Minister of Safety and Security v F (592/2009) [2011] ZASCA 3 (22 February 2011)

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

Case No: 592/09

In the matter between:

THE MINISTER OF SAFETY AND SECURITY Appellant

and

F Respondent

Neutral citation: The Minister of Safety and Security v F (592/09) [2011] ZASCA 3 (22 February 2011)

Coram: NUGENT, MAYA, SNYDERS and BOSIELO JJA and R PILLAY AJA

Heard: 11 NOVEMBER 2010

Delivered: 22 FEBRUARY 2011

Summary: Vicarious liability – police officer – crime committed while not on duty but on ‘call’ – whether state liable.

_____________________________________________________________________

ORDER

_____________________________________________________________________

On appeal from: Western Cape High Court (Bozalek J sitting as court of first instance).

The appeal is upheld. The orders of the court below so far as they relate the appellant (first defendant in the court below) are set aside and the following is substituted:

‘The claim against the first defendant is dismissed’.

_______________________________________________________________________

JUDGMENT

_______________________________________________________________________

NUGENT JA (SNYDERS JA and R PILLAY AJ concurring)

[1] K v Minister of Safety and Security¹ concerned a claim by a woman who was raped by three policemen. She had encountered them at a petrol station where she was stranded in the early hours of the morning. The policemen were on duty at the time, they were in uniform, and they were in a marked police vehicle. They offered to take her home and she readily accepted. Instead she was driven to a quiet place where she was raped. Needless to say, the policemen were all delictually liable for their conduct, but that was not the issue in the case. The issue was whether the State – nominally represented by the Minister of Safety and Security – was vicariously liable for their conduct. The Constitutional Court held that it was.

[2] This case seeks to take that a step further. The respondent in this case – who I will refer to as Ms F – was also raped by a policeman. Ms F had similarly found herself stranded late at night and the policeman offered to drive her home. Instead he drove to a remote spot where he raped her. The distinction between this case and K is that on this occasion the policeman was not on duty. Once more the question is whether the State is vicariously liable for the consequences of his conduct. The court below (Bozalek J in the Western Cape High Court) held that it is.² The Minister now appeals with the leave of that court.

[3] The policeman concerned was Mr van Wyk, a detective who was stationed at George. He was off duty but on ‘standby’ or on ‘call’ – as it was variously called in the evidence – at the time the incident occurred. A former senior police officer – Mr Du Toit, who was then a Provincial Commander: General Investigations, in command of all detectives in the Western Cape – explained how the ‘standby’ system worked at the time.

[4] The ordinary working hours of detectives were from 07h30 to 16h00 but every station had to make officers available on a ‘standby duty roster’. This meant that a detective who had been rostered was entitled to return home and go about his or her private business in the ordinary way after completing his or her shift but could be called upon to resume duty at any time between shifts. To enable him or her to respond to a call to resume duty a state vehicle was made available to the detective. If called upon to resume duty the detective was required to note the period for which duty was performed in his or her pocketbook and later to present it to his or her commanding officer.

[5] Detectives who were rostered were not paid for the time that they performed duty while on call but they received a ‘standby allowance’. The standard conditions of service in that regard provided as follows:

‘A non-pensionable Standby Allowance at a tariff of R16.80 is payable to officials who must be available for 24 hours per day for the performance of duty (This must be regarded as payment for overtime duties performed.)’

The conditions went on to provide:

‘The Standby Allowance was instituted to compensate for the restriction of movement placed on personnel on Standby duty and their households. This implies that these personnel have to be available at their dwelling in order to be available for duty at short notice unless where special alternative arrangements have been made’.

[6] The incident with which we are concerned occurred in 1998 when Ms F was thirteen years old. Summons was issued in 2008, after she had reached majority, and the action was tried in 2009. Both Ms F and Mr van Wyk gave evidence. The evidence of Mr van Wyk was in some respects remarkably candid and by and large it coincides with that of Ms F but in some respects their evidence diverges. In some such instances the evidence of Ms F was vague or uncertain or even contradictory but bearing in mind her age at the time and the long interval before she gave evidence that is not unexpected. The court below rejected the evidence of Mr Van Wyk where it diverged from the evidence of Ms F. On the view that I take of the matter I need not evaluate that finding – I have accepted the evidence of Ms F wherever it diverges from that of Mr Van Wyk.

[7] On 14 October 1998 Mr van Wyk was on the standby roster. His shift ended at 16h00 in the ordinary course but he remained on duty until about 20h00, apparently to complete various tasks. He then left the office and returned home in a police vehicle, which was unmarked but equipped with a communication radio. Various dockets that he had been working on were in the vehicle. He was not called out to resume duty at any time that night or the following morning.

[8] At about 22h00 Mr van Wyk visited a nightclub with two friends to have a few drinks. Although he was prohibited from using the police vehicle for private purposes he nonetheless used it to drive to the nightclub. He was dressed in civilian clothes.

[9] Ms F also visited the nightclub that night in the company of friends. At about midnight she fell out with one of her friends and decided to return home. When she left the nightclub she encountered Mr Van Wyk and his two friends – Mr Petrus Faniso and Mr Edward Botha. There is a slight lack of clarity in the evidence on this point (which is not really material) but I accept that the three men were standing alongside the vehicle when Ms F first saw them and that she then joined them and they stood talking for a while. Ms F did not know Mr van Wyk but she was casually acquainted with Mr Faniso. Ms F said that while they were talking she noticed ‘from the corner of her eye’ that the vehicle was equipped with a radio of the kind that is used by the police. When asked what was troubling her she told the men that she had fallen out with her friend and that she wanted to go home and one of the men said that they would drive her home.

