01ÿ3456789574

1ÿ5ÿ48ÿ5ÿ74559574
1ÿ6
1ÿ74895
1ÿ59ÿ9574
1ÿÿ9574
1ÿ95
1ÿ67494
!1ÿ"67948ÿ7ÿ#95 574
0$1ÿ%5&ÿ4ÿ74565ÿ'984ÿ( 574ÿ4ÿ5
001ÿ)'974ÿ7ÿ)*7&6+ÿ%5&ÿ%&ÿ5ÿ7*4574ÿ76ÿ, 9*574ÿ34-96ÿ48ÿ ÿ( 5
0$ÿ7ÿ0!!
0 1ÿ)'*574ÿ9
01ÿ.66*574
01ÿ,74
01ÿ/45ÿ05545
01ÿ.96ÿ) 747 ÿ6
01ÿ345664ÿ15ÿ745659ÿ6574
01ÿ2419ÿ7*5574
0!1ÿ.67895ÿ%5&
$1ÿ36ÿ7ÿÿ559576&ÿ895&
01ÿ.9%ÿ95765
1ÿ.6774ÿ%5&
1ÿ34-96&ÿ76ÿ85ÿ7ÿ4756ÿ*674
1ÿ478ÿ( 845ÿ948ÿ( 5ÿÿ7ÿ0!!
1ÿ.4ÿ48ÿ9 64
1ÿ)7574ÿ75
1ÿ346445ÿ7ÿ%78&ÿ4565&
1ÿ346445ÿ7ÿ845&6ÿ*67&ÿ48ÿ8455&
!1ÿ346445ÿ7ÿ6*95574
$1ÿ"67948ÿ7ÿ-95 574ÿ7 58ÿ15ÿ46445ÿ7ÿ*6745&ÿ4565
01ÿ8565ÿ%5&
 1ÿ9 679ÿ%5&
1ÿ6ÿ656574ÿ48ÿ*77&6ÿ48ÿ4568 5
1ÿ489574ÿ48ÿ**7657445ÿ7ÿ8
Chapter 1

Introduction

1.1 Introduction

1.2 The nature

of the law of delict
1.2.1 Loss allocation and corrective justice
1.2.2 Regulatory framework
1.2.3 Morality and fairness

1.3 The premise

1.4 Definitions of a delict

1.5 The functions of the law of delict

1.5.1 Compensate for harm that has been suffered or an interest that has been
infringed
1.5.2 Protect certain interests
1.5.3 Promote social order and cohesion
1.5.4 Educate and reinforce values
1.5.5 Provide socially acceptable compromises between conflicting moral views
1.5.6 Deter the injurer from behaving similarly in future and to warn and deter others
from behaving in a similar way
1.5.7 Reallocate and spread losses

1.6 Delict and insurance

1.7 Delict and technology

1.8 Remedies

1.9 Law of delict or law of delicts?

1.10 Protected interests

 1.11 Essential elements where liability is based upon fault

1.12 Essential elements where liability is strict

1.13 A systematic approach to delictual problem-solving

1.13.1 Step 1: Identify the parties
1.13.2 Step 2: Primarily fact-based decisions
1.13.3 Step 3: Primarily normative decisions
1.13.4 Step 4: Determining the remedy
1.13.5 Step 5: Apportionment of damages

1.14 General principles modified for specific actions

1.14.1 The actio legis Aquiliae
1.14.2 The Germanic remedy for pain and suffering
1.14.3 The actio iniuriarum
1.14.4 The actio de pauperie
1.14.5 The actio de pastu
1.14.6 Vicarious liability

