2014 3 SA 394 (SCA)

 TJ Scott
  University of Pretoria

 47 Volume 2 2014 pp 374
  Download Case Note in PDF
  Read the judgment of the court on SAFLII

Determination of constitutional nature of contractual and delictual claims - strict contractual liability of security company - vicarious liability of security company for wrongful and negligent conduct of employee

1 Introduction

The case under discussion is of general interest to those interested in the law of contract and delict, and of particular interest to practitioners in the private security industry field. It also provides a sterling example of the risks involved in modern litigation in South African courts: Here the plaintiffs who had succeeded fairly easily in the High Court and subsequently had to bear the disappointment of seeing their judgment overturned on appeal by a majority of the Supreme Court of Appeal, finally won the day when the Constitutional Court granted them leave to appeal and finally upheld their appeal against the judgment of the Supreme Court of Appeal, in effect reinstating the judgment of the High Court.

In this case note the focus will fall on certain aspects of the law pertaining to breach of contract as well as on the application of the principles of the law of delict dealing with wrongfulness and negligence in the context of causing pure economic loss. The set of facts that confronted the court would ordinarily be interpreted as an instance of pure economic loss, seeing that the actions of third parties (robbers) causing the damage in casu was made possible by the preceding conduct of one of the defendant’s employees, which conduct was therefore indirectly causally linked to the plaintiffs’ harm.

Van der Westhuizen J (with Moseneke ACJ, Skweyija ADCJ, Cameron J, Dambuza AJ, Froneman J, Jafta J, Madlanga J, Nkabinde J and Zondo J concurring) provided an interesting introduction to the judgment, highlighting the growing importance of the private security industry in contemporary South Africa. He referred to the high incidence of crime and provided startling recent statistics of serious offences such as murder and armed robbery and then remarked:

“The South African Police Service is not always perceived to be capable of meeting its constitutional mandate. Hence, the private security industry is a large and powerful feature of South Africa’s crime-control terrain. While it should and could not be a substitute for state services, it fulfils functions that once fell within the exclusive domain of the police” (398A).

The court attributed this tendency to historical factors, pointing out that it began developing since the late 1970s as one of the consequences of the apartheid regime’s focus on state security and political control, which caused a strain on the availability of members of the police to perform their ordinary protective duties. The court continued to sketch a scene where security officers in the private security industry nowadays even outnumber the members of the South African Police Service (398B-C). It is suggested that this phenomenon in particular should kindle the interest in this judgment of anybody involved or associated within the mechanisms - public as well as private - by means of which public safety and security are maintained in our country.

2 Facts and Judgment

The respondent, iMvula (defendant in the South Gauteng High Court and appellant in the Supreme Court of Appeal), is a private security company who was contracted to provide an armed guarding service for the protection of the home of the appellants, Mr and Mrs Loureiro and their two minor sons (plaintiffs in the South Gauteng High Court and respondents in the Supreme Court of Appeal), on a 24-hour basis seven days a week. In addition, Mr Loureiro had an extensive security system installed at the property (electric fences, perimeter beams, multiple alarm systems, a guard house with bulletproof windows, an intercom system and closed-circuit television). Access to the property could be gained through two entrances, namely an armoured pedestrian gate and a driveway gate. Both gates could be observed from the guard house.

In terms of their agreement iMvula initially undertook, inter alia, to take all reasonable steps to prevent persons gaining unauthorised access and/or entry to the premises, to take all reasonable steps to protect the appellants and their property and to take all reasonable steps to ensure that no persons gained unlawful access to the premises. (These terms were not put in writing and their content was reflected in clauses 6.5.1 - 6.5.2 and 6.5 7 of Mr Loureiro’s heads of argument presented to the High Court.) As a result of a lapse of security due to one of the guard’s allowing someone to enter the premises through the main gate without obtaining Mr Loureiro’s prior approval, the latter had the intercom system partially disabled to prevent the guards from opening and closing the main gate. This arrangement unfortunately affected the movement of the armed guards during their shift changes. To alleviate this problem Mr Loureiro furnished the guards with a key to the smaller pedestrian gate. This course of action was accompanied by an express prohibition to the effect that the key should not be used for any purpose other than to enable the guards to change shifts. The guard supervisor was emphatically instructed that the key should not be used to open that gate to allow anybody to gain access without Mr or Mrs Loureiro’s prior authorisation. This arrangement was accepted by iMvula and therefore constituted a further term of their oral agreement. (The full content of this further term to their oral agreement, reflected in clause 6.8 of Mr Loureiro’s heads of argument, was as follows: “[iMvula] was not entitled to permit any person to gain access to the [Loureiro family’s] residence other than [Mr and Mrs Loureiro] and their two minor sons, unless [iMvula] had obtained such prior authorisation from [Mr Loureiro] alternatively [Mrs Loureiro] to allow such person access” (399H-400A).) According to Mr Loureiro this term amended the initial terms alluded to earlier to the effect that it imposed strict liability on iMvula for harm pursuant to the latter’s failure to observe it. However, iMvula denied that it had incurred such strict liability.

About one month after Mr Loureiro had issued this prohibition, a robbery took place at the premises. The circumstances were as follows: In the early evening an unmarked white BMW sedan with a flashing blue light mounted to its dashboard drove into the driveway. A man wearing dark blue clothing, a reflective vest marked “Police” and a cap with a logo resembling a police logo stepped from the car, proceeded in the direction of the guard house and flashed an identity card in the direction of Mr Mahlangu (henceforth M), the guard on duty. M was unable to examine the card and tried in vain to speak to the stranger through the intercom system. Relevant to these circumstances was the fact that M, a qualified Grade-A security guard, had never been informed of the prohibition against allowing unauthorised strangers onto the premises, nor had he been issued with a detailed job description pertaining to the specific services required by the Loureiros. In addition he had received no proper instructions on how to deal with police officers demanding entry (viz in conformity with section 25(3) of the Criminal Procedure Act 51 of 1977), nor how to identify police officers. Finally, the unarmed M’s only means of contacting iMvula was his personal cellular phone which had no airtime at that crucial moment. Caught off guard by believing the man to be a police officer, M left the guard house, walked to the pedestrian gate and without attempting to speak to the visitor through the gate unlocked it without further ado. After unlocking and opening the gate, M was confronted by an armed robber pointing a firearm at him. Thereupon several accomplices joined the robber, entered the house and accosted the house staff and children. When Mr and Mrs Loureiro returned from a function about an hour later, they were also confronted and finally coerced to provide the robbers with the keys to their safes. The robbers thereafter fled with booty worth R11 million.