[10] They all left in the vehicle. First Mr Botha was dropped off, then Mr Faniso. When Mr Faniso was dropped off Ms F, who had been seated in the rear, moved to the front passenger seat. Ms F was then alone in the vehicle with Mr van Wyk. Mr van Wyk told her that he wanted to drive to Kaaimans to see if friends of his were there and they set off in that direction. Ms F said that he stopped the vehicle at a place that was dark – apparently in the vicinity of Kaaimans. She felt that ‘something was wrong’ and she leapt from the vehicle and hid herself. Mr van Wyk drove off. After a while Ms F returned to the road hoping to flag down a passing vehicle. A little later Mr van Wyk returned. Pretending not to know her he asked what she was doing there. She replied that he knew very well why she was there and she told him that he must take her home, which he undertook to do. Once more she entered the vehicle and they drove in the direction of George. Shortly before they reached George he suddenly turned off the road. Ms F tried to leap from the vehicle but Mr van Wyk restrained her. He then stopped the vehicle and assaulted and raped her. Afterwards he drove her to her home in George, threatening her with harm if she disclosed what had occurred. Later that morning Ms F reported the events to her mother, the police were informed, and Mr van Wyk was arrested.

[11] There are further aspects of the evidence of Ms F that I need to deal with specifically. I have said that there were police dockets in the vehicle. There is a suggestion in one part of the evidence of Ms F, in answer to an ambiguous question, that she observed the dockets before she first entered the vehicle. But I think it is clear from other parts of her evidence that she noticed the dockets only later. According to Mr van Wyk the dockets were at first on the floor at the foot of the front passenger seat. He said that when Mr Faniso was dropped off he took them from the floor and placed them at the foot of the rear seat before Ms F took her place in the front passenger seat. From the evidence as a whole, seen in the sequence within which events occurred, I think it is probable that Ms F saw them only then.

[12] Ms F said that in reply to her query relating to the dockets Mr van Wyk told her that he was a ‘private detective’, which she understood to mean that he was a police officer. Mr van Wyk denied that he had told her that he was a ‘private detective’ but I accept her evidence in that regard, and I accept that she understood that to mean that he was a policeman. Ms F was not able to say precisely when on the journey this occurred. But I think it can be inferred that it was probably shortly after the two men had been dropped off and while they were driving to Kaaimans, which is consistent with the finding of the court below.

[13] Although Ms F said that she noticed the police radio before she first entered the vehicle I agree with the finding of the court below that ‘on the probabilities when [Ms F] first accepted a lift from Van Wyk the fact that he was or may have been a police officer played no role in her decision’. In a statement that she made to the police on 15 October 1998 there was no suggestion that she entered the vehicle in the belief that Mr van Wyk was a policeman nor was that suggested in her evidence in chief. In cross-examination it was put to her that the reason that she had entered the car was that she knew Mr Faniso to which she replied: ‘I think so’. It was also put to her more than once that it was not because she thought Mr van Wyk was a policeman that she had entered the vehicle to which she consistently answered that she could not remember. If she had indeed been motivated to enter the vehicle by a belief that he was a policeman I have no doubt that her evidence would have been explicit in that regard.

[14] But by the time they reached Kaaimans she was indeed aware that he was a policeman (for the reasons I have given). Ms F said that she entered the car on that occasion because she ‘trusted’ him because he was a policeman. No doubt her knowledge that Mr Van Wyk was a policeman played some role in her decision but I do not think it can be inferred from her evidence that Ms F would not otherwise have done so. He had already given her good reason to believe that being a policeman was by itself no guarantee that he was to be trusted. She was stranded in the dark well after midnight, with no other apparent means of getting home, and she might just as well have accepted his renewed promise to drive her home even if he had not been a policeman. Nonetheless, I accept the conclusion of the court below that ‘this played a role in her decision to accept [assistance], in the desperate circumstances in which she found herself’.

[15] Vicarious liability has a long but uncertain pedigree.³ In essence it may be described as the liability that one person incurs for a delict that is committed by another, by virtue of the relationship that exists between them. There are two features of vicarious liability in its traditional form that are trite but they bear repetition. The first is that vicarious liability arises by reason of a relationship between the parties and no more – it calls for no duty to be owed by the person who is sought to be held liable nor for fault on his or her part.⁴ The second feature is that it is a secondary liability – it arises only if there is a wrongdoer who is primarily liable for the particular act or omission. I will return to those features presently.

[16] Vicarious liability might arise from various relationships but we are concerned only with the relationship of employment. The circumstances in which vicarious liability will arise were described by this court in Minister of Law and Order v Ngobo⁵ – with reference to two cases that it described as ‘lodestars in this firmament’ (Mkize v Martens⁶ and Estate Van der Byl v Swanepoel⁷) – as follows:

‘The critical consideration is … whether the wrongdoer was engaged in the affairs or business of his employer. (I shall refer to it as the “standard test” or “general principle”.) It has been consistently recognised and applied, though – since it lacks exactitude – with difficulty when the facts are close to the borderline’.⁸

[17] The question whether the employee was ‘engaged in the affairs or business of his employer’ – sometimes expressed as whether the employee was acting ‘in the course’ or ‘within the scope’ of his or her employment – at the time the delict was committed is often problematic. At one extreme the delict might be committed by the employee while going about his or her employment in the ordinary way – in which case the employer will be liable. At the other extreme the delict might be committed by a person who, albeit that he or she is an employee, is going about his or her own private business, unconnected to that of the employer – in which case the employer is not liable. Between those extremes is ‘an uncertain and wavering line’.⁹

[18] Most of the decided cases fall somewhere between those extremes. They are mainly cases in which the employee, starting out on the business of the employer, then deviated from the employer’s business to attend to business of his or her own. An example is the often-cited case of Feldman (Pty) Ltd v Mall,¹⁰ in which a delivery driver deviated from his ordinary route to conduct business of his own, in the course of which he negligently killed the respondent’s husband. Notwithstanding that he was not strictly about the employer’s business when the delict was committed a majority¹¹ held the employer liable. Elaborating upon whether the employee could be said to have been engaged upon the business of his employer Watermeyer CJ said:

‘If the servant’s abandonment of his master’s work amounts to mismanagement of it or negligence in its performance and is, in itself, the cause of harm to third parties, then the master will naturally be legally responsible for that harm . . .. If, on the other hand, the harm to a third party is not caused by the servant’s abandonment of his master’s work but by his activities in his own affairs, unconnected with those of his master, then the master will not be responsible’.¹²

Tindall JA expressed it as follows:

