said to have relationships of some inappropriate manner with other people and then he also went on then to let me know that he has been having fantasies about me which were so intense that he wished he was away on business more often because then at least when he is away on business he is able to then help himself and act on his fantasies. I then asked him why he would entertain such fantasies and his response was that I have no business asking him or telling him who to or not to fantasise about.   So that is none of my business but what he can do commit to doing is not to act on his fantasies.

MR PATEL: Good.  And what happened then?

MS S: So as Mr Makanda is stating all things physically he has now changed so his face is flushed and he is (sic) whole demeanour has changed and he is fidgety so he is fidgeting in and out of the table.

…

MS S: And then as he is fidgeting it draws then my attention to the fact that he is now erect and this was now during the period where he is very vividly explaining the types of fantasies he has about me.

MR PATEL: And what does he say, what are those fantasies?

MS S: He is explaining how he holds me.

MR PATEL: Yes?

MS S: And also that there is activities – yes, for lack of a better word, activities of things that he like to do to me.

MR PATEL: Sexual activities?

MS S: Sexual Activities, yes.

….

MS S: They were sexual activities explaining how he would hold me and do to me.

MR PATEL: Explain that to us so that we have that on record if you can?

MS S: He described my physical structure and how he would have me on top of him.”[2]

[27]          Ms S testified that Mr Makanda repeatedly expressed his feelings for her. This led Ms S to suspect that Mr Makanda was subliminally suggesting they have an affair. Ms S testified as follows:

“MS S: Up until at some point I asked him but if you really actually did not intend on ever professing your feelings or letting me know of your feelings, why is it that every other sentence you keep on getting back to emphasising the feelings you have for me. At the time I could not remember the term ‘subliminal messaging’ and I said to Mr Makanda I do not know why it is that you are doing or if you are aware that you – I think that you are trying to plant this in my mind anyway but I could not remember the term ‘subliminal messaging’. I then said to him I will remember and once I remember what it is that I am trying to refer to which you are doing to me I will then let you know what it is. So that is what brought about this text message later on that evening.”[3]

[28]          Finally, Ms S testified that this meeting ended with Mr Makanda agreeing not to hug her, or call her for anything other than work-related reasons.[4]

[29]          Later that night, Ms S recalled the meaning of subliminal messaging and sent the following text message to Mr Makanda:

“A subliminal message is a signal or message designed to pass below (sub) the normal limits of perception. For example, it might be inaudible to the conscious mind (but audible to the unconscious or deeper mind) or might be an image transmitted briefly and unperceived consciously and yet perceived unconsciously.”

[30]          A short while later, Mr Makanda replied with the following text message:

“Now it makes sense, it’s truly a psychological game. Not sure I was consciously playing it or it occurred unconsciously. Thanks I learnt something today.”

[31]          On Ms S’s version, the next few weeks passed without incident. 

[32]          During November 2016, Ms testified that she encountered Mr Makanda in the basement parking lot. Mr Makanda said he was struggling to comply with the agreement they reached on 18 October 2016.[5] According to Ms S, at this stage, she realised that the harassment would not stop. 

[33]          However, Ms S did not lodge a formal complaint about the alleged harassment until February 2017. In the intervening period, Ms S procured a transfer to Old Mutual’s Sandton office as an HR Consultant. This transfer took effect from 15 January 2017.

[34]          During the remainder of January 2017, Ms S was in the process of handing over her duties to the person who replaced her. After settling into her new environment, Ms S lodged a formal complaint.[6] 

[35]          I now turn to Mr Makanda’s version.

Mr Makanda’s version

[36]          Mr Makanda testified that, during April 2016, Mr Nethengwe introduced him to Ms S. At that stage, Ms S had just joined Old Mutual, and would be part of the HR team supporting him in his function.

[37]          During the introduction, Mr Nethengwe made a remark about Ms S’s beauty and said “please do not touch her.” Although Mr Makanda considered this remark inappropriate, he did not pay heed to it.

[38]          According to Mr Makanda, that same month, he had a meeting with Ms S to discuss performance contracts. Mr Makanda testified that, during this meeting, the following took place:

“MR MAKANDA:  Soon after she got into my office, what came as a shock to me, as she was sitting – we were looking at the computer sitting next to me, instead of focussing on the computer she started rubbing my left hand with the palm of her right hand, and stopped talking about the contracts, started talking about her love life.

MR MAJA: What did she say when she was talking about her love life?

MR MAKANDA: She was talking about why she is cheating for her husband. She is cheating because she wants to sleep with another man, she wants to have fun.

MR MAJA: What does that mean?

MR MAKANDA: It made no sense to me. I did not know where the talking came from and I was still a bit shocked though I though it is an innocent touch, but I was shocked but I was assisted by my regional – one of my regional managers walked through, Zoleka Cima, and as soon as Zoleka Cima walked through the door she removed her hand changed the topic back to the performance contract.”[7]

[39]          Although this arises from the disciplinary hearing, Mr Makanda repeated this evidence in the arbitration.[8] Mr Makanda testified that he informed his personal assistant about the incident.

[40]          This led Mr Makanda to suspect that Ms S had been placed in his office to frame him for sexual harassment. At any rate, in the subsequent months, Mr Makanda continued to work with Ms S without incident.  

[41]          Mr Makanda denied that the 2 August 2016 incident took place at all. He maintained that, although he was in the office that day, he did not encounter Ms S, much less force himself upon her. He testified that it would not have made sense for him to do so. Being a weekday, there were many people in the office. To reach Ms S’s office, he would have had to pass the tea room, in which employees often converged. He could not have passed by unnoticed. Given his seniority, employees would have interacted with him. What is more, it was implausible that the incident could pass unnoticed because Ms S’s office was visible from the corridor. Finally, Mr Makanda tendered his travel logs to demonstrate that he did not travel to Bedfordview that day, as alleged by Ms S.

[42]          Mr Makanda also denied inviting Ms S to late dinners. He tendered his telephone records to demonstrate that he did not telephone Ms S at all.

[43]          According to Mr Makanda, his relationship with Ms S was collegial and professional. She continued to assist him with various tasks even after 2 August 2016.

[44]          Mr Makanda testified that, on 18 October 2016, his secretary informed him that Ms S sought an urgent meeting with him. He accepted this meeting and Ms S arrived late in the afternoon. Mr Makanda hotly disputed Ms S’s version of this meeting. During his evidence in chief, Mr Makanda testified as follows:

“MR MAKANDA: … The first thing [Ms S] said to me is that Mr Makanda, I am not here about a work-related issue.  I am here about us. Then I asked who is us. Me and you. Then I started to be short. I do not know us between me and you. What is it all about. I am coming from a meeting with Maxwell Khubeka and Joshua Motlhabane, my bosses. Maxwell reprimanded me because he heard that me and you were caught sleeping in your office and you threatened the person who caught us sleeping, that should she tell anyone you will fire her. I was shocked. Who said that? It is Maxwell. The first this (sic) [Ms S] – my first question response, how can I be caught sleeping with you in my office when I am not even flirting with you. That is my first question. My second question, can I call Maxwell after that. [Ms S] said no, I came to you in confidence, please do not call Maxwell. Then again I went back. Then I am shocked, how can I be caught sleeping with you when I am not even flirting with you. Then I asked [Ms S], why are you here [Ms S]. Is it because you are not sure whether we slept in my office or we did not. [Ms S] did not answer because I was also – I do not remember her answer there, I was shocked but what I remember is what she said [Ms S], Mr Makanda, this is the third time now you are mentioning that you are not flirting with me and yet I have got a subliminal message playing in my head.

MR COMMISSIONER: Come again?