Subsequently the Loureiros claimed damages from iMvula in the High Court (Loureiro and Others v iMvula Quality Protection (Pty) Ltd [2011] ZAGPJHC 140 (GSJ)) on two grounds: Mr Loureiro alleged that his damage had been caused by iMvula’s breach of contract, whereas the claims of Mrs Loureiro and their two sons were brought in delict on the basis of iMvula’s alleged wrongful and negligent conduct which had been a contributing cause of their damage. The parties agreed to separate the liability and quantum issues and the latter issue therefore did not feature in the litigation at all. The Loureiro family was successful due to the fact that Satchwell J, who approached both types of claim from the angle of negligence, found iMvula to have been negligent. She held iMvula to have failed to take reasonable steps to eliminate or reduce the foreseeable harm, from which she concluded that imVula was contractually liable to Mr Loureiro on the basis of a breach of contract and delictually liable to Mrs Loureiro and their two children on the basis of delict for failing to act in accordance with its so-called “duty of care” and in conformity with the required standard for security companies - therefore both wrongfulness and negligence had been established.

iMvula lodged an appeal in the Supreme Court of Appeal (Imvula Quality Protection (Pty) Ltd v Loureiro and Others 2013 3 SA 407 (SCA)). Mhlantla JA (with Mthiyane DP, Bosielo JA and Mbha AJA concurring) approached the issue on the basis of the reasonableness of M’s conduct. She held that the contract did not impose strict liability on iMvula vis-à-vis Mr Loureiro and, in addition, that the contract had been subject to a tacit term excluding the police from the group of people who had no access to the property and that M had not acted unreasonably in his belief that the robber was a police officer. Mhlantla JA therefore concluded that iMvula had not breached its agreement with Mr Loureiro . In respect of the delictual claims, she concluded that no wrongfulness and negligence had been established on iMvula’s part. iMvula’s appeal was therefore upheld. In his minority judgment Cloete JA drew directly opposing conclusions, finding iMvula to be strictly liable in contract vis-à-vis Mr Loureiro, as well as delictually liable towards the other Loureiro family members on the basis of M’s wrongful and negligent conduct that had contributed to their damage.

In the final appeal proceedings in the Constitutional Court the judgment of the Supreme Court of Appeal was reversed in respect of both the contractual and delictual claims. Van der Westhuizen J concluded that iMvula’s liability in contract was strict and that M’s conduct had constituted an act of breach of contract towards Mr Loureiro. The court also found that the appellants had established both wrongfulness and negligence on the part of M and that iMvula could thus not escape vicarious liability in delict vis-à-vis Mrs Loureiro and the two children. In upholding the appeal, the Constitutional Court thus gave effect to the judgment of the court of first instance.

3 Critical Evaluation

3 1 Introduction

The issues that the Constitutional court had to determine were three in number. In the first place the court had to determine whether it should at all grant the aggrieved parties leave to appeal; if this question were to be answered in the affirmative, the court then would have to ascertain whether iMvula was liable to Mr Loureiro for breach of contract; and, thirdly, whether iMvula had to answer in delict for the losses sustained by the rest of the family (398F-G).

3 2 The Appeal Issue

In urging the Constitutional Court to overturn the judgment of the Supreme Court of Appeal, the Loureiros raised two arguments to afford a basis for the court’s jurisdiction in the matter at hand. The first argument was that a constitutional issue was involved, namely “the extent to which common-law actions in contract and delict give effect to the rights to security of the person, privacy and property” (404E), viz the fundamental rights enshrined in sections 12, 14 and 25 of the Constitution, 1996. This argument addresses the provision of section 167(3)(b) of the Constitution which provides that the Constitutional Court “may decide only constitutional matters, and issues connected with decisions on constitutional matters”. In the alternative, the Loureiros argued that the court had jurisdiction in terms of a recent amendment to section 167 by the Constitution Seventeenth Amendment Act, 2012 (which came into force on 23 August 2013) that authorises the Constitutional Court to hear non-constitutional matters whenever “the matter raises an arguable point of law of general public importance which ought to be considered by it”. The Loureiros’ application for leave to appeal had been filed before the amendment came into force, but it was argued on their behalf that the general tenet of statute construction prohibiting the retrospective operation of legislation did not apply due to the recognised exception to this rule against retrospectivity pertaining to procedural matters which do not detrimentally affect a party’s substantive rights (Curtis v Johannesburg Municipality 1906 TS 308 311, 313; Veldman v Director of Public Prosecutions, Witwatersrand Local Division 2007 3 SA 210 (CC) 224A-226G, 232C-233B).