‘In my view the test to be applied is whether the circumstances of the particular case show that the servant’s digression is so great in respect of space and time that it cannot reasonably be held that he is still exercising the functions to which he was appointed; if this is the case the master is not liable. It seems to me not practical to formulate the test in more precise terms; I can see no escape from the conclusion that ultimately the question resolves itself into one of degree and in each particular case a matter of degree will determine whether the servant can be said to have ceased to exercise the functions to which he was appointed.’¹³

[19] Minister of Police v Rabie¹⁴ (decided before Ngobo) was rather different. It concerned a policeman who was not on duty but he was nonetheless exercising police powers when he unlawfully arrested and assaulted Mr Rabie. A majority nonetheless found the state vicariously liable for his conduct. Jansen JA (writing for the majority) said that there were two stages of the enquiry:¹⁵

‘It seems clear that an act done by a servant solely for his own interests and purposes, although occasioned by his employment, may fall outside the course or scope of his employment, and that in deciding whether an act of the servant does so fall some reference is to be made to the servant's intention . . .. The test is in this regard subjective. On the other hand, if there is nevertheless a sufficiently close link between the servant’s acts for his own interests and purposes and the business of his master, the master may yet be liable. This is an objective test. And it may be useful to add that according to the Salmond test (cited by Greenberg JA in Feldman (Pty) Ltd v Mall 1945 AD 733 at 774):

“a master … is liable even for acts which he has not authorized provided that they are so connected with acts which he has authorized that they might rightly be regarded as modes – although improper modes – of doing them ....”’

[20] It is commonly accepted that at least one rationale for the existence of the rule of vicarious liability is that the employer creates the risk of harm and should thus be liable when the harm occurs.¹⁶ Having reiterated what is in effect the standard test later referred to in Ngobo the learned judge in Rabie went on to say ‘[our] leading cases mostly deal with deviations by the servant from his duties at a time he is actively engaged on his master’s work,¹⁷ and that the tests applied in such cases were not wholly apposite to the case before him. He went on to describe the enquiry that was called for in such a case:

‘In my view a more apposite approach to the present case would proceed from the basis for vicarious liability mentioned by Watermeyer CJ in Feldman (Pty) Ltd v Mall (at 741):

“… a master who does his work by the hand of a servant creates a risk of harm to others if the servant should prove to be negligent or inefficient or untrustworthy; that, because he has created this risk for his own ends he is under a duty to ensure that no one is injured by the servant’s improper conduct or negligence in carrying on his work …”

By approaching the problem whether [the policeman’s] acts were done “within the course or scope of his employment” from the angle of creation of risk, the emphasis is shifted from the precise nature of his intention and the precise nature of the link between his acts and police work, to the dominant question whether those acts fall within the risk created by the state.’

[21] Ngobo rejected that part of the reasoning in Rabie as clearly wrong. Kumleben JA pointed to the distinction between the rationale for a rule and the content of the rule itself, citing what was said by Schreiner JA in Carter & Co (Pty) Ltd v McDonald:¹⁸

‘It is often useful to examine the reason which probably gave rise to the rule, in order to discover the rule’s limits, but the reason, even if certainly established, is not the same as the rule.’¹⁹

Kumleben JA went on to say that

‘. . . whatever direct liability may in certain circumstances attach to an employer as a result of a risk created by him, this consideration in my opinion is not a relevant one to be taken into account when the standard test is to be applied in order to decide whether the master is vicariously liable.’²⁰

And later, after referring to a note by Professor Van der Walt²¹ dealing with the subject:

‘The writer proceeds to contend that [creation of risk] is the justification for State liability as a result of unlawful police conduct and thus for the decision in the Rabie case. This conclusion is reached, one should stress, not on the basis of vicarious liability, which is the ground of liability pleaded in the present case, but as an independent source of State liability.’²²

[22] As for the suggestion that the test that had been applied in deviation cases was not apposite Kumleben JA said the following:

‘If the standard test is to be accepted as the appropriate one for cases in which at the relevant time the servant had deviated from the course of his regular employment, it follows, in my view, that this test applies ad eundem – indeed more pertinently – where the servant cannot be said to have deviated for the reason that he was not even remotely engaged in his master's affairs at any relevant stage prior to the commission of the delict . . .’²³

[23] But while vicarious liability has traditionally been founded upon no more than the existence of the employment relationship – thus directing the enquiry only to whether the wrongdoer was engaged in the affairs or business of his or her employer when the delictual act was committed – recent cases in Canada and England reflect a principial shift by introducing into the enquiry duties on the part of the employer. Cases in which that has occurred all concern intentional acts of employees – which are usually difficult to conceive as having been committed within the course of the wrongdoer’s employment (or, on the ‘Salmond test’ referred to earlier,²⁴ as being ‘improper modes’ of doing an act authorised by the employer).

[24] The three principal cases in that regard were Bazley v Curry²⁵ and Jacobi v Griffiths²⁶ (both in the Supreme Court of Canada) and Lister v Hesley Hall Ltd²⁷ (in the House of Lords). They were all cases in which an employee of an institution of one kind or another sexually abused children who were in his care.

[25] In Bazley a foundation that operated residential care facilities for the treatment of emotionally troubled children was held vicariously liable for the sexual abuse by an employee of children who were under its care. After reviewing decided cases McLachlin J said:²⁸

‘Underlying the cases holding employers vicariously liable are the unauthorized acts of employees is the idea that employers may justly be held liable where the act falls within the ambit of the risk that the employer’s enterprise creates or exacerbates’.

The learned judge concluded:²⁹

‘In summary, the test for vicarious liability for an employee’s sexual abuse of a client should focus on whether the employer’s enterprise and empowerment of the employee materially increases the risk of the sexual assault and hence the harm. The test must not be applied mechanically, but with a sensitive view to the policy considerations that justify the imposition of vicarious liability – fair and efficient compensation for wrong and deterrence. This requires trial judges to investigate the employee’s specific duties and determine whether they gave rise to special opportunities for wrongdoing. Because of the peculiar exercises of power and trust that pervade cases such as child abuse, special attention should be paid to the existence of a power or dependency relationship, which on its own often creates a considerable risk of wrongdoing’.