MR MAKANDA: I have got a subliminal message playing in my head. I said what is a subliminal message [Ms S], I do not know that word. It is a bit new, that word was new to me. No Mr Makanda, it is a psychological game sort of. But [Ms S] I am not playing psychological games here. Okay.

COUNSEL: Just stop there…

MR MAKANDA: Then she said I will explain what it means when I am at home.”[9]

[45]          Mr Makanda admitted that he was fidgeting. He attributed this to the unsettling contents of their discussion. He said:

“MR MAKANDA: All I know is that on the day I was shocked and a bit angry, controlling my anger because the reason of that I did not want to show my upset, my anger because [Ms S] has been in my office almost – that was the fourth time on that day in a space of one and a half months because in the month of September she has been in my office three times, assisting me with the acquiring of my PA.”[10]

[46]          Mr Makanda denied that he was aroused and, even if he had been, Ms S was not in a position to observe this. On Mr Makanda’s version, the meeting ended as follows:

“MR MAKANDA: What I told [Ms S] was that as from today she must stop coming to my office. If I need any human resource assistance, I will ask Joshua, her boss because I dot (sic) not like these nuances – I will not be associated with this kind of nuances in future. So as from today she must stop coming to my office.  It is incorrect for her to mention things like stopping hugging and what is this, shaking hands, fidgeting, because what she came and spoke about – about a rumour. She has been reprimanded by Maxwell about us having been caught sleeping in my office. So those nuances I do not like them.  I said from now onwards, please let us agree if I need any human resource help, I will ask Joshua because like I said before, that was the 4^th time she has been to my office in a space of one and a half month, because she was assisting e with the appointment of a BE move. She came with shortlisted CVs, employment contract and amended employment contract third time. Fourth time with this outrageous rumour of us being caught sleeping. Then at my position, I did not need that. That was the only agreement, to limit our contact or she must keep the distance. That was the agreement because there was never a mention of 2^nd August at any stage like she said. There was never mention of I must stop my advances. Never ever. No mention of those things…”[11]

[47]          Mr Makanda did not report this conversation to anybody else. 

[48]          Mr Makanda denied that he met Ms S in the basement during November 2016. He testified that he had no reason to be in the basement parking lot because he parked at surface level in front of the building.

[49]          Mr Makanda suspected that Ms S was placed in his department to trap him. He based this conclusion on the April 2016 incident, and the timing of Ms S’s formal complaint (which coincided with a competitive process for a promotion he wanted). Further, in the build-up to his disciplinary hearing, Mr Makanda received a text message from another employee with whom Ms S was friendly. In this exchange, Ms S warned that employee of the upcoming hearing and the possibility of Mr Makanda alleging that she was conducting an affair with him. When that employee asked whether Mr Makanda had touched Ms S, she said he had. However, this was after he asked several times.

The disciplinary and arbitration proceedings

[50]          As I already mentioned above, in February 2017, Ms S filed a formal complaint of sexual harassment against Mr Makanda. Pursuant to this complaint, Old Mutual charged Mr Makanda with sexual harassment, arising from the incidents of 2 August 2016, 18 October 2016, November 2016, and the telephone calls.

[51]          After a lengthy hearing, chaired by a member of the Bar, Mr Makanda was dismissed for sexual harassment. In turn, Mr Makanda engaged the dispute resolution mechanisms in the Labour Relations Act (the LRA)[12] to challenge the procedural and substantive fairness of his dismissal. 

[52]          As to substantive fairness, Mr Makanda contended that (i) the incident of 2 August 2016 did not happen, (ii) the incident of 8 October 2016 did not amount to sexual harassment, and (iii) he made no telephone calls to Ms S during that period.

[53]          In his procedural challenge, Mr Makanda contended that Old Mutual was time-barred from proceeding with the disciplinary enquiry. This was so because Old Mutual’s Sexual Harassment Policy required Ms S to report the incident within two months, as required.

[54]          The Commissioner arbitrated the dispute and found Mr Makanda’s dismissal to be substantively unfair. The Commissioner did not express a finding on procedural fairness, despite being called upon to do so.

The award

[55]          The resulting arbitration award is difficult to decipher. During argument, counsel for both parties agreed that the award is based on five conclusions. In what follows, I shall crystalise these conclusions.

[56]          The first conclusion is this: The Commissioner rejected Ms S’s version of the 2 August 2016 incident. He found it improbable that the incident took place because Ms S failed to report the incident timeously. The Commissioner drew an adverse inference against Ms S, from her failure to disclose the text message exchange of 8 October 2016 during evidence in chief. He inferred that Ms S was trying to downplay her role in the matter. The Commissioner expressed this line of reasoning as follows:

“Ms S’s evidence has to be evaluated in its totality. She alleged that the applicant took advantage of her when he extended his hand as if for a hand shake. The applicant attempted to forcefully hug and kiss her in the office on 2 August 2016.  Ms S broke off the grip and the applicant ran out of the office.

Ms S was the HR consultant in the area of the applicant and Nethengwe had told the applicant to keep his hands off her.  This alone should have assured Ms S that she was protected.  Since Nethengwe said she had a direct line to him, she should have raised the issue timeously.

The respondent’s disciplinary code was specific about the timeline within which to report the offence. The applicant as a custodian of the HR policies was familiar with the time constraints. There was, however, no provision for condonation or pardon though in consideration of the serious allegation I have to deal with the matter as it was presented.

Ms S did not disclose that she wrote the whatsapp message relating to subliminal feeling. I am, therefore, not persuaded that she was numbed to the extent that she could not have raised the alarm.  Ms S claimed to have been afraid that she did not have enough material evidence to implicate the applicant at the time.  Ms S claimed to have raised the issue of the whatsapp in the hearing though it would have been imperative to raise the issue before me.  I am wary that this emerged only through cross-examination. I therefore draw a negative inference against her that she wanted to underplay her part in the process.

I find that Ms S was not so afraid that she could not raise the alarm on or immediately after 2 August 2016.”[13] (my emphasis)

[57]          Having done so, the Commissioner moved to the events of 18 October 2016. The Commissioner appears to have accepted Ms S’s version of the events of 18 October 2016. From his reasoning, one can only deduce that the Commissioner did not construe Mr Makanda’s alleged conduct as sexual harassment. This reasoning forms the second and third conclusions upon which the award is based.

[58]          The Commissioner’s second conclusion, was that Ms S’s behaviour on 18 October 2016 gave Mr Makanda the wrong impression. This was because (i) Ms S did not directly confront Mr Makanda about his previous conduct; (ii) she chose to conduct the meeting in the evening; and (iii) she commenced the discussion with a false rumour. Again, it is apposite to repeat the Commissioner’s reasoning for illustrative purposes.

“The second incident took place on 18 October 2016 when she was advised to confront the applicant.  Ms S took a route which gave the applicant a wrong impression in that she spoke about the rumour making rounds regarding the fact that they were caught sleeping together.

In my considered view the route taken suggested that Ms S had something to hide.  I would have expected Ms S to have been brutally honest with the applicant and said: I am not at ease with your antics or stop touching me inappropriately and I don’t relish it.

I have regard to her position as HR consultant and she could have shown others that she was worthy of her position.”[14]

[59]          The Commissioner’s third conclusion is that Ms S’s approach to the incident of 18 October 2016 was inconsistent with a victim of sexual harassment. The Commissioner expressed this as follows:

“Ms S claimed to have noted that the applicant was flushed, erect and fidgety when he told her to stop seeing one another.

The manner in which she addressed the matter was, in my view, not consistent with a person who was sexually harassed.

Ms S did not state to me what gave rise to the applicant assuming this state of being erect, fidgety and flushed. It was only under cross-examination that it emerged that the conversation was characterised by sexual undertones.  Ms S was free to deal with issues which were characterised by sexual undertones, considering that she was afraid.