Both arguments were opposed by iMvula and the court therefore commenced with an evaluation of the first of these. Van der Westhuizen J properly pointed out the well-established principle that both the law of contract and the law of delict give effect to and provide remedies for violations of constitutional rights (404H-405A, referring to Barkhuizen v Napier 2007 5 SA 323 (CC) parr 28-30, 35 and Fose v Minister of Safety and Security 1997 3 SA 786 (CC) par 58). He then proceeded to issue a warning that one should not oversimplify this issue: “[T]he mere fact that a matter is located in an area of the common law that can give effect to fundamental rights does not necessarily raise a constitutional issue” (405A). He explained this by first making a positive statement and then following up with a negative statement: First, the matter must in addition raise questions about the interpretation and development of the common law that will force the Constitutional Court to consider constitutional rights or values, notwithstanding the fact that the arguments raising these issues ultimately fail, or the common-law rule in question ultimately did not need to be altered (referring to Minister of Safety and Security v Luiters 2007 2 SA 106 par 23 and Barkhuizen (supra)); and, secondly, the matter must not simply deal with the application of an uncontroversial legal test to the facts (referring to Mbatha v University of Zululand [2013] ZACC 43 par 194).

The court then went about applying these principles to the facts (405C-406C). Van der Westhuizen J immediately pointed out that if an appellant would base his application on the fact that the lower court has in its finding on wrongfulness failed to take account of the normative imperatives of the Bill of Rights, this would ordinarily be regarded as raising a constitutional issue. Strong authority for this is to be found in Steenkamp NO v Provincial Tender Board, Eastern Cape (2007 3 SA 121 (CC) par 19) in which it was held that the mere fact that an aggrieved party has lodged an appeal against a court’s decision on wrongfulness in effect transforms it into a constitutional issue. Likewise, the judgment in Phumelela Gaming and Leisure Ltd v Gründlingh and Others (2007 6 SA 350 (CC) par 23) was referred to as authority for the contention that if a court, in applying the test for wrongfulness, ignores the sprit, purport and objects of the Bill of Rights, that would involve a constitutional matter affording the Constitutional Court the necessary jurisdiction to hear the matter. Obviously eager to establish the correct basis for making a finding on the constitutionality of the type of issue confronting the court, Van der Westhuizen J proceeded as follows (with reference to the leading cases of Minister of Safety and Security v Van Duivenboden 2002 6 SA 431 (SCA) par 21; Minister van Polisie v Ewels 1975 3 SA 590 (A) 597A-C); Carmichele v Minister of Safety and Security and Another (Centre for Applied Legal Studies Intervening) 2001 4 SA 938 (CC) par 56; Van Duivenboden par 17; and Olitzki Property Holdings v State Tender Board and Another 2001 3 SA 1247 (SCA) par 12):

“This is because of the nature of the wrongfulness element in delict. An enquiry into wrongfulness is determined by weighing competing norms and competing interests. Since the landmark Ewels judgment, whether conduct is wrongful is tested against the legal convictions of the community. These now take on constitutional contours: the convictions of the community are by necessity underpinned and informed by the norms and values of our society, embodied in the Constitution. In this case the wrongfulness enquiry ... invokes the convictions of a community plagued by crime on the crucial issue of respect for the police, its role and its interaction with the ever-growing private security industry”(405C-406A).

The court then proceeded to evaluate the decision of the Supreme Court of Appeal to the effect that M’s conduct in allowing access to police officers or those claiming to be police officers had not been unlawful. The court wasted no ink in concluding that that court failed to take constitutional considerations into account and in pointing out that if its decision were allowed to stand, “the harm-causing conduct of security companies and their employees who mistake robbers for police would not be wrongful and thus not attract delictual liability” (406B). It is suggested that at this point the court had already fully substantiated a stance that the issue at hand was of a constitutional nature which justified its hearing in that court. The next sentence should therefore be judged as an additional substantiation ex abundanti cautela, where Van der Westhuizen J expresses himself as follows: “Whether this Court should overturn the decision of the Supreme Court of Appeal poses questions about the interpretation and development of the common law” (406C). This is reminiscent of the wording of section 39(2) which enjoins every court, tribunal or forum to promote the spirit, purport and objects of the Bill of Rights. The court’s final conclusion that “[a] constitutional matter is therefore raised” (406C) is in fact the only logical finding that it could make on the basis of quoted authority and should be welcomed.

It is noteworthy that the court’s substantiation of its jurisdiction was up to this point based entirely on the nature of the wrongfulness issue in delict. It would appear that the court regarded its decision that the wrongfulness aspect of the set of facts served as a “connecting factor” to effect jurisdiction on the basis of the presence of a constitutional issue as sufficient to ensure its jurisdiction in respect of the contractual issue as well. This is clear from the following statement: “Mr Mahlangu’s negligence and the interpretation of the contract are - in this case - issues connected to the required decision on a constitutional issue” (406C-D). Two comments on this will suffice: First, the adjudication of the negligence issue in delict is in its own right a process that brings constitutional values into play. It is well established that the diligens paterfamilias test for negligence entails, similar to the boni mores test for wrongfulness and the imputability test for legal causation, the application of open-ended or flexible delictual principles that can be described as the indirect application of the Bill of Rights. Such indirect application entails that all the rules and principles of private law should be given content in the light of the basic values contained in the Bill of Rights (see, in general, Neethling and Potgieter Neethling-Potgieter-Visser The Law of Delict (2010) 21; Visser “Enkele beginsels en gedagtes oor die horisontale werking van die nuwe Grondwet” 1997 THRHR 297 299, “Enkele gedagtes oor die moontlike invloed van fundamentele regte ten aansien van die fisies-psigiese integriteit op deliktuele remedies” 1997 THRHR 495; Van der Walt and Midgley Principles of Delict (2005) 18; Neethling, Potgieter and Visser Neethling’s Law of Personality (2005) 75; Loubser and Midgley (eds) The Law of Delict in South Africa (2012) 34). It is therefore suggested that it was unnecessary to “link” negligence to wrongfulness to enable the court to adjudicate on the presence or absence or negligence. In the second place it is clear that the only way in which the court could justify its adjudication of the issue of breach of contract was to connect it “to the required decision on a constitutional issue”. The reason for this lies in the fact that fault in the form of negligence is not a requirement for breach of contract in terms of modern South African contract law, so that the issue of breach of contract is in fact to be regarded as an issue simply dealing with the application of an uncontroversial legal test to the facts, viz whether the actions of a party to a contract are factually in contravention of the terms and conditions of contract. No indirect application of the Bill of Rights was involved in this respect and on its own the issue of breach of contract would not have afforded the Constitutional Court a platform to adjudicate on the issue. As it would have been most unsatisfactory for the court to decide that it could resolve the delictual issue, but not the question pertaining to breach of contract, it is suggested that the stratagem of “connecting” or “linking” the breach of contract issue with the wrongfulness issue to enable adjudication on it, conforms to the spirit, purport and objects of the Bill of Rights and must therefore be welcomed. One could even describe the method applied in this judgment of applying the open-ended rules of delict pertaining to wrongfulness, which moves this branch of private law into the constitutional law sphere, as a “vehicle” for conveying the rules of (breach of) contract into the same constitutional law sphere. This is indeed a novel way of enhancing the influence of our Constitution.