1.15 Conclusion

1.1 Introduction
The South African law of delict is a true hybrid. Its genealogy stretches from the Twelve Tables in
Roman Law to the Bill of Rights in our Constitution.1 At various stages in its history other legal
systems have influenced its principles; most notably, Roman-Dutch law and English law. However,
the influences have on occasion not been compatible with our law’s Roman-Dutch core
foundations, resulting in rules that are sometimes unclear and contradictory. On the surface, the law
of delict provides a system for compensating those who have been wrongfully harmed by the
culpable conduct of others; but it also expresses society’s views on what it considers acceptable
behaviour, and what it does not. At the heart of the delictual principles lie society’s legal
convictions, or boni mores, which embody legal and public policy considerations as well as
constitutional rights and norms. These views are not static and continue to develop over time, and
so past expressions of public policy (and also policy in other jurisdictions) are not necessarily
expressions of contemporary acceptable behaviour in South Africa.
For students, the difficulty often lies not in identifying the principles of delictual liability, but
in applying these principles so that one can reliably predict likely outcomes for a specific factual
scenario. It is this difficulty that causes some to lack confidence in their ability to resolve delictual
problems. On the other hand, the law of delict also traps the unwary – those who either forget that
delict is inherently a flexible set of principles that embody social policy, or those who over-
confidently and mechanically apply rules that have been rotelearnt without proper understanding.
The best way to understand delict is to keep reading cases, to stay abreast of the latest legal
developments and to ensure that the thought processes found in case law become ingrained.
This chapter focuses on:
• Introducing the law of delict
• The nature and purpose of the law of delict
• Issues that relate to the law of delict as an academic discipline, as opposed to a set of rules for
application in practice
• The social significance of delict
• Understanding delictual principles, and deciphering ‘the delict code’ – those mixed messages
that unfortunately are found in cases and textbooks.

The problem-solving process and the tools for isolating the core issues are as important as the rules
themselves. So, this chapter also sets out the backbones of the various actions, which are expanded
in later chapters, providing a small picture of the bigger puzzle, which this book, hopefully, will
assist readers in building successfully.

1.2 The nature of the law of delict

The law of delict forms part of private law and is primarily concerned with the circumstances in
which a person can claim compensation from another for harm that has been suffered. The Romans
classed it as part of the law of obligations, because when a delict has been committed, someone is
obliged to compensate another for the harm that has been suffered.

1.2.1 Loss allocation and corrective justice

The enquiry into whether or not a delict has occurred, is in fact a loss-allocation exercise to
determine whether or not the burden should shift from the plaintiff to the defendant. This point was
emphasised in an earlier edition of Street on Torts:2
Tort is often described as centrally concerned with corrective justice – that is, the
circumstances in which a wronged party is able to obtain recompense or reparation from a
wrongdoer. In consequence, the law of torts is often judged by its success or otherwise as a
compensation system. In simple terms, since most tort actions have as their objective
monetary compensation for a loss inflicted on the claimant by the defendant, the question
that most often arises is ‘who should bear the relevant cost?’ Should it lie where it falls on
the unfortunate claimant; or is the conduct of the defendant such that the law should shift
the loss to him? In the tort of negligence, and many other torts … , loss fixing is a core issue.
Tort’s ‘success’ in these areas must therefore be judged, at least in part, by its efficacy as a
compensation system.

TERMINOLOGY ‘Tort’ vis-à-vis ‘delict’

‘Tort’ is an Anglo-French word (with Latin origins) used by systems
based on the English common law that describes a civil wrong.
Civil law systems on the European Continent and those systems
that retain a strong link to Roman law, such as Scotland and South
Africa, prefer the term ‘delict’. Originally, this term included both
civil and criminal wrongs,3 but this is no longer the case.

1.2.2 Regulatory framework

The law of delict, like all other law, forms part of a regulatory framework for society. Its purpose is
also to set standards of behaviour for human conduct. So, underlying the overt compensatory
regime of delict is a ‘hidden agenda’: it also serves a normative purpose and it prescribes a set of
ethical rules and principles for social interaction. Boberg states this ‘hidden agenda’ quite openly in
the next passage:4
 Law … is a form of social engineering. This is particularly true of the law of delict, which is
close to the core problem of balancing individual freedom against collective security … .