I have regard to the time at which she approached the applicant at about 18h00 late after work.  I am not persuaded that Ms S felt threatened in the presence of the applicant.”[15]

[60]          Fourth, the Commissioner made a credibility finding against Ms S, pertaining specifically to the incident of November 2016. The Commissioner preferred Mr Makanda’s version because he remained consistent under cross-examination, while Ms S prevaricated and was ruffled. This, in the Commissioner’s view, tilted the probabilities in Mr Makanda’s favour. The Commissioner expressed this finding as follows:

“Ms S claimed the third incident took place around the parking bay in the basement. The applicant claimed that his parking bay was on the surface not underground.

I am persuaded to prefer the version of the applicant as he was consistent and did not appear ruffled under cross-examination.  Ms S appeared to prevaricate in the course of cross-examination. This, in my considered view, tilts the scale of probability in favour of the applicant.”

[61]          Fifth, the Commissioner found that the telephone calls and invitations to lunch could not amount to sexual harassment in the context. He accepted that Mr Makanda’s telephone records contained several calls to Ms S, but he could not “second-guess the contents of these calls”. The Commissioner expressed this reasoning as follows:

“The issue of invitations to lunch could not, in my view, amount to sexual harassment when I have regard to the working environment in the assurance service.

I have regard to the extent the applicant had gone to refute the allegation against him in that he retrieved the itemised billing for his landline and cell phone for the relevant period.

There were calls made to Ms S by the applicant, however I don’t know the content thereof. I can’t second guess the content of the calls thus the presence of Ms S’s calls could not be construed to amount to sexual harassment.”[16]

[62]          This is the sum total of the Commissioner’s reasoning. In the premises, the Commissioner found that Mr Makanda was not guilty of sexual harassment and awarded him retrospective reinstatement.

The test on review

[63]          Broadly, Old Mutual mounts two main attacks against the award. It contends that (i) the Commissioner committed a gross irregularity; and (ii) the award is so disconnected from the evidence, that no reasonable decision-maker could have reached that conclusion.

[64]          The starting point is that no appeal lies against an arbitrator’s decision. This Court is constrained to determine the application on the grounds set out in section 145(2)(a) of the LRA (as suffused by reasonableness); or on the basis that the decision is one that a reasonable decision-maker could not reach.

[65]          The threshold for interference on review is high. Awards should not be overturned lightly on review. Minor imperfections cannot, on their own, vitiate an award. More substantial irregularities, such as errors of fact or law, can only unseat an award if they are material.

[66]          An irregularity becomes material if it leads the arbitrator to misconceive the inquiry or renders the result unreasonable. In Herholdt v Nedbank Limited (Congress of SA Trade Unions as Amicus Curiae),[17] the Supreme Court of Appeal (SCA) captured this as follows:

“In summary, the position regarding the review of CCMA awards is this: A review of a CCMA award is permissible if the defect in the proceedings falls within one of the grounds in section 145(2)(a) and the LRA. For a defect in the conduct of proceedings to amount to a gross irregularity as contemplated by section 145(2)(a)(ii), the arbitrator must have misconceived the nature of the inquiry or arrived at an unreasonable result. A result will only be unreasonable if it is one that a reasonable arbitrator could not reach on all the material that was before the arbitrator. Material errors of fact, as well as the weight and relevance to be attached to particular facts, are not in and of themselves sufficient for an award to be set aside, but are only of any consequence if their effect is to render the outcome unreasonable.”[18]

[67]          In Head of the Department of Education v Mofokeng and Others ,[19] the Labour Appeal Court (LAC) held that materiality is determined by evaluating the extent to which the error warped (or distorted) the result. If, without the error, a different result would have been reached, then the error is material. The LAC captured this as follows:

“Irregularities or errors in relation to the facts or issues, therefore, may or may not produce an unreasonable outcome or provide a compelling indication that the arbitrator misconceived the inquiry. In the final analysis, it will depend on the materiality of the error or irregularity and its relation to the result. Whether the irregularity or error is material must be assessed and determined with reference to the distorting effect it may or may not have had upon the arbitrator’s conception of the inquiry, the limitation of the issues to be determined and the ultimate outcome. If but for an error or irregularity a different outcome would have resulted, it will ex hypothesi be material to the determination of the dispute. A material error of this order would point to at least a prima facie unreasonable result. The reviewing judge must then have regard to the general nature of the decision in issue; the range of relevant factors informing the decision; the nature of the competing interests impacted upon by the decision; and then ask whether a reasonable equilibrium has been struck in accordance with the objects of the LRA. Provided the right question was asked and answered by the arbitrator, a wrong answer will not necessarily be unreasonable. By the same token, an irregularity or error material to the determination of the dispute may constitute a misconception of the nature of the inquiry so as to lead to no fair trial of the issues, with the result that the award may be set aside on that ground alone. The arbitrator however must be shown to have diverted from the correct path in the conduct of the arbitration and as a result failed to address the question raised for determination.”[20]

[68]          And later, the LAC further went on to state that:

“Lapses in lawfulness, latent or patent irregularities and instances of dialectical unreasonableness should be of such an order (singularly or cumulatively) as to result in a misconceived inquiry or a decision which no reasonable decision-maker could reach on all the material that was before him or her.”[21]

[69]          This formulation requires the court to first determine whether the commissioner committed an irregularity. If so, then the next stage of the inquiry involves two independent prongs. 

[70]          The first, involves determining whether the irregularity resulted in the arbitrator misconceiving the nature of the inquiry or his duties in connection therewith.[22] If it did, then the award must fall.       

[71]          The second, involves determining the extent to which the irregularity distorted the result, rendering the outcome unreasonable. This begs the question: when is a decision so unreasonable as to warrant setting aside on review? In Coega Development Corporation(Pty) Ltd v Commission for Conciliation, Mediation, and Arbitration and Others,[23] this Court grappled with that question. After surveying numerous authorities, the Learned Myburgh AJ held:

“[70] There is a another issue that warrants some consideration for present purposes — what is the threshold for unreasonableness?  Traditionally, the answer is that the decision must fall outside the range of reasonable decisions. But this, in itself, is not particularly helpful, because how does one determine that range?  To my mind, the issue turns on the intensity with which a review for reasonableness should be undertaken in the context of this court having been tasked (through its review powers) to supervise the reasonableness of CCMA awards — the higher the intensity of the review, the narrower the range of reasonable decisions (and vice versa).

[71] In my view, on an overall assessment of the jurisprudence of the LAC (whose judgments are, of course, binding on this court and from which this court takes guidance), it adopts a relatively high intensity reasonableness review. As a result of this, on my assessment, where an award is obviously wrong, the LAC will typically set it aside on review on the grounds of unreasonableness — it does not have to be hopelessly wrong or absurd before it will do so (which is what the threshold in a lower intensity review might be).  Seen thus, the permissible margin for errors by a CCMA commissioner is between what is objectively right and what is obviously wrong.  Put differently, where a decision is obviously wrong, it falls outside of a range of reasonableness.”

[72]          This formulation does not blur the distinction between review and appeal. It sets the outer limits for interference between what is objectively right, and obviously wrong. A review court might well disagree with a finding that falls between these outer limits. That, on its own, is not sufficient to vitiate an award on review.[24] To do so, would be to overstep this Court’s powers.

[73]          Generally, if the reasons for the award rationally support the outcome, interference on the basis of unreasonableness will be unjustified.[25]

[74]          So much for the threshold by which this matter should be determined. In sum, this review must succeed if (i) the Commissioner committed a reviewable irregularity that rendered the decision unreasonable, and (ii) the outcome is not otherwise justifiable on the evidence before the Commissioner.