Having decided that the issues involved were of a constitutional nature, the court did not need to consider the alternative argument based on the possible retrospective operation of the amended section 167 of the Constitution (406D). Should the issue have been pursued and the court have come to the conclusion that that section could be applied retrospectively, such decision could arguably have been applied alternatively to hold that the court had jurisdiction to decide the issue of breach of contract, as such issue would certainly fall under “any other matter... on the grounds that the matter raised an arguable point of law of public importance which ought to be considered by [the Constitutional] Court (s 167(3)(b)(ii) as amended).

Notwithstanding his positive finding on the constitutional nature of the issues involved - which, on its own, would logically suffice as reason for granting leave to appeal - Van der Westhuizen J, out of abundant caution, posed a further compelling reason for holding that it would be in the interests of justice to grant leave to appeal, namely the “substantial difference in the approaches of the majority and minority judgments in the Supreme Court of Appeal (as well as the High Court judgment)” which required the court to provide guidance on legal certainty (406E, quoting as authority F v Minister of Safety and Security and Others (2012 1 SA 536 (CC) par 38).

3 3 The Breach of Contract Issue

Mr Loureiro argued that the clause of his oral agreement with iMvula contained in clause 6.8 of the original heads of argument introduced strict contractual liability on iMvula’s part and that the Supreme Court of Appeal had erred in deciding that it was subject to a reasonableness qualification. On the other hand, iMvula contended that Mr Loureiro failed to prove that the contract was in fact amended as reflected in clause 6.8 of Mr Loureiro’s heads of argument and further that, if the contract had in fact been amended, it did not impose strict liability, because that would be inconsistent with the other applicable terms which imposed a reasonableness standard (reflected in clauses 6.5.1-6.5.2 and 6.7 of Mr Loureiro’s heads of argument). It was furthermore argued that iMvula’s conduct had been reasonable and that it had therefore not breached that contractual term (407A-C).

The court identified three issues to be decided in determining iMvula’s possible liability, viz: (a) whether the original contract had been amended by the term containing a prohibition against opening the smaller pedestrian gate without prior authorisation; (b) whether such prohibition had imposed strict liability on iMvula, or included a qualification as to reasonableness; and (c) whether the contract was breached (par 40). Although Van der Westhuizen J did not spell it out in such detail, the second issue was dependent upon a positive answer to the first question, whereas the third issue in turn depended on a positive answer to the second question. Be that as it may, the court’s line of reasoning is abundantly clear.

The court did not experience the slightest difficulty in coming to the immediate conclusion on question (a) that the term including the prohibition had, on the evidence presented in the court of first instance, amended the agreement (407F). This was a simple finding of fact and involved no legal argument.

The court then proceeded to evaluate the second issue (question (b)) and commenced by pointing out that “[i]n the absence of a contrary stipulation, the law of contract does not require fault (even in the form of negligence) for breach [of contract]” (407G). In light of the approach followed in both the majority and minority judgments in the Supreme Court of Appeal, this correct statement of the law has to be welcomed. In that court Mhlantla JA, writing for the majority, stated that “this court is required to consider the reasonableness of the conduct of the security guard in both legs of the respondents’ claims” (413H, own emphasis), viz in respect of the contractual and delictual claims. Perhaps this erroneous dictum was inspired by the equally incorrect statement of Satchwell J in the court of first instance, where she expressed herself as follows: “At the end of the day, the liability of defendant (whether on claim A in contract or claim B in delict) essentially boils down to the question of negligence” (par 43). This flies in the face of established case law to the contrary. In this respect Van der Westhuizen J referred to the judgments of Thoroughbred Breeders’ Association v Price Waterhouse (2001 4 SA 551 (SCA) par 66) and the older judgment of Administrator, Natal v Edouard (1990 3 SA 581 (A) 597E-F) as authority for his commencing statement on this aspect (see also Van der Merwe, Van Huyssteen, Reynecke and Lubbe Contract General Principles (2007) 330-331). To this one can add the recent judgment of the Supreme Court of Appeal itself in Scoin Trading (Pty) Ltd v Bernstein (2011 2 SA 118 (SCA) 122I) where Pillay AJA clearly pointed out that “contractual damage does not depend on fault”. It is indeed strange that such a recent own judgment should have slipped the attention of the Supreme Court of Appeal.