1.2.3 Morality and fairness

Underlying delict is a sense of morality and fairness. The law of delict should give substance to
these concepts, and sometimes it requires judges to decide which conflicting moral principles
should be given practical effect when regulating behaviour in society. As Lord Atkin said in
M’Alister (or Donoghue) v Stevenson,5 possibly the most famous case in the world:
The liability for (delict) … is no doubt based upon a general public sentiment of moral
wrongdoing for which the offender must pay. But acts or omissions which any moral code
would censure cannot in a practical world be treated so as to give a right to every person
injured by them to demand relief … . The rule that you are to love your neighbour becomes
in law: You must not injure your neighbour … .

The core moral principle upon which the South African system of personal responsibility rests is
that there can be no liability without fault. There is also strict liability (that is, liability without
fault) which expresses a different morality; one in which society determines whether the nature of
the conduct, or the risk associated with the conduct, is such that a person should compensate
anyone who suffers harm as a result. Cane concludes:6
Tort law, then, is a complex mixture of principles of personal responsibility for conduct
(whether intentional, reckless or negligent) and personal responsibility for outcomes (strict
liability). Different ethical imperatives underlie these two forms of responsibility. That
underlying conduct-responsibility is not to engage in the liability-attracting conduct; and
that underlying outcome-responsibility is to compensate for adverse outcomes of the
relevant activity. Viewed in this way, tort liability based on outcome-responsibility is a sort
of tax on activities which attract such liability rather than a penalty for engaging in it.
Liability based on conduct-responsibility, by contrast, implies a disapproval of the liability-
attracting conduct which does not attach to outcome responsibility.

PAUSE FOR Fault liability

REFLECTION Writing from an English law perspective, Deakin, Johnston and
Markesinis7 point out that fault liability stems from the nineteenth-century
notion that a person committing a delict should not only be legally in the
wrong but also morally wrong. This notion corresponded with the moral,
social and economic views at the time. There was also a poorly
developed insurance industry, and so compensation came from the
injurer. Therefore, the law of delict served to compensate, show
disapproval of unethical behaviour, and deter harm-causing behaviour.
Fault liability served to limit the scope of liability and ‘meant that during
the period of industrial expansion and increased industrial accidents, an
enterprise could be shielded against the costs of accidents’.8
However, different economic conditions and socio-political views
prevail now, with ideas moving from loss-shifting to loss-spreading
functions. For example, would no-fault regimes for product liability or road
accident compensation not better reflect current societal morality in
welfare-orientated states? Would such strict liability not be less arbitrary
and more fair, given that all injured persons would be compensated
irrespective of whether someone else was at fault? Would it not also be
more fair on plaintiffs, who might well be bankrupted by massive pay-
outs?
 Nonetheless, Linden and Feldthusen point out that the fault system ‘is
riddled with imperfections; it is costly, difficult to administer, denies
compensation to many injured people, and is replete with delays’; but the
authors also contend that ‘it is also a mark of nobility when a society
directs its members to conduct themselves reasonably in their relations
with their fellow citizens or pay for the consequences’.9

1.3 The premise

A famous American judge, Oliver Wendell Holmes, wrote that ‘sound policy lets losses lie where
they fall, except where a special reason can be shown for interference’.10 Our courts adopt the same
principle and when considering whether delictual liability ought to be imposed, the starting point is
always that loss should lie where it falls.11 This means that as a general rule, plaintiffs must suffer
the infringement of their interests and bear any financial loss associated with such infringements;
unless they can show that there is a special reason for shifting the burden of making good any loss
to someone else. The special reason for reallocating the loss arises when a plaintiff proves all the
elements of a delict. These elements are discussed in more detail later in this chapter and in the rest
of the book.