[75]          To my mind, the Commissioner committed several irregularities. Principally, the Commissioner failed to evaluate the evidence before him. Moreover, in the evaluation he did perform, the Commissioner clearly misconceived the inquiry and committed numerous factual errors. I now interrogate these in turn. 

The commissioner’s failure to evaluate the evidence

[76]          The Commissioner’s primary duty was to resolve the factual controversy between the parties.

[77]          The Commissioner was faced with mutually destructive mutually destructive versions tendered by Ms S and Mr Makanda. If Ms S’s version was true, then Mr Makanda was guilty of sexual harassment. There was no scope to find otherwise. Conversely, if Mr Makanda’s version was true, then he was not guilty of sexual harassment. 

[78]          There was no middle-ground between their versions. Nor was there direct evidence, which is often difficult to come by in sexual harassment cases. The Commissioner was bound to accept one version or the other in respect of each incident.

[79]          The technique to be employed in this process was laid out by the SCA in Stellenbosch Farmers Winery Group Limited and Another v Martell et Cie and Others .^[26] 

“…The technique generally employed by courts in resolving factual disputes of this nature may conveniently be summarised as follows. To come to a conclusion on the disputed issues a court must make findings on (a) the credibility of the various factual witnesses; (b) their reliability; and (c) the probabilities. As to (a), the court's finding on the credibility of a particular witness will depend on its impression about the veracity of the witness. That in turn will depend on a variety of subsidiary factors, not necessarily in order of importance, such as (i) the witness' candour and demeanour in the witness-box, (ii) his bias, latent and blatant, (iii) internal contradictions in his evidence, (iv) external contradictions with what was pleaded or put on his behalf, or with established fact or with his own extracurial statements or actions, (v) the probability or improbability of particular aspects of his version, (vi) the calibre and cogency of his performance compared to that of other witnesses testifying about the same incident or events. As to (b), a witness' reliability will depend, apart from the factors mentioned under (a) (ii), (iv) and (v) above, on (i) the opportunities he had to experience or observe the event in question and (ii) the quality, integrity and independence of his recall thereof. As to (c), this necessitates an analysis and evaluation of the probability or improbability of each party's version on each of the disputed issues. In the light of its assessment of (a), (b) and (c) the court will then, as a final step, determine whether the party burdened with the onus of proof has succeeded in discharging it. The hard case, which will doubtless be the rare one, occurs when a court's credibility findings compel it in one direction and its evaluation of the general probabilities in another. The more convincing the former, the less convincing will be the latter. But when all factors are equipoised probabilities prevail.”

[80]          In Sasol Mining(Pty) Ltd v Ngqeleni NO and Others ,[27] this Court endorsed the above dictum when it held:

“[9] One of the commissioner's prime functions was to ascertain the truth as to the conflicting versions before him. As I have noted, this much the commissioner appears to have appreciated. What he manifestly lacked was any sense of how to accomplish this task, or which tools were at his disposal to do so. The commissioner was obliged at least to make some attempt to assess the credibility of each of the witnesses and to make some observation on their demeanour. He ought also to have considered the prospects of any partiality, prejudice or self-interest on their part, and determined the credit to be given to the testimony of each witness by reason of its inherent probability or improbability. He ought then to have considered the probability or improbability of each party's version. The commissioner manifestly failed to resolve the factual dispute before him on this basis. Instead,  he summarily rejected the evidence of each of the applicant's witnesses on grounds that defy comprehension…”[28]

[81]          When faced with mutually destructive versions, an arbitrator must be equal to the task and resolve them. If he does so, the scope for interference on review is limited, and was summarised in Simani v Mosselbay Municipality and Others,[29]  as follows:

“That having been said, if a commissioner comes to a conclusion and the reasons provided by him or her for such conclusion objectively speaking do not constitute any support for the conclusion, the reviewing court may be entitled to intervene. Whether the court will be entitled to intervene is dependent upon the nature and scope of the commissioner's erroneous reasoning in relation to the final conclusion. If, for instance, none of the reasons provided by the commissioner in fact support the ultimate conclusion, it appears obvious that he or she cannot be said to have determined the substantive merits of the dispute and the award will be set aside.”

[82]          By parity of reasoning, arbitrator commits a gross irregularity (possibly of a worse order) when he simply elects a version without engaging with the factual dispute at all. In such cases, the award falls far short of the arbitrator’s duty to resolve the dispute on the evidence. The result is more akin to a vote than a quasi-judicial decision.   

[83]          A further point bears mentioning. The approach in SFW is inherently comparative. It requires the Commissioner to assess the (i) credibility of both witnesses, (ii) their reliability, and (iii) the probabilities. Its purpose is to discern which of the mutually destructive versions is more probable than the other. This being so, it cannot be conducted solely by reference to one of the versions, while completely ignoring the competing version. If only one version is put to the test, then only half of the task is performed.

[84]          The duty to resolve factual disputes by reference to both versions takes on an enhanced symbolic significance in disputes concerning sexual harassment. It is imperative that the process does not only interrogate the complainant and her version. Equal attention must be paid to the alleged perpetrator’s version.

[85]          The contrary position, in effect, puts the complainant on trial. A careful consideration of both versions ensures, at the very least, that the process does not put only the complainant on trial. It is difficult enough for victims of sexual harassment to come forward, and subject themselves to a process in which they inevitably re-live their trauma. This trauma can only be worsened by evidentiary evaluations that place complainants, and not alleged perpetrators, on trial.

[86]          At any rate, the Commissioner does not appear to have evaluated the competing evidence and conducted the test in SFW. In his “analysis of arguments and evidence,” there is no reference whatsoever to the credibility, reliability, and probabilities of Mr Makanda’s version in respect of each event. This alone is a material irregularity that vitiates the award.

Misconstruing the issues

[87]          The award contains several errors of fact and law, which render it deficient, both at the level of his understanding of the issues and the reasonableness of the award.  

[88]          I shall consider these errors in respect of each of the award’s five foundations.

[89]          I am minded to commence this analysis by repeating this Court’s dictum in Rustenburg Platinum Mines:[30]

“[49] The lessons coming out of the global anti-sexual harassment movements mentioned elsewhere in this judgment are that the so-called ‘victims’ of sexual harassment react to their own ordeals and circumstances differently, and in most instances, long after the fact. Astute lawyers will always attack the credibility of complainants because of the time lapses between the incidents and when they get reported, and the inability to proffer specifics or corroborating evidence. There is of course always a danger in accepting at face value that an incident took place simply because it was reported immediately thereafter. The consequences could be dire for both the accuser and accused if the allegations are found to be without merit. The stigma of being a sex pest remains forever even if, in the end, the allegations are found to be unsubstantiated. There is however an even greater danger when it is not accepted that the incident took place because the complainant took long to report it, or that he or she cannot recall details with I clarity. Without vindication because of such technicalities, the trauma persists indefinitely for the complainant.

[50] Common sense, however, and a bit of appreciation of the human mind dictate that one must look deeper and objectively into the reasons why incidents of sexual harassment are not immediately reported. This examination again has to do with how human beings react differently to the same or a similar set of circumstances. Depending on the nature and character of the individual complainant, in some instances, and immediately when an incident takes place, the harasser may be told in unequivocal and impolite terms to cease and desist the conduct, and to find the nearest cold shower. At best, the incident may even be reported immediately. Of course, this would be the ideal scenario, and the workplace would be free from sex pests and harassers if every incident was to be dealt with in that manner.”

[90]          When seen in this light, the five foundations of the Commissioner’s award cannot withstand scrutiny. Each premise either lacks foundation on the evidence, or betrays an insensitivity to the nature of sexual harassment.

[91]          I am alive to previous injunctions against adopting a piecemeal approach to reviews. By engaging with the enquiry in this way, I merely aim to show how, individually and cumulatively, the Commissioner erred.