Apart from this, the evidence established that the parties had expressly agreed that there would be an absolute, viz “strict-liability” prohibition which belied a stipulation to the contrary. The court furnished three reasons for this: (i) In light of the well-established rule that contractual obligations are determined by the intention of the contracting parties as well as the fact that Mr Loureiro’s prohibition was unequivocal, and judged against the reason why the prohibition was imposed in the first place, one can only agree with the court’s conclusion that “the prohibition does not, in fact, require fault for breach” (408B). In addition Van der Westhuizen J pointed out (408G-H: n 39) that the rule “expressum facit cessare tacitum” (that which is expressed excludes (or supersedes) what is implied) ruled out any conclusion that there had been a tacit imposition of a reasonableness standard (relying on Penderis and Gutman NNO v Liquidators, Short-Term Business, AA Mutual Insurance Association Ltd 1992 4 SA 836 (A) 842I; Rashid v Durban City Council 1975 3 SA 920 (D) 924-5; and Glennie, Egan & Sikkel v Du Toit’s Kloof Development Co (Pty) Ltd 1953 2 SA 85 (C) 94). (ii) On the strength of the other terms of the agreement that required “all reasonable steps” to be taken by iMvula, the latter argued that the prohibition contained in clause 6.8 of its original heads of argument should be read in context and therefore be interpreted as containing a reasonableness standard. This was in fact also the conclusion of Mhlantla JA in the Supreme Court of Appeal (414C). Van der Westhuizen J approached this issue strictly on the basis of the evidence produced in the lower courts and reached the only logical conclusion in the following words:

“While contractual terms must be understood in context, this is no reason to think that all the terms must impose the same fault standard. And in this case, other obligations explicitly imposed a reasonableness standard and the prohibition deliberately omitted that standard. Exactly because of this, the conclusion that strict liability was imposed is compelling” (408C).

(iii) Finally, the court went out of its way to establish an additional explanation for holding that the term containing the prohibition was strict and not subject to a reasonableness proviso. The argument was raised that the terms containing a reasonableness proviso pertained to positive steps that iMvula had been obliged to take in terms of its agreement (viz “to prevent” (clause 6.5.1), “to protect” (clause 6.5.2) and “to ensure” (clause 6.7)), while the clause containing the prohibition imposed a negative obligation not to admit third parties without the applicable prior authorisation: “It makes sense that parties would contract to require a reasonableness standard for a positive obligation to do something, while not for a negative obligation not to do something - especially not to open the gate, which was at the very heart of iMvula’s contractual obligations” (408E-F). The logic of this statement escapes me. It would appear that this statement was inspired by the notion that the breach of a negative obligation is in law regarded as more serious that the failure to fulfil a positive obligation (on the authority mentioned at 408J: n 42, viz Honoré “Are Omissions Less Culpable” Responsibility and Fault (1999) 60, 65-66). Whether this supports the court’s argument, based on pure logic, is to my mind doubtful. It is suggested that the first two grounds proffered by the court are more than sufficient for holding the term in question to impose strict adherence to the contract.

The answer to question (c) posed no problem at all. The court simply referred to the facts, pointing out that M disregarded the prohibition by allowing the robbers unauthorised entry: “This amounts to a breach of contract. Whether or not he was negligent is irrelevant. iMvula is liable”(409A).

Obviously eager to quell any dint of uncertainty that may have remained in this respect, the court addressed the further argument raised by iMvula that it had in terms of section 25(3) of the Criminal Procedure Act been compelled to allow the robber who had masqueraded as a police officer to enter the premises. This argument was in fact sustained in the majority judgment of the Supreme Court of Appeal (414E), but emphatically rejected in the minority judgment (422D). In effect Van der Westhuizen J adopted Cloete JA’s minority judgment on this point and by implication rejected the patently wrong decision of Mhlantla JA by simply stating: “The demand was not lawfully made by a police officer” (409B).

The court’s final word in respect of the breach of contract issue represents a contrary approach for mere argument’s sake, viz that the clause containing the prohibition was in fact qualified by a reasonableness standard. To this the equally simple answer was that iMvula “would in any event be liable, since the standard was breached” (409B). This is obviously a mere obiter dictum in view of the conclusions reached on the basis of the strict nature of the clause in question and need not be pursued any further.

3 4 The Delictual Issue
3 4 1 Preliminary Issues

The delictual liability that iMvula could have incurred vis-à-vis Mrs Loureiro and their two children could be either vicarious - depending on the conduct of its employee, M - or direct, based on its own conduct (409C-D). Van der Westhuizen J commenced with the issue of vicarious liability by setting out the three well-established rules for holding an employer vicariously liable for the delict of his employee, namely that the employee should have committed a delict, that an employer-employee relationship should have existed and that the employee should have committed the delict while acting within the course and scope of employment (referring to the two most recent cases on vicarious liability handed down by the Constitutional Court, viz F v Minister of Safety and Security par 40 n 33 and K v Minister of Safety and Security 2005 6 SA 419 (CC) par 21, as well as and Neethling and Potgieter 365-368 who concisely set out and explains these rules). This issue was uncontentious and the court therefore proceeded to apply the law of vicarious liability to the facts. The only aspect that merited attention was whether M, iMvula’s employee, had in fact committed a delict which had caused the appellants’ damage. The court thus tacitly accepted that the two remaining requirements for this type of liability had been complied with.

Van der Westhuizen J emphatically referred (409I-J: n 45) to one of iMvula’s written arguments to the effect that M’s conduct had not been the legal cause of the Loureiros’ harm and that factors such as that the alarm system had partially been deactivated and the back door to the house had been left unlocked constituted “nova acta interveniens” (sic) (the present participle should read “intervenientes” to accord to the plural form “nova acta”). This argument involved the determination of legal causation but it had been abandoned, causing the court to conclude that it was “thus not in dispute that Mr Mahlangu’s conduct caused damage to the Loureiros”. In this way the delictual elements of conduct, causation and damage were accepted as proven. This explains why the rest of this judgment deals exclusively with the two remaining delictual elements of wrongfulness and negligence on the basis of the appellants’ arguments that M had acted both wrongfully and negligently, which was denied by the respondent (409F-410B).