[26] Jacobi, on the other hand, concerned an employee of a children’s club, who sexually assaulted children of the club, but at the employee’s home and outside working hours. A majority held the club not to be vicariously liable. Following the ‘creation of risk’ approach adopted in Bazley Binnie J pointed out that to justify imposing liability on that basis there must be a strong connection between the created risk and the wrongful act, which was held to be absent.³⁰ ‘To find a strong connection’, the learned judge said, ‘there must be a material increase in the risk of harm occurring in the sense that the employment significantly contributed to the occurrence of the harm’.³¹

[27] Lister was decided along similar lines. In that case the owner and manager of a boarding house attached to a school was found vicariously liable for systematic sexual assaults upon children resident in the boarding house committed by the warden. Lord Steyn articulated the approach to be taken to vicarious liability and then said the following:³²

‘If this approach to the nature of employment is adopted, it is not necessary to ask the simplistic question whether in the cases under consideration the acts of sexual abuse were modes of doing authorised acts. It becomes possible to consider the question of vicarious liability on the basis that the employer undertook to care for the boys through the services of the warden and that there is a very close connection between the torts of the warden and his employment. After all, they were committed in the time and on the premises of the employers while the warden was also busy caring for the children’.

[28] As Watermeyer CJ said in Feldman,³³ the rationale underlying the imposition of liability for ‘risk creation’ is that by creating the risk of harm the employer has a duty to ensure that the harm does not eventuate. Because he has created the risk for his own ends

‘he is under a duty to ensure that no one is injured by the servant’s improper conduct or negligence in carrying on his work and that the mere giving by him of directions or orders to his servant is not a sufficient performance of that duty’.

[29] While ‘risk creation’ might indeed be capable of giving rise to liability on the part of the employer, it was said in the passage from Ngobo to which I referred earlier,³⁴ that the true basis for liability in such cases is the failure of the employer, acting through the instrument of the employee, to fulfil the duty that is cast upon the employer to avoid harm occurring through the risk that has been created. For on the traditional approach vicarious liability arises from the existence of the relationship alone and not from any failure of duty by the employer. But adopting the approach in those cases I cannot see that any material risk was created by the state in this case – and least of all that there was a ‘strong connection’ between the delict and such risk as might have been created.

[30] But the introduction into the principle of vicarious liability of a duty owed by the employer was taken a step further in K. Adopting the ‘two-stage’ enquiry that was laid down in Rabie, O’Regan J said that the first question ‘requires a subjective consideration of the employee’s state of mind and is a purely factual question’.³⁵ Needless to say, the state of mind of the policemen in that case, as it was in this case, was entirely self-directed, and the case turned rather on the second ‘objective’ question, namely, ‘whether, even though the acts done have been done solely for the purpose of the employee, there is nevertheless a sufficiently close link between the employee’s acts for his own interests and the purposes and the business of the employer’. That question, said the learned judge

‘does not raise purely factual questions, but mixed questions of fact and law. The questions of law it raises relate to what is ‘sufficiently close’ to give rise to vicarious liability. It is in answering this question that a court should consider the need to give effect to the spirit, purport and objects of the Bill of Rights.’³⁶

[31] The learned judge pointed out that the rape of K was ‘clearly a deviation from their duties’ but went on to observe that when committing the rape the three policemen were ‘simultaneously omitting to perform their duties as policemen’, which was said to be ‘relevant to answering the . . . question . . . was there a sufficiently close connection between that delict and the purposes and business of the employer’. She then listed three considerations³⁷ that had founded her conclusion.³⁸ First, that

‘the policemen all bore a statutory and constitutional duty to prevent crime and protect the members of the public. That duty is a duty which also rests on their employer and they were employed by their employer to perform that obligation’.

Secondly

‘in addition to the general duty to protect the public, the police here had offered to assist the applicant and she had accepted their offer’.

And thirdly,

‘the conduct of the policemen which caused harm constituted a simultaneous commission and omission. … Their simultaneous omission lay in their failing while on duty to protect her from harm, something which they bore a general duty to do, and a special duty on the facts of this case’.

The learned judge then concluded:

‘In my view, these three inter-related factors make it plain that viewed against the background of our Constitution, and, in particular, the constitutional rights of the applicant and the constitutional obligations of the respondent, the connection between the conduct of the policemen and their employment was sufficiently close to render the respondent liable.’

[32] There are three observations to make in relation to those findings. The first is that the court found that both the state, and the policemen personally, were under a duty to protect K, and that they omitted to fulfil those duties. It follows that in acting as they did the policemen committed two separate delicts – one was their positive delictual act of assaulting K, and the other was their delictual omission in failing to protect her. The second observation is that I think the inference is clear from the three reasons that were advanced by the learned judge that the delict for which the state was held liable was not the positive acts of the policemen – for otherwise their omissions would have been immaterial – but instead their delictual omissions. And the third observation is that the conclusion in that case was expressly founded upon vicarious liability for the delicts of the policemen³⁹ – and not upon direct liability of the state – from which it follows that the policemen must have been considered to be personally liable for their omissions (for otherwise there would have been no scope for vicarious liability).

[33] In his insightful commentary on the case Stephen Wagener⁴⁰ has expressed the view that the court ‘confused personal and vicarious liability’ and that ‘[a] breach of an employer’s duties, in this case the state’s alleged constitutional ones, can only affect its personal liability’. Later he says that ‘logically, the Bill of Rights can only affect personal liability outcomes, and is irrelevant to vicarious liability’.⁴¹

[34] I think that criticism is only partly correct. I have pointed out that both the state, and the policemen personally, were held to be under a duty to protect K. In those circumstances it might be that the court could justifiably have found that the state, acting through its employees, was directly liable for its own delictual omission. That would have been consistent with a line of cases that have been decided in this court (Minister of Safety and Security v Van Duivenboden,⁴² Van Eeden v Minister of Safety and Security,⁴³ Minister of Safety and Security v Hamilton,⁴⁴ and Minister of Safety and Security v Carmichele⁴⁵) that purport to be founded upon vicarious liability, but might better be said to have been founded upon direct liability of the state, acting through the instrument of its employees.⁴⁶ In an illuminating article François du Bois⁴⁷ construes those cases as reflecting

‘[a] subtle but vital shift . . . in which state liability is no longer viewed in terms of the traditional vicarious liability paradigm of the common law model, but rather, á la civilian systems, as a form of direct liability arising from an organizational failure or faute de service.’