The first conclusion: 2 August 2016

[92]          The Commissioner’s first conclusion relates to the events of 2 August 2016. 

[93]          As a starting point, it is not clear whether the Commissioner found (i) that the incident took place, but did not amount to sexual harassment, or (ii) that the incident did not take place at all. The Commissioner’s analysis seems to focus solely on Ms S’s failure to report the incident and her alleged complicity. The Commissioner simply concludes that “…Ms S was not so afraid that she could not raise the alarm on or immediately after 2 August 2016.”

[94]          This finding sheds no light on the issue that the Commissioner was required to resolve. The issue was whether the incident took place and, if so, whether it amounted to sexual harassment.

[95]          The conclusion is capable of two interpretations. On one hand, it can be read to mean that the incident likely did not take place because Ms S did not report it timeously, with no good reason. On the other hand, it could mean that the incident indeed occurred, but that Ms S was time-barred from raising the complaint.

[96]          Neither of these constructions is reasonable or justifiable on the evidence before the Commissioner.

[97]          If the Commissioner intended to infer from Ms S’s failure to report the incident immediately after it occurred (or within the time periods set in the Policy) that it did not take place, then this finding is plainly wrong.

97.1       First, this finding jars with the uncontroverted evidence that Ms S reported the matter to Mr Mothlabane on 28 September 2016. This was within the two-month period contemplated in the Policy. Both Ms S and Mothlabane testified that Ms S reported the incident. Ms S also repeated her version of the incident to Mr Khubeka on 8 October 2016. 

97.2      Second, assuming that Ms S reported the incident late, one cannot infer from this alone, that the incident did not take place.

97.3      Third, Ms S tendered two reasons for not reporting the incident. Ms S testified that she was traumatised and was afraid that she did not have sufficient evidence of the incident. Given that there were no other witnesses to the incident, there is nothing unreasonable about this evidence. It is not uncommon for victims of sexual harassment to fear being disbelieved and the consequences thereof. Nor is it uncommon for victims to try to process the incident in their own minds before taking an action.

97.4    This cannot reasonably form the basis for rejecting her version. The Commissioner failed to appreciate that there is no standard reaction to trauma. This is not the last time the Commissioner commits this error.

[98]          If the Commissioner intended to convey that Ms S was time-barred, then the Commissioner misconceived the enquiry and did not actually decide the issue. 

98.1    The issue was simply (i) whether the incident of 2 August 2016 took place, and (ii) if so whether it amounted to sexual harassment. Neither of these questions are answered with reference to the timing of Ms S’s complaint.  

98.2    This finding has its roots in Mr Makanda’s procedural complaint. He contended that the Policy required an incident to be reported within two months, failing which it could no longer be raised. He relied on clause 6.1,[31] and Appendix 1,[32] to the Policy.

98.3     The proposition that these provisions of the Policy bar a complainant from raising a complaint about one of the most heinous forms of workplace discrimination, and one of the worst invasions of a woman’s bodily integrity, lacks rigour. It cuts against an employer’s fundamental duty to ensure a workplace that is free from sexual harassment.

98.4       It seems plain that the purpose of these provisions was to encourage victims of sexual harassment to lodge their complaints as soon as possible, to procure expeditious resolution of the issue and to avoid losing evidence. It cannot be that the employer’s duty to prosecute a disciplinary hearing prescribes by virtue of the late referral. A proper reading of the Policy demonstrates this, because it does not prevent the employer from prosecuting a complaint that is filed outside the two-month window.

98.5      In any event, time-barring is irrelevant to the dispute. At best for Mr Makanda, this could only go to procedural fairness. It does not tend to prove or disprove, as the Commissioner suggests, that the incident did not take place.

[99]          What is perhaps more pungent, is the Commissioner’s inference that Ms S sought to “…underplay her part in the process.” I describe this finding in such strong terms for good reason.

99.1   First, it is based on the assumption that Ms S “led on” Mr Makanda. This, in the face of her evidence that Mr Makanda’s advances were unwelcome. Even if they had been welcomed at a stage (which was not the evidence before the Commissioner) it ignores the basic principle that prior consensual participation does not mean that the conduct continues to be welcome.

99.2     Second, to reach this conclusion the Commissioner must necessarily have found that the incident of 2 August 2016 took place. The award is bereft of any such finding.  

99.3   Third, the inference is illogical. The Commissioner drew this inference from the text messages exchanged on 18 October 2016. He used this to suggest that Ms S had a role to play in the events of 2 August 2016, over two months earlier. This conclusion does not follow. Nothing in the text message invited the Commissioner’s conclusion. The fact that Ms S sent a text message describing the definition of a subliminal message two months later, does not tend to prove or disprove that the incident of 2 August 2016 took place. Nor does it prove that she was somehow complicit in her harassment.

99.4    Most fundamentally, to draw this inference, the Commissioner had to believe that (i) Mr Makanda’s defence was that his alleged advances were consensual (and thus not unwelcome), or (ii) that Ms S invited Mr Makanda’s alleged advances. This was never the case before the Commissioner. Mr Makanda denied the events of 2 August 2016. He could not simultaneously maintain that denial and contend that the same alleged advances were consensual.

[100]       Finally, the Commissioner was wrong in his observation that Ms S did not testify about the text messages during her evidence in chief. I have already referred, in the summary of her version above, to the portion of the transcript in which Ms S testified about the text message during evidence in chief.[33] In any event, the text messages did not assist in determining whether the incident of 2 August 2016 took place.

[101]       In summary, the entire objective of the inquiry appears to have eluded the Commissioner. His findings in respect of the 2 August 2016 incident are misconceived. Virtually nothing mentioned in the Commissioner’s reasoning determines the actual issue. Precious little is rationally connected with the evidence before the Commissioner. Each constituent conclusion is so patently flawed as to fall outside the limits of reasonableness.

The Second Conclusion

[102]       The Commissioner’s second conclusion was that Ms S’s conduct on 18 October 2016 was inconsistent with a victim of harassment. This was because Ms S did not confront Mr Makanda directly and her approach gave Ms Makanda an incorrect impression.

[103]       Again, the Commissioner failed to grasp or determine the issue. The issue was whether the discussion between Ms S and Mr Makanda took place as Ms S alleged it, or as Mr Makanda alleged it. If the Commissioner found the former, it had to follow that Mr Makanda was guilty of sexual harassment.

[104]       The Commissioner’s conclusion seems to suggest that he accepted Ms S’s version, but found that Ms S invited Mr Makanda’s advances by confronting him in the way that she did. Thus, consent was achieved and Mr Makanda’s actions were welcome. There are significant difficulties with this.

[105]       Fundamentally, the Commissioner’s finding is a throw-back to times where anything short of rebuffing an unwelcome advance in the strongest terms, sufficed for consent. This finding is well outside the bounds of reasonableness. What is more, it is out of step with the Legislature’s laudable attempts to ensure a gender-responsive dispute resolution mechanism, which are set out in the Code of Good Practice on the Handling of Sexual Harassment in the Workplace.

105.1          Item 5.2.1 of the Code recognises that: “There are different ways in which an employee may indicate that sexual conduct is unwelcome…”

105.2          This item expresses an understanding that there is no standard uniform reaction to sexual harassment. It recognises the invidious position in which victims find themselves. I cannot express it better than the Industrial Court did J v M Ltd:[34] 

“The victims of harassment find it embarrassing and humiliating. It creates an intimidating, hostile and offensive work environment. Work performance may suffer and career commitment may be lowered. It is indeed not uncommon for employees to resign rather than subject themselves to further sexual harassment. The psychological effect on sensitive and immature employees, both male and female, can be severe, substantially affecting the emotional and psychological well-being of the person involved. Inferiors who are subjected to sexual harassment by their superiors in the employment hierarchy are placed in an invidious position. How should they cope with the situation? It is difficult enough for a young girl to deal with advances from a man who is old enough to be her father. When she has to do so in an atmosphere where rejection of advances may lead to dismissal, lost promotions, inadequate pay rises, etc - what is referred to as tangible benefits in American law - her position is unenviable.