3 4 2 Wrongfulness

It is to be welcomed that the court commenced with an investigation of the wrongfulness issue, although this is (unfortunately) not to be accepted as an indicator that the court thereby unequivocally endorsed the approach that the determination of wrongfulness is logically anterior to ascertaining negligence. (It is not my intention to pursue this thorny issue on which much ink has been spent over the last few years. For an overview of recent literature on this point, see Neethling and Potgieter 123 n 6). The court’s uncritical acceptance of the following dicta from Trustees, Two Oceans Aquarium Trust v Kantley and Templer (2006 3 SA 138 (SCA) 10) that “[n]egligent conduct giving rise to damages (sic) is not ... actionable per se. It is only actionable if the law recognises it as unlawful” in fact creates the impression that the negligence issue could be determined before the wrongfulness issue. The same conclusion could be drawn from the court’s assertion that “[i]t is recognised, however, that there are cases where conduct that would not be wrongful if negligent, may be wrongful if intentional, where the subjective state of mind may thus be relevant to the wrongfulness inquiry (with reference to the judgment in Minister of Finance and Others v Gore NO 2007 1 SA 111 (SCA) par 86 and academic sources such as Visser “Delict” in Du Bois (ed) Wille’s Principles of South African Law (2007) 1187 and Boberg The Law of Delict Volume I - Aquilian Liability (1984) 33) as well as from the court’s reference (410I-411G: n 50) to recent case law in which it was held that deciding which test should be applied first, is a matter of convenience (with reference to cases such as Haweka Youth Camp and Another v Byrne 2010 6 SA 83 (SCA) par 24; Gouda Boerdery BK v Transnet Ltd 2005 5 SA 490 (SCA) par 12; and Local Transitional Council of Delmas and Another v Boshoff 2005 5 SA 514 (SCA) par 20; as well as recent academic sources, such as Brand “Reflections on wrongfulness in the law of delict” 2007SALJ 76 79; Nugent “ Yes, it is always a bad thing for the law; a reply to Professor Neethling” 2006 SALJ 557 559-562 and Fagan “ Rethinking wrongfulness in the law of delict” 2005 SALJ 90 109). Van der Westhuizen J probably did not specifically apply his mind to this aspect of the law, in view of the general tenor of his judgment and thus one should not endeavour to draw conclusions of a general nature in this respect.

Equally to be welcomed is the court’s formulation of the nature of the wrongfulness test as opposed to that of negligence. After appropriately referring to Nugent JA’s warning in the leading case of Minister of Safety and Security v Van Duivenboden (2002 6 SA 431 (SCA) par 12) not to conflate the enquiries into wrongfulness and negligence - a warning not heeded by the majority of the Supreme Court of Appeal - the court expressed itself as follows:

“The wrongfulness enquiry focuses on the conduct and goes to whether the policy and legal convictions of the community, constitutionally understood, regard it as acceptable. It is based on the duty not to cause harm - indeed to respect rights - and questions the reasonableness of imposing liability. Mr Mahlangu’s subjective state of mind is not the focus of the wrongfulness enquiry. Negligence, on the other hand, focuses on the state of mind of the defendant and tests his or her conduct against that of a reasonable person in the same situation in order to determine fault” (410B-D).

Although one could embark on a theoretical discussion on virtually each and every one of the above statements, only the following comments are relevant to this discussion: First, it should be welcomed by those such as Neethling and Potgieter (123), who doggedly maintain that the wrongfulness issue should be decided before the negligence issue, that the court places conduct as the focal point of wrongfulness, and not negligent conduct. Secondly, authors in the same camp as Neethling and Potgieter will not value the court’s general approach that wrongfulness in general presents itself as a breach of a duty not to cause harm; the orthodox approach is that the duty approach should only be followed where the issue turns on the possible wrongfulness of an omission, or where wrongfulness has to be ascertained in the sphere of causing pure economic loss (see eg Neethling and Potgieter 54-56, 291 sqq). Finally, it is suggested that the court essentially contradicted itself where it was said in one and the same sentence that negligence deals with the defendant’s state of mind (which implies an absolutely subjective test) and that one should test the defendant’s conduct against the criterion of the diligens paterfamilias (which is essentially an objective ex ante facto test). It would have been better for the court to describe negligence as the defendant’s legal blameworthiness, whereas there is no harm in describing intent (dolus malus) as the defendant’s blameworthy state of mind. (On this topic, see the thought-provoking discussion by Van der Merwe and Olivier Die Onregmatige Daad in die Suid-Afrikaanse Reg (1989) 110 n 90.)

The way in which the court applied the reasonableness test for wrongfulness can be fully supported. In rejecting the finding of the Supreme Court of Appeal to the effect that M was obliged to allow the intruders masquerading as police officers access to the premises “because of his duty to cooperate with the police” (411B), the court simply declared: “The intruders were as a matter of fact robbers, not police officers” (ibid). This is a classic example of the ex post facto arm-chair (or diagnostic) approach to be taken in assessing wrongfulness, where one may be “wise after the event”, as opposed to the ex ante facto (or prognostic) approach of the diligens paterfamilias test for negligence (411C; see, in general, Neethling and Potgieter 158). It is clear that this conclusion reached by the court fortunately eliminates any possible detrimental effect of Mhlantla JA’s decision in the Supreme Court of Appeal on the future application of the reasonableness test for wrongfulness, where she flatly rejected application of the test of “an armchair critic with the benefit of hindsight” (417A). Although her assertion was, of course, quite correct in respect of the negligence element, she did not distinguish clearly between negligence and wrongfulness, which created the risk of uncertainty when applying the wrongfulness test in future. The Constitutional Court’s judgment on this point must therefore be welcomed, as it “saved” the theory of the law of delict from being subjected to unnecessary uncertainty in future.