He describes them as representing ‘a further rupture in historical link to the common law tradition with its failure to develop a distinctive notion of the state and its mission’. Elaborating he says that they concern themselves

‘with the duties of the state and the need for holding government accountable for breaches rather than with the question whether the individual official has committed the delict’.

[35] But I have pointed out that the court found that not only the state, but also the policemen personally, were under a duty that they omitted to fulfil. In view of its express finding of vicarious liability it must be taken that the policemen were considered to be personally liable for their omissions, thus rendering the state vicariously liable, albeit that it might equally have held the state to be directly liable for its own omission. In this case, too, we are concerned with whether the state is vicariously liable for delictual conduct on the part of Mr Van Wyk, and not with whether the state is directly liable.

[36] Clearly Mr Van Wyk is liable for the consequences of his positive delictual acts (as the policemen were liable for the consequences of their positive acts in K). But as I have already observed I think the inference from the findings in K is that the state was considered not to be vicariously liable for the positive delictual acts of the policemen in that case – for if it was vicariously liable for those positive acts then the omissions would have been immaterial.

[37] If the state was not vicariously liable for the positive delicts of the policemen in K then I think that, a fortiori, it is not vicariously liable for the positive delict that was committed by Mr Van Wyk. Indeed, it would seem to me to be rather extreme to find that a policeman is ‘engaged in the affairs or business of his employer’⁴⁸ when he commits the crime of rape, or that that could ‘rightly be regarded as a mode – although an improper mode’ of exercising the authorisation conferred by his employment.⁴⁹ In the words of Kumleben JA in Ngobo, Mr Van Wyk ‘cannot be said to have deviated [from his employer’s business] for the reason that he was not even remotely engaged in his master's affairs at any relevant stage prior to the commission of the delict.’⁵⁰ Or as Watermeyer CJ would have said it,⁵¹ the harm was ‘not caused by the servant’s abandonment of his master’s work but by his activities in his own affairs, unconnected with those of his master’. Or in the words of Tindall JA,⁵² ‘his digression from the business of his employer was so great in respect of space and time that it cannot reasonably be held that he [was] still exercising the functions to which he was appointed.’

[38] Three factors that were considered by the court below to be indicative of vicarious liability – seemingly for that positive act (the other considerations taken account of, which I will come to, were directed to the alleged omission) were, first, that Mr Van Wyk was in possession of a police vehicle, which, the court said, provided him with the means to commit the offences (this was said by the court below to be the ‘single most important connection’). Secondly, that the discovery by Ms F that Mr Van Wyk was a police officer ‘to some extent . . . operated to lull her suspicions’. And the third consideration was what the court described as ‘the coincidence between the nature of the assistance which Van Wyk pretended to offer . . . and the normal duty of a police official’.

[39] That an employee uses property of the employer might in some cases indicate that he or she is about the business of the employer but I think that is no more than an evidentiary factor. In this case it is clear that Mr Van Wyk was not engaged on police business at the relevant time and I do not see that his unauthorised use of his employer’s property is then relevant.⁵³ I might add that in Ngobo the delict was committed directly by the use of a service firearm yet that was not considered to be significant. That Ms F might have relied upon the fact that Mr Van Wyk was a police officer also does not seem to me to assist: I do not think the question whether an employee is about the business of an employer can be dependant upon whether the victim knows or does not know that he or she is an employee. And the fact that Mr Van Wyk’s offer to drive her home coincided with what might be expected from a police officer also does not seem to me to take the matter further when in fact the offer was not made in fulfilment of police functions.

[40] I might add that all those factors – in accentuated form – were also present in K yet they were not sufficient to expose the state to liability for the positive acts of the policemen concerned. In that case the vehicle, which announced itself openly to be a police vehicle, was being used by the policemen to go about their official business, from which they then diverted. In that case K knew full well that they were policemen and that they were on police duty when she accepted their offer. And in that case the assistance that they offered did not merely coincide with what could be expected of a policeman – they were doing precisely what their police duties required.⁵⁴

[41] Turning to the omissions that were found to exist in K I have no doubt that the state had a duty to protect Ms F against harm and that that duty necessarily falls upon functionaries who execute the duties of the state – which might render the state directly liable if its functionaries omit to do so. But the basis for the finding in K was that the policemen were also under an equivalent personal duty – thus rendering them personally liable (and the state vicariously liable) – for omitting to fulfil that duty. It seems to me, then, that what this case comes down to is whether Mr Van Wyk was under a similar duty at the time he committed his criminal act. And that depends upon whether the duties that were held to exist in K persist when a police officer is not on duty.

[42] There is some suggestion in the judgment of the court below that a police officer is never off duty – that his or her obligations are of a ‘continuing nature’ – and that was said to be supported by the decision of the trial court in Rabie.⁵⁵ It was also suggested that a police officer who is on ‘standby duty’ is not off duty at all but is on duty in an attenuated form. The court below expressed that as follows:

‘In my view it would be mistaken to see only a sharp distinction between being on and off duty and then to treat Van Wyk as being off duty. His status as being on stand-by at the material time fell rather somewhere between these states.’

Various shades of those suggestions also surfaced in argument before us. I do not think they are correct.

[43] Neither in the evidence of Mr Du Toit nor in the standing orders do I see the presence of any of the characteristics of employment⁵⁶ while a detective is between shifts. Until such time as a detective is called upon to resume duty he or she is not subject to the control or direction of his or her employer – the detective is free to engage at will in hobbies or to go about his or her personal affairs. The only obligation is to resume duty when called upon to do so – thereby reverting to the control and direction of the employer – just as the detective is obliged to do so when the next shift commences. That the scope of his or her private activities are attenuated during that time is no more than a concomitant of the uncertainty that exists as to whether he or she might be required to resume duty. The detective may not drink alcohol, for example, not so as to be sober while going about his or her personal business, but so as to be sober if called upon to resume duty, just as a detective must not drink at 7.00 in the morning so as to be sober when resuming duty at 7.30. In my view ‘standby duty’ as described in the evidence is precisely what the language conveys – a detective is on standby to resume duty if duty calls and is off duty until that occurs.