Fear of the consequences of complaining to higher authority whether the complaint is made by the victim or a friend, often compels the victim to suffer in silence. That sexual harassment of an employee in an inferior position is despicable is only fully realized when one has to comfort a young girl crying her heart out in a quiet corner.”[35]

105.1          The Commissioner appears to have expected Ms S to react in a confrontational manner, leaving no room for her to approach the issue in the manner she saw fit. In the Commissioner’s mind, anything short of this justified the inference that she invited what followed.

105.2          Even the Commissioner’s formulation of the standard response, such as it is, is problematic to say the very least. The subtext is that, unless the victim expressly says “…I don’t relish it…” or words to that effect, the perpetrator can safely assume that she does. 

[106]       At any rate, Ms S chose to introduce the conversation with a false rumour. It is not difficult to imagine why, given the sensitivity of the topic, and the power dynamics at play. It is also difficult to conceive how the Commissioner could infer that Ms S “had something to hide” by approaching the issue the way she did. Nothing in the award sheds light on what this could be. 

[107]       There was no factual basis to infer that, by introducing the conversation as she did Ms S “…gave the applicant a wrong impression.” It was never Mr Makanda’s version that he gained the wrong impression from Ms S. He did not testify that he understood the rumour to be an invitation to express the fullness of his alleged attraction to Ms S. Quite the contrary, Mr Makanda maintained that he took exception to the rumour.

[108]       In short, the Commissioner failed to decide the dispute on the evidence before him.  He misconceived the kernel of the actual dispute insofar as it related to the October incident.  As a result, his conclusions, and inferences drawn, cannot be traced to the evidence before him.

The Third Conclusion

[109]       The third foundation is that Ms S did not state the facts which gave rise to her conclusion that Mr Makanda was aroused during their conversation on 18 October 2016. The Commissioner said this only arose in cross-examination. It did not. The Commissioner appears to have missed the portion of Ms S’s evidence where she stated, in some detail, what happened during the discussion. There could be no reasonable conception of her description of the conversation being anything other than overtly sexual, let alone one characterised by sexual undertones.

[110]       Having read the record, I can only assume that this conclusion was formed from an unfortunate line of cross-examination by Mr Makanda’s representative:

“COUNSEL: Okay, look at Mr Makanda’s physical structure would you see when he has an erection.

MR PATEL: Sorry Mr Arbitrator, I am not sure what is happening.  Is he giving evidence?

MR COMMISSIONER: No, he is just demonstrating, he has already said that the way he is physically you cannot see his erection and so he has gone backwards. It is not like giving evidence.

COUNSEL: Yes.

MR COMMISSIONER: He is merely showing that this is just ocular, there is nothing here.  He is not saying anything.  He is simply saying that if you look, he has a pouch and he says if you are there he says you cannot see, so there is nothing here…(inaudible)… What is your comment Ma’am?

MS S: Yes, I saw his erection.

MR COMMISSIONER: That is it.

COUNSEL: By the show of hand, how big was it?

MS S: I do not know... (inaudible)…”[36]

[111]       The last question was gratuitous. Counsel sought to extract the concession that Ms S was not in a position to see Mr Makanda’s state. The witness maintained that she could. That was enough. Unless counsel sought to put a factual version to the witness concerning Mr Makanda’s endowment (which in itself would be of questionable value), then he could go no further. There was no value in the remaining line, except to embarrass the witness. This is precisely the type of questioning that leads to the notorious under-reporting of sexual harassment incidents. The Commissioner did not intervene. On the contrary, he appears to have adopted the underlying reasoning lock stock.

[112]       That aside, the issue begins and ends with the Commissioner’s failure to consider Ms S’s evidence.

The fourth conclusion

[113]       The fourth conclusion is a credibility finding against Ms S. The Commissioner preferred Mr Makanda’s version because he remained consistent under cross-examination. According to the Commissioner, Ms S prevaricated and appeared ruffled. This, in the Commissioner’s view, tilted the scale of probability in favour of the applicant. 

[114]       Although credibility findings are not immune from review, a court should be loath to interfere with such findings. It should only do so if the finding is completely out of kilter with the evidence on the record and the probabilities on the whole.[37]

[115]       Factors such as demeanour are indeed best observed by the Commissioner. However, others should manifest themselves in the record. As this Court observed in Harmony Gold Mining Co Limited v Commission for Conciliation, Mediation and Arbitration and Others [38]

“What the analysis above makes clear is that observations of demeanour are merely one factor among many in assessing credibility and many factors bearing on credibility will be apparent from the transcript of evidence. It is also obvious that credibility findings based on observation of the witness are not the only or the first recourse in assessing credibility and even less so in evaluating probabilities. Adjudicators should be wary of making definitive credibility findings based on their supposed omniscient ability to detect unreliable evidence solely from observing a witness.”

[116]       The pith of the dispute is whether the Commissioner’s acceptance of Mr Makanda’s version over Ms S’s is a decision that a reasonable decision-maker could not reach. This formulation was recently endorsed by the LAC in Department of Health, Kwa-Zulu Natal v Public Servants Association of South Africa and Others. [39]

[117]       The Commissioner’s credibility finding was based on Ms S’s prevarication and her demeanour. 

[118]       The Commissioner does not indicate in what respect Ms S prevaricated. This is unsatisfactory. The Commissioner’s central role is to determine the dispute between the parties. If this dispute turns on a credibility finding, it behoves the Commissioner to set out the reasons for that credibility finding in sufficient detail. This does not require painstaking detail. The Commissioner should simply cite the facts from which he concluded that Ms S prevaricated. The parties should not be left guessing about the basis on which their dispute was determined, as they have been in this case. 

[119]       The specific instances where Ms S prevaricated under cross-examination should be manifest from the record. They are not. Mr Makanda’s Counsel suggests three main inconsistencies from which the Commissioner concluded that Ms S prevaricated. None of these are manifest from the record. Nor can they reasonably form the basis of a credibility finding. 

[120]       The first inconsistency, is that Ms S was untruthful about the advice she received from Mr Khubeka. This can only be based on Mr Khubeka’s evidence where he denied that he advised Ms S to commence the discussion with a rumour. Mr Makanda appears to conflate internal inconsistencies within Ms S’s evidence and evidence that is rebutted by another witness. In the former, a witness can be said to “prevaricate”. In the latter, a witness would remain consistent in their version, but their evidence would be contradicted by another witness. Ms S remained consistent throughout cross-examination on this score. She consistently maintained that Mr Khubeka advised her to commence the discussion with a rumour. She did not prevaricate in this regard.

[121]       Properly construed, counsel’s argument on this score is that Mr Khubeka’s evidence contradicted Ms S’s. Even this does not bear scrutiny because Mr Khubeka testified that commencing the conversation with a rumour was certainly part of the discussion that evening. He only denied that it was his advice.

[122]       The second inconsistency relates to the manner in which Mr Makanda telephoned Ms S. The trouble with this argument, is that it was common cause that Mr Makanda telephoned Ms S. How he did so, is immaterial and cannot reasonably form the basis of a credibility finding. A witness cannot be expected to remember, in meticulous detail, whether a discussion was conducted by landline or mobile. 

[123]       The third inconsistency relates to whether the conversation in November 2016 took place in the basement parking. It was put to Ms S that Mr Makanda parks at surface level. Again, this is immaterial. It does not tend to prove or disprove that the conversation did not take place. 