Although one would have thought that the court’s conclusion up to this point on wrongfulness conclusively demonstrated that the wrongfulness of M’s conduct had been established, the court chose to base its conclusion on the presence of wrongfulness on a further base, namely that of policy (411D-E). In this respect the following factors were isolated as pointing towards the wrongfulness of M’s conduct: (a) The constitutional right to personal safety and protection from theft or damage to one’s property; (b) the public interest in ensuring that private security companies who assume their roles for remuneration should succeed in preventing harm; and (c) the deterrent effect of imposing delictual liability. Perhaps one should draw the conclusion that the result of the application of the ex post facto reasonableness test which yielded a conclusion of wrongfulness was merely prima facie and that it could be rebutted if the court found that policy considerations militated against such finding. However, if policy considerations dictated otherwise, the prima facie finding would become conclusive, as in the instant case. This is clearly the only way in which one could explain the court’s final verdict on this point, which reads as follows: “The convictions of the community as to policy and law clearly motivate for liability to be imposed” (411E; italics supplied).

It is noteworthy that the court only declared M’s conduct to be wrongful in the next brief paragraph (411E), which amply demonstrates that its finding at the end of the previous paragraph (see also 410C) that liability should be imposed is the true basis of its verdict on wrongfulness. This represents a clear choice for the “new” test for wrongfulness that originated in the judgment of Harms JA in Telematrix (Pty) Ltd t/a Matrix Vehicle Tracking v Advertising Standards Authority SA (2006 1 SA 461 (SCA) par 13), in deviation from the orthodox academic approach followed by authors such as Neethling and Potgieter (78-82). The Constitutional Court thus followed the line of its recent judgments in Le Roux and Others v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae (2011 3 SA 274 (CC) par 122) and Lee v Minister of Correctional Services (2013 2 SA 144 (CC) par 53) in which the “new test” has met with its approval (see also the court’s positive reference to the decision of the Supreme Court of Appeal in the Two Oceans Aquarium Trust case (par 11) in which Brand JA applied the new test). The scope of this discussion does not allow any discussion of the merits or demerits of the new approach to the wrongfulness test (see the interesting opposing views reflected in the following recent academic publications: Brand “The contribution of Louis Harms in the sphere of Aquilian liability for pure economic loss” 2013 THRHR 57; Neethling and Potgieter “Wrongfulness in delict: A response to Brand JA” 2014 THRHR 116). However, it should now be apparent that this test has won the day and that the appeal of Neethling and Potgieter (80-81) to the judgment of Moseneke ACJ in Steenkamp NO v Provincial Tender Board, Eastern Cape (2007 3 SA 121 (CC) 138) to the effect that “[f]ortunately, the Constitutional Court... and in contrast to Harms JA in the Supreme Court of Appeal, has not yet bestowed explicit recognition upon the new variation [of the wrongfulness test]”. It is crystal clear that the reasonableness of imposing liability has now undoubtedly been accepted by our highest court as an acceptable test for wrongfulness.

If the set of facts of the case under discussion represents an instance of the causing of pure economic loss by iMvula to the Loureiros, the court’s conclusion does not need to raise the eyebrows, for it has already generally been accepted by its proponents (and adversaries) that the new version of the wrongfulness test operates exclusively in the spheres of omissions and pure economic loss. However, it is suggested that an alternative interpretation of this set of facts is possible: When M took the fatal decision and unlocked the gate to allow the robbers entry to the premises, that conduct can certainly be regarded as a positive act which contributed to the ultimate damage. The mere fact that M failed to observe the terms of the contract and in general failed to act reasonably does not transform the positive act into an omission (Van der Walt and Midgley 66). The normal rule in respect of positive acts (commissiones) is that the causing of physical damage to a plaintiff’s person or property is regarded as prima facie wrongful and that in such a case the onus is on the defendant to rebut the inference of wrongfulness arising from such harm (Gouda Boerdery BK v Transnet par 12; Telematrix par 13; Roux v Hattingh 2012 6 SA 428 (SCA) par 32; Mabaso v Felix 1981 3 SA 865 (A) 871F-874F; Santam Insurance Co Ltd v Vorster 1973 4 SA 764 (A) 780G-781G). One could speculate on the outcome of this judgment had this line of reasoning been followed. Then the defendant (respondent) would have been saddled with the onus of proving that its employee had not acted unreasonably and therefore not wrongfully. Possibly the defendant would then simply have concentrated on the fault (negligence) element, which would in all probability have yielded the same practical outcome. However, this interpretation of the set of facts as an instance of liability for positive conduct should not mean that the “new test” for wrongfulness has now been extended to cover instances of positive conduct. It is quite clear that the court approached the case from a pure economic loss perspective.

3 4 3 Negligence

The court proceeded (411F-412A) from the authoritative negligence test formulated by Holmes JA in Kruger v Coetzee (1966 2 SA 428 (A) 430E-F) which has been constantly applied in our courts for almost 50 years. The test entails three questions. A positive answer to all three of these will yield a conclusion of negligence on the part of the party being tested. In respect of the facts at hand the court had to determine whether a reasonable person in the shoes of M at the time of his actions (a) would have reasonably foreseen the damage; (b) if so, would have taken reasonable precautions to avoid the harm; and (c) if so, whether M failed to take such steps (411F-412A). The court then proceeded to apply the foreseeability and preventability stages of this test to the facts. iMvula’s endeavour to portray the negligence test as purely factual (therefore not involving a constitutional issue) in order to avoid it being considered at all by the Constitutional Court was summarily disposed of in the following terms: “It is generally understood in our law that the enquiry into negligence is at least partly normative” (412B). A moment’s reflection points to the correctness of this conclusion, for what else is the diligens paterfamilias than an objective criterion, a standard of reasonableness? Two lines later the court emphasised the objective nature of this test by correctly denying that the enquiry of negligence focuses on the defendant’s belief (in the process contradicting its earlier statement that the negligence test focuses on the defendant’s state of mind (see 410D and my comments under § 3.4.2 above)).