[44] I also see no basis for finding that the obligations of a police officer are of a ‘continuing nature’. What does continue when a police officer goes off duty is his or her authority to exercise police powers. The Criminal Procedure Act 51 of 1977, for example, confers upon peace officers, by virtue of their appointment alone, the authority to arrest without warrant, whether the police officer is on or off duty. That police officers are entitled to exercise police powers when they are off duty does not imply that they are obliged to do so.

[45] In his evidence Mr Du Toit said that even when not on duty a police officer must intervene when he or she encounters a crime that is being committed and ‘automatically puts himself on duty’ when he or she does so. I have no doubt that many police officers consider it their duty to exercise their police powers whenever they encounter crime and that is meritorious. But we are concerned in this case not with what is meritorious but instead with whether they are legally obliged to do so. I can see no grounds for finding that a police officer is obliged to perform his or her ordinary functions when not on duty. Indeed, I think that the consequences of such a finding would be far-reaching indeed. Its effect would be to make the state a guarantor of good behavior on the part of police officers at all times by virtue alone of their appointment. Without a duty to protect Ms F against harm – and thus personal liability for omitting to do so – there is no scope for secondary liability of the state for the omission to protect Ms F. And while a police officer who chooses to exercise police powers might render the state vicariously liable for delicts committed in the course of doing so – which was the case in Rabie and in Minister of Safety and Security v Luiters⁵⁷ – Mr Van Wyk did not purport to be exercising police powers in this case.

[46] There was one further factor that was considered by the court below to strengthen its conclusion. Applying what it described as the ‘contentious “creation of harm” approach articulated by Jansen JA’ in Rabie, the court concluded that by retaining Mr Van Wyk in its employ notwithstanding that he had criminal convictions⁵⁸ the state ‘accepted the risk that his propensity for criminal conduct might continue and cause harm to others’.

[47] It might not have been brought to the attention of the learned judge that Rabie was later reversed by Ngobo on that point (although the idea is far from dead, as appears from the foreign cases I have referred to⁵⁹). But I think that the view expressed by the court below in that regard draws attention to the pitfalls of introducing concepts of duty into the doctrine of vicarious liability. For the question that it inevitably raises – as it does on the view taken by the learned judge – is whether the breach of duty is causally connected to the delict – an enquiry that does not fit easily into vicarious liability. If the state ought indeed not to have retained Mr Van Wyk in its employ because he had been convicted of crimes, and its breach of its duty in that regard was causally connected to the offence, then I think that might render the state directly liable for breach of its own duty. But I cannot see how it is material to determining whether it is vicariously liable.

[48] In my view this case fails the test for vicarious liability that was articulated in K. It seems to me that upon proper analysis the finding of liability in that case was founded upon the personal liability of the policemen concerned – and consequently the vicarious liability of the state – for omitting to fulfil their constitutional and statutory police duty and Mr Van Wyk had no such police duty in this case. (I think it is important to add that because Mr Van Wyk was not purporting to act as an instrument of the state at the time he committed his crime the state could equally not have been found directly liable had the case been advanced along those lines.)

[49] For those reasons I would uphold the appeal. Because this case raises an important principle of broader significance for the state I do not think that the respondent should pay the costs in this court or in the court below. The orders of the court below so far as they relate to the appellant (first defendant in the court below) are set aside and the following is substituted:

‘The claim against the first defendant is dismissed’.

_____________________

R W NUGENT

JUDGE OF APPEAL

MAYA JA (BOSIELO JA concurring)

[50] I am indebted to my colleague, Nugent JA, for the benefit of his judgment. After much thought and vacillation, I find myself respectfully unable to agree with his conclusion. The salient facts are set out fully in his judgment and they need not be dealt with in any great detail.

[51] As I see the case, the fundamental issue is whether the Minister of Safety and Security is vicariously liable for delictual damages flowing from a rape committed by a police officer whilst on standby duty.

[52] The court below found that Mr Allister Van Wyk, who was a police detective in the South African Police Service at the time (he was dismissed from the police force after his conviction), had acted in his capacity as a police officer when he committed the assault and rape and consequently held the Minister vicariously liable for his conduct. In reaching this conclusion, the court below, inter alia, made strong credibility findings in favour of the victim, Miss F, and attributed the occasional vagueness in her evidence to the trauma of her brutal attack and her tender age (she was 13 years old) at the time of the incident.

[53] Van Wyk’s evidence was, however, found wanting. In the court’s view, it was ‘false and self-serving’ and Van Wyk’s seeming candidness regarding the rape act itself was, by his own admission, born out of sheer necessity as he had already been convicted of the rape which he had denied at his trial. The court then preferred F’s version of the events where it conflicted with that given by Van Wyk. I have found no reason to depart from the court’s assessment of these witnesses and its reasoned findings in that regard. I would merely add that the hiatus of some 11 years between the rape incident and the criminal trial must have contributed to F’s uncertainty on some of the fine detail of the events.

[54] There was no dispute around the key facts ie that F was assaulted and raped at an isolated, dark spot around midnight by Van Wyk after she accepted a lift home from him at a nightclub. Van Wyk was on standby duty and wore plain clothes. He was driving an unmarked police vehicle, which he had been allocated for official duties, for his own private purposes when he met F and offered to drive her home.

[55] What appears from F’s testimony is that she initially agreed to travel in Van Wyk’s vehicle mainly because Mr Petrus Faniso, an acquaintance whose family lived in her block of flats, was a passenger in the vehicle. She readily recognised the unmarked vehicle even before accepting the lift at the nightclub as she stood next to it because it had a police radio attached to it which had its power light switched on. She became aware that Van Wyk was a police officer before accepting his second offer of a lift at Kaaimans River. This, she said, happened when she saw police dockets on the front passenger floor. She could not recall the exact stage in the chain of events but, taking Van Wyk’s evidence in this regard into account, this must have occurred when she took the front seat and he put the dockets on the floor at the back after they left Faniso at his home. When she asked Van Wyk why the police dockets which were marked ‘Sgt Van Wyk’ on their covers were in the vehicle, he told her that he was a private detective and she understood this to mean that he was a police officer.