1.4 Definitions of a delict

Definitions of a delict vary according to authors’ views of the legal framework governing the law
of delict. Consider the following definitions:

Van der Merwe and Olivier:12

A delict is understood to be a wrongful and culpable act that causes another harm or
infringes another’s personality interest. Within this realm of the law of delict belong all the
rules that determine the private-law liability of a person who has caused harm or a
personality infringement to another in a wrongful and culpable way.13

Boberg:14
Van der Merwe and Olivier regard fault as an essential characteristic of delictual liability,
so that instances of no-fault liability (such as liability for damage caused by animals and the
vicarious responsibility of a master for the delicts of his servant) are in their view not
delictual, but arise ex variis causarum … . For the same reason an interdict (which can be
obtained without showing fault) does not seem to them a delictual remedy … . There seems
to be no warrant for this curtailment of the ambit of delict … .

Neethling, Potgieter and Visser:15

Boberg … defines a delict as the ‘infringement of another’s interests’. This description is,
however, both incomplete and misleading. On the one hand there is an omission to state
fault as a general delictual requirement, and on the other hand the erroneous impression is
created that all individual interests, and not only those that are legally recognised and
protected, are relevant in this regard.

Van der Walt and Midgley:16

 In general terms a delict can be defined as a civil wrong … . A narrower definition considers
a delict to be wrongful and blameworthy conduct which causes harm to a person.

Although these authors define delict differently, these extracts show that delict is part of civil, or
private, law. (In fact, together with the law of contract, it is one of the central features of the law of
obligations.) It deals with situations in which one person has wronged another, and sets the rules for
determining whether the person whose interests have been infringed should have a remedy against
the wrongdoer or another person in respect of the harm caused. Only those infringements that the
law recognises as worthy of compensation will result in liability. In most instances, the law insists
that a wrongdoer or person from whom compensation is sought should also be at fault. However, in
some instances, society considers fault not to be an element and liability is said to be strict.

PAUSE FOR The definition of a delict

REFLECTION In these extracts, the authors appear to be talking past each other, with
some talking about delict as a concept and others talking about the
requirements of delictual liability. If this duality of meaning is accepted,
would Neethling, Potgieter and Visser’s criticism of Boberg’s view still be
valid?
Van der Merwe and Olivier equate delictual liability with fault liability
and treat instances of no-fault liability as exceptions to the rule. Others
define delict more broadly, to include instances of no-fault liability. Does
this really matter? Are these not merely differences of approach, which in
practice have no bearing on the outcome?
All these extracts also root delict in the common law. None of them
refer to statutory liability, for example, liability under the Electricity
Regulation Act 4 of 2006, the National Nuclear Regulator Act 47 of 1999,
or the dispensation set out in the Road Accident Fund Act 56 of 1996, as
part of the law of delict. Should this be so? Is there a case for including at
least some of these statutory liability or regulatory regimes in the
definition of delict?

1.5 The functions of the law of delict

According to Cane, the functions of delict ‘are those purposes or ends which people seek to further
or achieve through tort law’.17 However, in isolating these purposes, it is worth remembering that
the law is not static and that ideas may fluctuate over time. Deakin, Johnston and Markesinis note:18
The aims of the law of tort [delict] have changed throughout its history: appeasement,
justice, punishment, deterrence, compensation, and loss-spreading can be counted amongst
them. None of them has offered a complete justification for the law … . Overall, however, it
can be said with some measure of confidence that at different stages of development of tort
law one of its functions may have been more prominent than the rest. Moreover, each in its
historical setting reveals something about the socio-economic and philosophical trends of the
day.

From the information so far, one can isolate the following inherent functions of the law of delict,
some of which articulate wider social goals and others that relate to those who are party to a
specific action.
1.5.1 Compensate for harm that has been suffered or an interest that has
been infringed
Compensating for harm is the primary, but not the sole function of the law of delict.19 At the same
time, one should recognise that delict is not a victim’s only source of compensation – insurance,
social security and statutory compensation schemes play a similar role – and one should not over-
estimate delict’s ability to serve as a compensatory mechanism, especially since the civil
proceedings are cumbersome, time-consuming and costly.20 While compensation in the form of
damages may be significant, financial costs are high and only a small number of people benefit
from this system.