[124]       I turn now to the more difficult question of demeanour. In principle, a review court should be deferent to the Commissioner’s findings on demeanour. With good reason: the trier of fact is steeped in the atmosphere of a trial, while the review court is limited to the record. The trier of fact is, thus, uniquely placed to observe the demeanour of a witness. 

[125]       In President of the Republic of South Africa and Others v South African Rugby Football Union and Others,[40] the Constitutional Court sounded a warning, about the weight to be placed on the demeanour of a witness. It observed, with reference to S v Kelly, [41] that demeanour is a tricky horse to ride and that of all the factors involved in assessing credibility, the witness’s demeanour can be the most misleading. This is because, people differ in nature. A crafty witness can appear honest, while a truthful witness, who is shy in nature, can appear dishonest. As such, the truthfulness or otherwise of a witness can rarely be determined by reference to demeanour alone, especially without regard to the probabilities. It said:

“The advantages which the trial court enjoys should not, therefore, be over-emphasised ‘lest the appellant’s right of appeal becomes illusory’.  The truthfulness or untruthfulness of a witness can rarely be determined by demeanour alone without regard to other factors including, especially, the probabilities. As indicated above, a finding based on demeanour involves interpreting the behaviour or conduct of the witness while testifying. The passage from S v Kelly above correctly highlights the dangers attendant on such interpretation.  A further and closely related danger is the implicit assumption, in deferring to the trier of fact’s findings on demeanour, that all triers of fact have the ability to interpret correctly the behaviour of a witness, notwithstanding that the witness may be of a different culture, class, race or gender and someone whose life experience differs fundamentally from that of the trier of fact.”[42] (my emphasis)

[126]       I emphasise the last portion of this extract, because it bears direct relevance to this case. Having been steeped in the atmosphere of the arbitration, the Commissioner described Ms S’s appearance as “ruffled” at stages. Ms S might well have appeared “ruffled”. If she did, that would be understandable. Ms S was subjected to the most uncomfortable cross-examination, involving questions of the type I mentioned earlier. 

[127]       Commissioners should be alive to the sensitivity involved in matters of sexual harassment, involving as it inevitably does, gender imbalances in society. The mere fact that Ms S appeared ruffled, is not necessarily an indication that she was being untruthful. It could indicate the contrary, because one could expect a victim of sexual harassment to be traumatised. Worse still, when she is asked to relive the events in explicit detail.

[128]       The Commissioner was unsympathetic to this possibility. This is reflected throughout his award, which is underpinned by chauvinist assumptions such as —

128.1          The notion that a woman can only respond to sexual harassment in a directly confrontational manner.[43] Anything shy of that, invites the inference that the advances are welcome;

128.2          The notion that a woman who does not “raise the hue and cry” immediately is unlikely to have been the victim of sexual harassment;[44]

128.3          The notion that there is a standard way to confront sexual harassment at all is problematic. More so, the notion that a woman who refers to a rumour about sex, invites anything that follows;[45] and

128.4          The notion that, by sending an innocuous text message, Ms S invited what followed.[46]

[129]       As a result, I cannot reasonably defer to the Commissioner’s finding on Ms S’s demeanour. 

The fifth premise

[130]       The Commissioner found that, having regard to industry,  invitations to lunch could not amount to sexual harassment. Although the Commissioner accepted that Mr Makanda’s telephone records reflected several calls to Ms S, he was not inclined to “second-guess the contents of these calls”.  Both of these components warrant interrogation.

[131]       The general statement that invitations to meals outside the office cannot, on their own, amount to sexual harassment is acontextual. It loses sight of the fact that an invitation to lunch can, in certain circumstances, be an unwelcome advance of a sexual nature. It can also simply be purely work-driven and thus, not a sexual advance. Either way, the nature of the advance can only be determined by reference to the context. 

[132]       It is no good to have regard only to the prevailing context within the industry, as the Commissioner did. It is difficult to conceive of any industry in which co-workers do not invite one another to social events or meals. The real issue is whether the particular invitations amounted to sexual harassment. The marrow of the dispute lies in the particular context of those invitations. It would involve interrogating, inter alia, (i) the nature of the relationship between both parties; (ii) events that have passed; (iii) how the invitation was conveyed; (iv) the stated intention behind the invitation; and (v) the basis of the perception that the invitation was a sexual advance. These considerations are common sense, and by no means, a closed list.

[133]       The Commissioner was invited to consider these circumstances. Ms S testified that Mr Makanda made several telephone calls to her, in which he invited her to late lunches and dinners outside the office. Mr Makanda denied making any telephone calls to Ms S. He tendered his telephone records as proof of this. In cross-examination, he conceded that Ms S’s number appeared in his telephone log. Hence it was established, as a fact, that Mr Makanda made the telephone calls to Ms S. What remained was the content of these telephone calls. 

[134]       The Commissioner declined to “second guess” the content of the telephone calls. There was no need for him to do so. The Commissioner had Ms S’s version of the discussions. His role was to establish whether Mr Makanda made unwarranted sexual advances, by reference to the factors I have described above. He did not do so. In the result, the Commissioner actually failed to decide the issue and to grapple with the evidence. 

The Consequence of these Irregularities

[135]       The sum of the above analysis is that the award falls to be set aside on review. This is primarily so because the Commissioner failed to perform his dispute resolution function. In instances where he performed this function, the Commissioner appears to have either wholly misconceived the nature of the inquiry, or rendered an outcome that bore no rational connection to — or could not reasonably have arisen from — the evidence led.     

[136]       What is more, to the extent that it is relevant given the misdirections above, the Commissioner’s decision is not one that a reasonable decision-maker could reach.

Substitution

[137]       Having found the award to be reviewable, the next issue is whether to substitute the award or refer it back to the CCMA.

[138]       I am minded to determine the matter. There is a full record, comprising over 1400 pages, capturing both the disciplinary and arbitration proceedings. The evidence has been fully ventilated. This Court is as well-placed to decide the matter as the CCMA would be. I now turn to this task.

[139]       I have already summarised both parties’ versions of the events of 2 August 2016. In short, Ms S contends that Mr Makanda forced himself on her in her office. Mr Makanda admits that he was in the office all day, but denies ever seeing Ms S that day. This factual dispute is to be determined by reference to the credibility and reliability of both parties’ versions, along with the probabilities.

[140]       Nothing in the transcript suggests that Ms S was not a credible witness. She was entirely without guile. She made the necessary admissions, even when they were not ideal. 

[141]       Mr Makanda sought to challenge Ms S’s bias. He suggested that Ms S was part of a conspiracy to prevent his promotion. In support of this, Mr Makanda contended that Ms S’s report coincided with his interview for a promotion. Ms S gained nothing by reporting the alleged incidents of sexual harassment. Quite the contrary, in doing so, Ms S would have had to endure scrutiny into her private life throughout the course of several proceedings. Second, one cannot conclude that the timing of her report was designed to scupper Mr Makanda’s promotion, in an interview process that occurred in February 2017 — Ms S informally reported the incident from as early as September 2016. She repeated her version of the incident in a subsequent discussion with Mr Khubeka. This occurred months before the interview process. It might well be that her formal report coincided with the interview process in February 2018. Ms S’s explanation for the delay in reporting the incident, was that she needed time to (i) engineer her move to the Sandton office and, (ii) once that move succeeded, to settle in. There is nothing inherently problematic with this explanation. 

[142]       I cannot find any material contradictions, whether internal or external in Ms S’s version. Ms S consistently maintained her version of the incidents. Quite apart from the two occasions when she did so with her line managers, which were contemporaneous, Ms S later maintained that Mr Makanda inappropriately touched her in a text message to another colleague.[47] Even after that, Ms S’s version remained consistent during the disciplinary hearing and in the arbitration. This despite probing, uncomfortable, and at times inappropriate cross-examination.