[56] Van Wyk’s superior at the material time, Mr Du Toit, formerly the Provincial Commander in charge of General Investigations in the Western Cape, testified on the Minister’s behalf. He explained the nature of ‘standby duty’ as follows. A police officer on the ‘standby duty’ roster was required, in terms of the relevant standing order of the police⁶⁰ to be ‘available at his dwelling’ after the normal 7h30 – 16h00 shift to resume duty at short notice when so required by the police service centre. (In practice, however, a normal shift for a detective on official duty or standby duty extends much later than 16h00 according to Van Wyk who worked until 20h00 on the fateful day and Mr Johan van Dyk, another former member of the Investigations Unit at George who testified on F’s behalf.)

[57] The officer would be paid an allowance as compensation for the restriction on his movements and would be allocated a State vehicle, for which he signed, but only for the purposes of enabling him to resume official duty. A police officer who resumed duty during the standby period, whether by reason of being called by the service centre or placing himself on duty, was required to record the nature of the official duties performed in his pocketbook as proof that he had done the work.

[58] Du Toit explained that because detectives are experienced police officers they are trusted to be able to work individually ie not under a superior’s direct control and can place themselves on and off duty as they see necessary as long as this is recorded in the officer’s pocketbook. According to police procedure, an officer who is not on duty places himself on duty automatically by intervening in the commission of a crime or assuming police duties and whenever a police officer performs police duties he is on duty.

[59] Van Wyk, who testified, confirmed these police procedures. In cross-examination, he was asked why F got back in his vehicle on the second occasion and his reply was

‘It was late, it was dark. She was alone by the side of the road … Because I told her I would take her home.’

He said that the dockets in his vehicle concerned matters reported earlier that day which he was investigating and that he took them with him to prepare for inspection on the following day. After raping F he completed the vehicle’s logbook as required by procedure and falsely ascribed the mileage he had accumulated during his illicit jaunt to investigative duties he performed when he started his shift at 7.30 on 14 October until the end of the standby shift at 7.30 of 15 October. He also recorded the case numbers of the ‘three dockets’ as the cases he had been working on during the standby period and fuelled the vehicle at a police filling station in the amount of R45, which was subsequently paid by the State. He stressed that his objective when he gave F a lift was to rape and not to help her. But he conceded that he would have been obliged as a police officer to take her home had he been on duty and would in fact have done so.

[60] This is the essence of the evidence on which the court below determined the issue of the Minister’s liability.

[61] The legal principles underpinning vicarious liability of an employer, including the State, for the delict of an employee are firmly established. Put simply, vicarious liability arises for the employer if the delict is committed by the employee whilst acting in the course and scope of his or her employment or, in other words, whilst engaged in the affairs or business of his or her employer.⁶¹

[62] The foundation of the rule is public policy which this court, in Feldman (Pty) Ltd v Mall,⁶² said was based on the fact that because an employer’s work is done ‘by the hand’ of an employee, the employer creates a risk of harm to others should the employee prove to be negligent or inefficient or untrustworthy. (The ‘risk of harm’ referred to relates only to the reason underlying the rule and not its content; it is not an element of the rule. ) This places an employer under a duty to ensure that no one is injured as a result of the employee’s improper conduct or negligence in carrying on his work.

[63] Our courts have for many years found applying this rule a complex task as a result of difficult questions of fact which often arise especially in ‘deviation’ cases ie where a rogue employee in pursuit of his own interests commits a delict whilst ostensibly engaged in the affairs of his employer.⁶³ And determining whether the State should take responsibility for the misdeeds of police has proven particularly vexing.

[64] Drawing from the established common law principles with regard to vicarious liability as interpreted and applied by our courts in cases such as Feldman⁶⁴ and Minister of Police v Rabie,⁶⁵ the Constitutional Court has formulated the questions to be asked in applying the rule in deviation cases as follows in K v Minister of Safety and Security:

‘The approach [adopted in Minister of Police v Rabie] makes it clear that there are two questions to be asked. The first is whether the wrongful acts were done solely for the purposes of the employee. This question requires a subjective consideration of the employee’s state of mind and is purely a factual question. Even if it is answered in the affirmative, however, the employer may nevertheless be liable vicariously if the second question, an objective one, is answered affirmatively. That question is whether, even though the acts done have been done solely for the purpose of the employee, there is nevertheless a sufficiently close link between the employee’s acts for his own interests and the purposes and the business of the employer. This question does not raise purely factual questions, but mixed questions of fact and law. The questions of law it raises relate to what is ‘sufficiently close’ to give rise to vicarious liability. It is in answering this question that a court should consider the need to give effect to the spirit, purport and objects of the Bill of Rights.’⁶⁶

(The court acknowledged the subsequent criticism levelled against the majority’s adoption of ‘creation of risk’ as a factor relevant to the determination of vicarious liability in Rabie but pointed out that its statement of the standard test was not directly criticised.)

[65] The court below found that the Minister was vicariously liable mainly because (a) Van Wyk drove a police vehicle officially assigned to him and used it to commit the rape, (b) F formed a belief that he was a police officer before accepting his second offer of a lift and this lulled her suspicions and (c) the nature of the assistance Van Wyk pretended to offer coincided with the normal duty of a police officer to protect members of the public even when off-duty. The court also found it relevant for the enquiry that Van Wyk was on standby duty, which it termed an ‘attenuated form of duty’ falling ‘rather somewhere between’ being on and off-duty.