Van der Wetshuizen J commenced his treatment of the foreseeability tier of the negligence test by concluding that a reasonable person in M’s shoes would have foreseen that the person who had approached him had been an impostor and then furnished the following reasons for his conclusion: (a) The robbers’ car was unmarked; (b) the flashing blue light had been affixed to the car’s dashboard (and not to its roof, as is conventional); (c) the clothing worn by the person demanding access was not typical of that worn by on-duty police officers; (d) the impostor had not properly presented his “identity card”, but merely “flashed” it at M without giving M the opportunity to inspect it properly; and (e) it was foreseeable to a reasonable person “in his position as security guard” - which is an alternative way of referring to a reasonable security guard - that criminals may try to gain entry to the premises by masquerading as police officers (412C-G). This practical approach to the determination of foreseeability is to be welcomed.

In light of the positive answer to the foreseeability question, the court embarked on an application of the preventability tier of the diligens paterfamilias test, entailing asking what reasonable steps could have been taken by M to avoid the Loureiros’ harm. Here the court relied on the well-established criteria that have developed over a long period, namely (i) the degree or extent of the risk created by M’s conduct; (ii) the gravity of the consequences if the harm would occur; and (iii) the burden of eliminating the risk of harm (referring to the leading judgment of Ngubane v South African Transport Services 1991 1 SA 756 (A) 776H-I; see also Van der Walt and Midgley 179; Neethling and Potgieter 145-148; Loubser and Midgley 124-129; Boberg 274 sqq). The fourth consideration regularly applied in determining whether the harm was reasonably preventable is the utility of the actor’s conduct, but this did not play a role in this set of facts. Van der Westhuizen thereupon applied these criteria to the facts and found that preventability had been established on all of the three grounds mentioned: In respect of (i) he found that the risk created by not verifying the identity of the person demanding entry had been great, the gravity of the consequences of possible harm (factor (ii)) had been equally great and the burden of taking adequate preventative steps (factor (iii)) had been slight (412H-413A).

Having answered the second question in the affirmative as well, the court finally had no difficulty in concluding as follows:

“Mr Mahlangu failed to take any of these fairly easy precautions. When one is tasked with protecting a property against intruders, it is simply not reasonable to open a door for a stranger without adequately verifying who that person is or what he or she wants. Mr Mahlangu’s conduct fell short of that of a reasonable person” (413C).

As an afterthought the court referred, although not in so many words, to the application of the imperitia culpae adnumeratur rule which entails raising the standard of care expected of a defendant when he or she possesses certain expertise in the relevant field (on the more stringent test of the reasonable expert, see Neethling and Potgieter 139-141; Van der Walt and Midgley 192-194; Scott “Die reël imperitia culpae adnumeratur as grondslag vir die nalatigheidstoets vir deskundiges in die deliktereg” Petere Fontes: LC Steyn Gedenkbundel 124). The fact that M was an experienced security guard in possession of a “Grade-A qualification” prompted the court to declare that it would be appropriate to raise the standard of conduct expected of him “to be commensurate with this expertise” (413D). This is in stark contrast with the decision of the majority judgment of the Supreme Court of Appeal (see 416E-418A of that judgment), but in conformity with the minority judgment of Cloete JA (423H-424B) on this aspect.

The court concluded its findings on negligence by rejecting iMvula’s argument - obviously presented as a desperate last stand - that the court was unable to decide the negligence issue in the absence of expert evidence pertaining to security-industry standards. Referring to the judgment in Michael and Another v Linksfield Park Clinic (Pty) Ltd and Another (2001 3 SA 1188 (SCA) parr 34-40) Van der Wetshuizen J pointed out that the negligence issue must be resolved by the court itself and that if the court is able, on the facts, to draw a conclusion on the level of the appropriate reasonableness standard, then it is unnecessary to hear evidence on the matter (413F).

Finally, having in effect found iMvula vicariously liable for M’s conduct, the court declined to consider iMvula’s possible direct delictual liability (414B). It is suggested that such an enquiry would probably have presented the Loureiros’ with a more difficult case, judged from a theoretical perspective (see, eg, the minority judgment of Froneman J in the case of F v Minister of Safety and Security 564A-575A).

4 Conclusion

The judgment of the Constitutional Court is to be heartily welcomed. The majority judgment of the Supreme Court of appeal, overturned on both the issues of breach of contract and delict, was in fact potentially calamitous in its effect on both branches of private law: it disregarded the well-established principle that fault is no requirement for breach of contract and it confused the delictual elements of wrongfulness and negligence. If that judgment had not been overturned, the doctrine of stare decisis would have presented legal practitioners, academics and students, obliged to follow the judgments of a higher-ranking court, with unnecessary uncertainties in respect of clear rules which have come about after a long process of development. The judgment of Satchwell J in the court of first instance surpassed that of the Supreme Court of Appeal majority in quality, whereas the minority judgment of the latter court provides a good example of judicial clarity in which the glaring flaws in the majority decision are pointed out. The fact that both last-mentioned judgments were in effect reinstated/upheld, justifiably allows the legal fraternity to heave a sigh of relief.

Finally, the court’s willingness to hear this appeal on the basis that constitutional principles were involved, is noteworthy. It demonstrates that the open-ended principles of both our laws of contract and delict make these branches of law particularly susceptible to judicial interpretation on a constitutional basis. This is particularly true in respect of the law of delict, on which a judge of the Supreme Court of Appeal commented as follows in a recent article (Brand “Influence of the Constitution on the law of delict” Advocate (April 2014) 42 45):

“[A]lthough the overt purpose of the law of delict is to compensate, it also plays a covert role [in] which it prescribes a set of ethical rules for social interaction. As a natural consequence, the law of delict is underpinned by a sense of morality and fairness. In this light it seems logical that constitutional values would have a dramatic impact on delict, but that the impact would be through the application rather than the amendment of established principles”.