COUNTER Does delict provide the most appropriate compensatory

POINT mechanism?
Consider the following:
• The common law of torts in the United Kingdom has been criticised as being
ineffective in its principal aim of compensating harm especially in cases arising
from personal injury, disease and death.21 Dissatisfaction with the operation of
the tort system received widespread academic attention during the 1960s and
1970s.22 During the same time, mass tort litigation drew public attention to the
clumsy, time-consuming and costly nature of obtaining compensation by
instituting civil proceedings.23
• Since 1972, New Zealand has had a statutory compensation scheme for
personal injuries,24 which partially replaced the common law with a general
compensation scheme providing compensation for harm resulting from all
accidents and some diseases.25 The rationale was that this would address the
concerns about the high transaction costs of the civil litigation system.
• Against this background, Deakin, Johnston and Markesinis suggest26 that tort
claims represent only around 25% of all compensation paid out in the United
Kingdom. In other words, approximately 75% of victims’ compensation comes
from other sources, for example, insurance, social security and statutory
compensation schemes.
• In a similar vein, Atiyah’s Accidents, Compensation and the Law,27 citing figures
contained in the Pearson Report,28 suggests that approximately 6,5% of injured
persons in the United Kingdom receive compensation via tort claims, although
the payments amount to almost 50% of the total compensation paid and
account for about 40% of the total cost (payments and administration).29
• Figures for South Africa are not known, but they are likely to show similar
trends.

Is it fair to say that delict is a ‘slow, cumbersome and expensive’

compensatory mechanism? What other sources of compensation exist in
South Africa? Are they more efficient? Would a statutory compensation
scheme similar to the one in New Zealand be more efficient than the
common law in providing compensation for harm suffered?

1.5.2 Protect certain interests

Where activities take place in a common space and sometimes compete with one another, social
and business interactions can give rise to tensions. These are usually tolerated, but where interests
are impacted upon, the law offers protection, but only where it is socially desirable to do so. At first
the law protected personal security and personal property, and particular infringements of
personality. In time, as society became more sophisticated and demanding, and as our ability to
understand intangible interests improved, the scope of protected interests expanded to include:
• Mental health in the form of pain and suffering in relation to physical injury
• Mental health (psychiatric interests) generally
• Business interests, such as goodwill and customer bases
• Interests that are purely economic
• Personality interests such as privacy and identity.

However, the law of delict is not an exclusive vehicle for protecting victims’ interests. In some
instances, courts consider victims’ interests secondary to the interests of defendants (for example,
when recognised defences such as self-defence or fair comment are present).30 In other instances,
society’s interest prevails, for example, when courts deny claims because there is a need to limit the
scope of delictual liability.
Delict protects personal and property interests by either holding or threatening to hold people
to account. It enables society to call upon police and other government organs to be more
responsible, manufacturers to produce safer goods, and professionals to provide quality services.

1.5.3 Promote social order and cohesion

Delict promotes social order and cohesion by resolving particular types of disputes, ensuring justice
between the parties, keeping the peace and preventing self-help. Initially, in Roman law, where the
law of delict was seen as a supplement to criminal law, the purpose of delict was to buy off a
victim’s vengeance rather than compensate.31 Delict has since lost its quasicriminal function, yet it
still aims to vindicate, satisfy and appease. A delictual remedy, therefore, serves as a type of
‘ethical retribution’32 in some instances; but one cannot apply such moral vindication where liability
is strict, or even in some instances of negligent conduct where fault is more a statistical inevitability
than a form of ‘immoral’ behaviour.