[143]       There was nothing inherently improbable about Ms S’s version. Mr Makanda sought to challenge this in two ways.

[144]       First, Mr Makanda said it was improbable that he would approach Ms S in her office without being seen because (i) there were people in the office, and (ii) people could easily see into Ms S’s office. Ms S countered this by saying that, at the time, the office was relatively empty, given the impending public holiday. She admitted that her office was visible from the outside. However, since the glass was partly sandblasted, it would not be easy for people to see inside. This was not meaningfully challenged. To the extent that Mr Makanda’s questions went to his own risk assessment, this was irrelevant. Ms S cannot speak to Mr Makanda’s internal risk assessment. If this was tendered to demonstrate the unlikelihood of the event having occurred at all due to the lack of corroborative witnesses, this was sufficiently countered by Ms S’s evidence that there were few people in the office at the time.

[145]       Second, Mr Makanda challenged Ms S’s evidence that he told her that he was en route to Bedfordview. He tendered his travel log to prove that he did not travel at all that day. This fact is neutral. On one hand, it tends to prove that Mr Makanda did not tell Ms S that he was en route to Bedfordview. It does not, however, tend to prove that the incident did not take place. Quite the contrary, Mr Makanda’s travel logs place him in the office on the day of the alleged incident.

[146]       On the probabilities, there is a logical flow to Ms S’s evidence. It can simply be summarised as follows. On 2 August 2016, Mr Makanda attempted to push himself on her. She reacted with shock. She ultimately reported the incident to her line manager. In turn, he reported it to Mr Khubeka. Both suggested that Ms S confront Mr Makanda about the incident. She did so. This led to the 18 October 2016 meeting. Although she experienced further harassment during this incident, both she and Mr Makanda agreed to limit their contact. Ms S thought the issue had been resolved. However, in November 2016, she had a discussion with Mr Makanda which suggested otherwise. Consequently, she extricated herself from that working space, and moved to Sandton. Having settled down, Ms S decided to formally report the issue.

[147]       The same cannot be said of Mr Makanda’s version. Mr Makanda’s narration of the facts, from the very beginning, casts doubt on his veracity. Mr Makanda testified that, one of his first interactions with Ms S took place in April 2016, when she rubbed his hand and spoke of her dissatisfaction in her marriage. I find this version hard to believe. The very first time it was raised, was in the course of disciplinary proceedings. At no stage before that, did Mr Makanda think to raise it. On his own version, Mr Makanda did not even raise it during the October incident, when it was ripe for him to do so. He tendered no explanation for his failure to do so. The irresistible inference is that the April 2016 incident was contrived in an attempt to cast doubt over Ms S’s character. This inference is buoyed by the fact that Mr Makanda failed to call Ms Cima to testify in support of the April 2016 incident at the arbitration.

[148]       Mr Makanda’s version of the discussion of 18 October 2016 also warrants interrogation. Only two aspects of this discussion are common cause: how the discussion commenced, and how it ended. What happened in between is disputed. Mr Makanda’s version is problematic in several respects, which indicate that the conversation did not go as he says it did.

[149]       First, Mr Makanda’s narration of the October 2016 incident was inconsistent. During the disciplinary hearing, Mr Makanda testified that he encouraged Ms S to record their conversation.[48] He did not give this version during the arbitration. This inconsistency is relatively minor, and could be overlooked if it was the only one. However, it was not. 

[150]       Mr Makanda also testified that, at a stage, he wanted to telephone Mr Khubeka and Ms S asked him not to do so as “…she had approached him in confidence.” This beggars belief because Ms S confronted Mr Makanda on Mr Khubeka’s advice and instruction. In fact, Ms S told Mr Makanda that Mr Khubeka had reprimanded her for their relationship. It is inconceivable that she would tell Mr Makanda that she had approached him in confidence. 

[151]       Third, Mr Makanda did not challenge Ms S about her alleged actions in April 2016, when she rubbed his hand. Ms S had confronted him with rumours of an amorous affair between them. On Mr Makanda’s narration of events, these could only have arisen from the events of April 2016. Yet Mr Makanda remained silent. He did not confront Ms S about this event at all. It would have made sense for him to do so, given that this was the only incident of sexual contact between them. It is inconceivable that Mr Makanda could have taken exception to the allegations proffered by Ms S without so much as referring to this incident.

[152]       Fourth, pursuant to the discussion in October 2016, Mr Makanda faced a serious operational issue, to which Mr Makanda did not react as one would expect of a reasonable manager. On his version, Mr Makanda told Ms S, in no uncertain terms, never to darken his door again. If he is to be believed, the reason for this was serious: Ms S falsely implicated him in an affair with her and in an attempt to cover it up. These were serious allegations. It is inconceivable that he would not report that issue to Ms S’s line manager (Mr Khubeka) at all, let alone immediately afterwards. Any reasonable person would have done so, at the very least, to manage the risk of working with Ms S and being exposed to future accusations.   

[153]       The inescapable inference to draw from Mr Makanda’s failure to report the outcome of the meeting, is that the circumstances from which it arose were embarrassing. Such embarrassment could only have arisen from the manner in which he conducted himself in that meeting. I pause to note that Ms S also failed to report the October 2016 incident to Mr Khubeka. The difference is that Ms S explained her failure to do so. She said, despite the egregious nature of what occurred, she thought the matter was resolved by the agreement to maintain distance. No explanation appears at all from Mr Makanda.

[154]       Finally, Mr Makanda was untruthful about the telephone conversations. His initial version was that he did not telephone Ms S at all. He tendered telephone records as proof. However, under cross-examination, Mr Makanda conceded that Ms S’s telephone number appeared on his call logs.     

[155]       Seen in this light, Mr Makanda’s version is woefully deficient. It is riddled with fundamental internal and external contradictions. There are also inherent improbabilities and unexplained gaps. No reasonable decision-maker, having grappled with this version, could accept it over Ms S’s.

Conclusion

[156]       In summary, it is apparent that the Commissioner committed several irregularities. The most fundamental, is that he failed to properly assess and resolve the dispute between the parties. The Commissioner dismissed Ms S’s version as implausible, without analysing it against the competing version. In doing so, the Commissioner erroneously construed the evidence and the positions of both parties.

[157]       In cases involving mutually destructive versions of single witnesses, it is vital to assess the evidence of both parties. This is more so in cases involving sexual harassment, involving as it does, potentially devastating consequence of being labelled a “sex pest” or a liar. It is fundamental that the victim’s version be tested against that of the alleged perpetrator and the objective facts.

[158]       Only after a complete consideration of both versions can it be said that the dispute has been adjudicated. This analysis should appear from the Commissioner’s award. This is not to say that the Commissioner’s award should contain an exhaustive analysis such as the one I have attempted above. However, at the very least, it should show why one version was preferred over the other. Anything short of that, is not a quasi-judicial decision, but a vote. It is not enough to proffer a catch-all phrase to the effect that the Commissioner has considered the evidence holistically. If this is not apparent from the Commissioner’s reasons, then this phrase is a mere platitude.

[159]       In light of my findings above, there would be no point in remitting the matter to the CCMA for fresh consideration. In any event, little could be achieved by this, save for another painstaking process. As such, I am inclined to substitute the award.

[160]       In the premises, I make the following order:

Order

1.    The arbitration award issued by the Third Respondent is reviewed and set aside, and substituted with the following order:

“the dismissal of the first respondent was procedurally and substantively fair.”

2.    Each party is to pay its own costs.

M Sibanda

Acting Judge of the Labour Court of South Africa

Appearances:

For the Applicant:               Advocate F Boda SC

Instructed by:                     Cliffe Dekker Hofmery Inc

For the First Respondent:   Advocate Mashele

Instructed by:                      Maja Inc

---