[66] I must say at the outset that I do not understand our law to impose a duty on police officers to protect members of the public and prevent crime even when not on official duty. And our common law has not been developed to the extent that the State is answerable for delicts committed by off-duty police officers. Neither do I attach any significance to the fact that Van Wyk was on standby duty. In my view, he effectively remained off duty until summoned to resume duty or took action that placed him on duty as explained by Du Toit. It is well to bear in mind in the latter regard that for purposes of vicarious liability, there is no distinction between on-duty police officers and those who are off-duty but place themselves on official duty; they are in the same legal position.⁶⁷

[67] Regarding the first stage of the test set out in K, there is no question that Van Wyk’s conduct in raping F had nothing whatsoever to do with the performance of his duties as a police officer. But the enquiry goes further. It must still be considered whether despite the fact that Van Wyk’s unlawful conduct was totally self-serving, there nevertheless was a sufficiently close link between his acts for his own personal gratification and the State’s business.

[68] Contrary to the finding of the court below in this regard, I do not accept that Van Wyk attempted to hide the fact that he was a police officer. He did not conceal the radio in the vehicle, which, according to F, was switched on and which he conceded could be recognised as a police radio by any person near the vehicle. This piece of equipment clearly marked the vehicle as belonging to the police force to anyone who cared to notice. That he had removed the police radio aerial from the vehicle, as he testified, was not done in an effort to hide the vehicle’s identity. He said that cars were frequently burgled at the nightclub premises and that he removed the aerial only to prevent it from being stolen as it was merely screwed on the vehicle and would be easy to steal.

[69] More importantly, he told F that he was a detective, albeit a private one. He gave that explanation as a reason for the presence in his vehicle of police dockets which were casually placed in the vehicle, in clear view of passengers, and conspicuously bore the rank and name of a police officer (this is the very information F subsequently gave the police which enabled them to trace and promptly apprehend him). This conduct, in my view, cannot be reconciled with an intention to hide his official identity.

[70] Unlike my colleague Nugent JA, I find it quite pertinent that F was aware that Van Wyk was a police officer when she accepted Van Wyk’s second bogus offer to take her home at Kaaimans and I agree with the court below that this knowledge influenced her decision and quelled her earlier misgivings. This state of mind is plain from her words

‘I found out [that he was a policeman] during the evening before the rape …He told me …when I asked him why the [police dockets] were in the car … I went with him because he said he was a detective because I trusted him because he was a detective … I thought he was a policeman…And the police radio that was in the car.’⁶⁸

[71] As I understand the court’s reasoning in K, it was a relevant factor that the victim there identified the policemen by their uniform and the marked police van they drove and for that reason, placed her trust in them and accepted their lift even though she did not know them. The court said:⁶⁹

‘[I]n addition to the general duty to protect the public, the police here had offered to assist the applicant and she had accepted their offer. In doing so, she placed her trust in the policeman although she did not know them personally. One of the purposes of wearing uniforms is to make police officers more identifiable to members of the public who find themselves in need of assistance.’

At para 57 of the same judgment, the court continued:

‘In sum, the opportunity to commit the crime would not have arisen but for the trust which [K] placed in them because they were policemen, a trust which harmonises with the constitutional mandate of the police and the need to ensure that mandate is successfully fulfilled.’

[72] It seems to me that if the purpose of a police uniform is to enable members of the public who need assistance to readily identify the police, then, by parity of reasoning, the same must apply to a marked police vehicle. But in this case, it did not matter that van Wyk wore plain clothes and drove a vehicle which did not bear police insignia. He practically identified himself as a police officer and F placed her trust in him for that reason; a result which he most likely anticipated.

[73] Thus, by offering to rescue and take home in a police vehicle a lone, vulnerable child stranded on a dark, deserted riverside in the dead of night in those circumstances, Van Wyk subjectively placed himself on duty and acted in his capacity as a police officer. This is regardless of his intention which I find no different from that of the errant, off-duty, officer in Rabie who, actuated purely by malice, arrested a person he very well knew was innocent. As the learned judge below pointed out and Van Wyk himself conceded as previously stated, such conduct is precisely what would be expected of a police officer in that kind of situation. In my opinion, he placed himself on duty as he was empowered to do by law. And once he did, he assumed the status and obligations of an on-duty police officer. For that reason, I would find the Minister vicariously liable.

[74] Finally, there is another matter that, in my view, requires comment. It transpired in the trial that Van Wyk has a criminal record. He was initially convicted of assault with intent to do grievous bodily harm. Thereafter, he was convicted of the negligent discharge of a firearm whilst under the influence of alcohol or narcotic drugs. A further conviction for common assault followed. All these convictions occurred during his career in the police force and they do not seem to have hindered his rise within the police ranks as he became a detective vested with vast police powers and limited control by his superiors. More disconcerting was a statement made by Du Toit when asked if he was surprised that an officer with Van Wyk’s sullied track record was kept in the police force, that ‘there are guys that have done worse things that have stayed on in the [police] service’ and that it is only when a police officer has been declared unfit to carry a firearm that he will not be put on operational duties.

[75] I do not agree with the erroneous elevation by the court below of the apparent risk of harm created by the Minister, in keeping a police officer of Van Wyk’s questionable calibre in the police force, to an element of the vicarious liability rule. But I share the sentiments the court below expressed that it should not be a matter for surprise for the Minister that Van Wyk, who had no vehicle of his own and would most probably not have committed the rape had he not been provided with a vehicle as he would have been unable to give F a lift, acted as he did.

[76] In K,⁷⁰ the Constitutional Court exhorted our courts to ‘take account of the importance of the constitutional role entrusted to the police and the importance of nurturing the confidence and trust of the community in the police in order to ensure that their role is successfully performed’. In Luiters,⁷¹ the court reiterated the Minister’s responsibility ‘to ensure that police officers are properly trained and carefully screened to avoid the risk that they will behave in a completely improper manner’. I find it inimical to these objectives that our police force would, seemingly as a matter of course, have within its ranks police officers who have repeatedly committed serious crimes.

[77] I would accordingly dismiss the appeal.

__________________

MML MAYA

JUDGE OF APPEAL

APPEARANCES:

For appellant: R T Williams SC

J van der Schyff

Instructed by:

The State Attorney, Cape Town

The State Attorney, Bloemfontein

For respondent: L M Olivier

Instructed by:

De Klerk & Van Gend, Cape Town

Symington de Kok, Bloemfontein