PAUSE FOR Punishment

REFLECTION Although it was a strong feature in Roman times, punishment is not listed
here as a separate function of the law of delict. However, there is still
some controversy in modern law as to whether delict has a punitive
function. There is no doubt that both the lex Aquilia and actio iniuriarum
of Roman law were penal actions that prescribed penalties for wrongful
conduct. In Roman-Dutch law, the lex Aquilia became purely
compensatory, but the penal features of the actio iniuriarum remained.
There is a view that this situation is still in place today and it is possible to
find older cases that awarded additional sentimental damages as a
penalty for conduct that was especially outrageous. More recent cases,
including some decided by the Constitutional Court,33 view such a
purpose as unjustifiable in a modern system of law. On this view, the
function of delict is primarily to compensate; and of criminal law, to
punish.
Should this be the case? How does one explain the fact that courts
award larger sums in damages when the facts indicate that the delict
committed is more serious than in other instances? Would a punitive
delictual system withstand constitutional scrutiny? Note that in the
American system, and to a lesser extent in the English system, courts
may award punitive damages over and above compensatory damages.
Should South Africa follow suit?

1.5.4 Educate and reinforce values34

Delict is a complex mixture of principles of personal responsibility for conduct and outcomes. So,
an important aim of the law of delict is to articulate a set of normative rules of behaviour, to
provide guidance to individuals on how they ought to behave, and to express disapproval of certain
types of conduct. According to Linden and Feldthusen,35 the ‘fundamental goal is individual
restraint and respect for one’s fellow creatures, something which is required more than ever in mass
urban societies.’ Also, now more so than before, delict serves to uphold and vindicate fundamental
human rights.

1.5.5 Provide socially acceptable compromises between conflicting moral

views
There are often a number of morally acceptable approaches to a particular issue, usually based on
different points of departure. One example is the conflict between maintaining a person’s dignity
and freedom of expression. The issue in such instances is not to prefer or to deny either of these
moral outcomes, but to resolve each particular conflict by balancing these rights and freedoms in
the circumstances in a socially acceptable way.

1.5.6 Deter the injurer from behaving similarly in future and to warn and
deter others from behaving in a similar way
While one of the functions of delict is to deter people from behaving in a certain way, the extent to
which delict achieves this objective is open to doubt. For example, people probably think more
carefully about what they say about others because of the law of defamation, and the size of a
damages award, if publicised, might impact on the way others behave. However, this would mainly
be limited to intentional conduct and some instances of negligent conduct, such as negligent
misstatements, product liability or professional negligence. Some drivers of motor vehicles might
drive more carefully, but one cannot say for sure whether it is delict or criminal law, or both, that
constitute the deterrent. In any event, the extent of reckless and/or negligent driving, and driving
under the influence of alcohol that occurs on our roads might indicate that the deterrent effect is a
myth, or of a very low order.36 Also, for deterrence to be considered an effective function, one
would have to assume that people consistently behave rationally, weighing up the potential
consequences of their conduct before proceeding.

PAUSE FOR Deterrence in delict

REFLECTION The vast majority of delict cases involve traffic accidents, most of which
‘result from regrettable, but statistically inevitable, lapses of attention’.37
What role does deterrence play in such cases, and to what extent is this
deterrent function diluted when someone is very rich, or too poor to pay,
or is covered by insurance?

1.5.7 Reallocate and spread losses

Reallocating losses from victim to injurer is a generally accepted function of delict, but the idea
that delict should also spread losses is controversial, for then neither the victim nor the wrongdoer
would bear the responsibility for the harm if others have to share the responsibility of repairing it.
Nonetheless, there are instances of loss-spreading in the law of delict. An oft-cited rationale for
vicarious liability is that someone with deeper pockets than the wrongdoer – for example, a
person’s employer who benefits from the employee’s services – is better placed to compensate the
victim.
 PAUSE FOR Loss-spreading
REFLECTION An example of spreading loss is the compensation fund established
under the Compensation for Occupational Injuries and Diseases Act 130
of 1993. In terms of the Act, employers are statutorily obliged to
contribute towards a statutory compensation fund and the money in the
fund is then used to compensate employees who became injured or
diseased during the course and scope of their employment. Therefore,
the loss of occupational injuries and diseases is spread between
employers generally who, in turn, are no longer required to compensate
employees who may become injured or diseased as a result of their
culpable wrongdoing.

Cane argues 38 that delict is primarily a system designed to ensure personal responsibility for one’s
conduct. Delict is not, he says, a social welfare system, although he accepts that it partly serves to
further ‘desirable human and social goals.’ For Cane, ‘the ability of a party to spread losses by
insurance or passing-on is not the basis of a principle of personal responsibility’. 39 Linden and
Feldthusen similarly argue 40 that tort (delict) was not designed to undertake loss-spreading, and
those who advocate that all losses should be made good, move outside the realm of delict and enter
the domain of social welfare.
The advocates of loss-spreading as a delictual function use vicarious liability to justify their
view, pointing to its rationale that employers are better placed to ensure against employees’ delicts.
They also point to strict liability regimes for product liability, where manufacturers are better
placed than individual consumers to spread losses by means of insurance. Cane counters this by
saying that one could equally find justification for vicarious liability in the principle of personal
responsibility and argues that delict:
by reason of its structure, is not well-designed to perform this function … .41 If the court
were allowed to go beyond those two parties (the victim and the injurer), it might identify a
much better loss spreader than either of them.42

Linden and Feldthusen agree:43

[Tort recovery] is welcome enough if there is nothing else available, but if full and swift
compensation is the only task of tort law, it should be replaced by something else less costly
and less dilatory.

COUNTER What role should insurance play in determining delictual liability?

POINT The Australian High Court noted in Kars v Kars:44

[C]ompulsory insurance has for many years and in most

litigated motor vehicle injury cases, … substituted the deep
pocket of an insurer for what may be the empty pocket of
the tortfeasor as the source of the injured party’s
recovery.45

Whether this holds true for South Africa is moot, for there is no clear
evidence that courts take a pro-plaintiff stance in Road Accident Fund
cases, making delict an instrument of a welfare-orientated state.
However, outside the area of compulsory insurance, it is clear that the
existence of insurance, or the ability to insure, is usually disregarded and
 so a person might conclude that loss-spreading is not considered to be
one of delict’s functions.
The question of whether our society has changed sufficiently over the
years to warrant loss-spreading as an additional purpose, still remains.
The Supreme Court of Appeal has, however, noted:46

It would be fair to say that there has been great expansion

in recent years of the use of independent contractors, and
out-sourcing in the place of employees. It is unlikely that
vicarious liability for servants would ever have developed if
servants as a class had been capable of paying damages
and costs. The historical rationale for imputing liability to a
master, namely that they had deeper pockets hardly
applies, I daresay, to most modern contractors, who may in
fact be wealthier than their principals. Where both principal
and independent contractor are large firms or covered by
insurance, the incidence of liability may not matter much.
But where the principal is an individual without insurance,
the imposition of liability upon him may cause grave
hardship. From the point of view of a plaintiff, the only case
in which the liability of a principal is advantageous is where
the independent contractor is unable to pay damages.
Whether indeed this situation is sufficiently frequent to
warrant provision being made for it must be open to doubt,
particularly when it adds so greatly to the difficulty of the
law. Courts have to be pragmatic and realistic, and have to
take into account the wider implications of their findings on
matters such as these ….

1.6 Delict and insurance

People are able to insure against losses or harm that they might suffer, including loss or harm
caused by a delict. In theory, a victim of a delict who has also taken out indemnity insurance, has
both a delictual action against the wrongdoer and a contractual claim against the insurer. The
insured person will be entitled to both the damages recovered from the wrongdoer and the proceeds
of the insurance claim. However, in the case of indemnity insurance, an insured person who has
instituted a delictual claim against the wrongdoer and succeeded in recovering damages equal to, or
in excess of the insured amount, is obliged, in terms of the indemnity insurance contract, to repay to
the insurer any amount he or she had received under the contract. This is done to prevent the
insured receiving double compensation.
The doctrine of subrogation, which seeks to prevent double compensation and to ensure that
wrongdoers take some form of legal responsibility for the harm they have caused, is also relevant in
this context. In terms of this doctrine, an indemnity insurer may institute a delictual action against a
wrongdoer in the name of the insured person, without cession of action; provided that the insured
person has not claimed damages from the wrongdoer. The doctrine also entitles an insurer to limit
or exclude liability by defending delictual claims in the name of the insured person.
In Commercial Union Insurance Company of South Africa Ltd v Lotter 47 the Supreme Court of
Appeal described the operation of the doctrine of subrogation as follows:
 It is trite law that an insurer under a contract of indemnity insurance who has satisfied the
claim of the insured is entitled to be placed in the insured’s position in respect of all rights
and remedies against other parties which are vested in the insured in relation to the subject-
matter of the insurance. This is by virtue of the doctrine of subrogation, which is part of our
common law.

The doctrine of subrogation means that insurance companies in fact conduct many delictual actions
ostensibly instituted by the person who suffered harm, or ostensibly defended by the wrongdoer. In
many such cases, a person would not have instituted delictual action if indemnity insurance did not
exist.

PAUSE FOR The effect of insurance on delict’s deterrence function

REFLECTION Linden and Feldthusen argue:48

The advent of liability insurance has removed some of

[delict’s] prophylactic power because the civil sanction is
rarely applied against the tortfeasors themselves. When a
judgment against an individual is paid by the individual’s
insurer, whatever preventive force tort law retains is further
enfeebled.

• Does indemnity insurance reduce delict’s deterrent effect?

• Are all losses covered by insurance? Are there circumstances in which
insurance cover can be refused or forfeited and where delict would provide the
only compensatory recourse?
• Would insurance influence delict in instances of contributory negligence, where
compensation claims are not paid out in full? (The defendant might be assisted,
but the contributory negligent plaintiff would not be.)
• Are there ways in which insurance companies can influence their insured
persons to behave more carefully? What deterrent mechanisms are available to
insurers?

1.7 Delict and technology

The internet and electronic revolutions have created novel opportunities for people to commit
wrongs against others. There is no statutory provision regulating compensation for harm caused
through the use of electronic media and this begs the question as to whether or not traditional
delictual principles are appropriate for dealing with legal issues associated with cyberspace. Courts
are increasingly confronted with ‘cyber-delicts’, where harmful material is found on websites,
Facebook, emails and SMS messages, primarily in the fields of defamation49 and invasions of
privacy and identity.50 However, it is only a matter of time before courts are confronted with
patrimonial harm issues such as unfair competition, product liability and for causing pure economic
loss in the cyberspace arena.
Thus far, courts have found the common law sufficiently flexible in dealing with cyberspecific
delicts. In the same way as Roman and Roman-Dutch law evolved to cope with demands emanating
from the industrial revolution, so too has there been no need to deviate from standard principles in
the technological era and courts are still able to perform their loss reallocation and compensation
functions satisfactorily.
In many of the chapters that follow, readers will find examples of how courts have adapted the
law to modern situations while still remaining faithful to delict’s historical roots.
1.8 Remedies
The usual remedy when a delict has been committed is compensation, that is, to claim damages for
the harm that has been caused. Another commonly sought remedy is the interdict, where a person
seeks an order to prevent existing harm from continuing, or to prevent threatened harm from
occurring. However, strictly speaking, an interdict is not a remedy in terms of any of the actions,
because when a person asks for an interdict, that person is not seeking compensation, but is trying
to prevent loss. So, all the elements of delict, except the loss requirement, must be present for
courts to grant an interdict. More recently, courts have also explored retraction and apology as a
delictual remedy.
In Roman law there were a variety of separate civil wrongs, each with its own rules. In time,
almost all of these became subsumed under the actio legis Aquiliae and the actio iniuriarum,
causing McKerron to declare:51
The result is that today the Aquilian action and the actio iniuriarum are the foundation-
stones of the law of delict – the former having become the general remedy for wrongs to
interests of substance, the latter, as in the old law, affording a general remedy for wrongs to
interests of personality.

This statement is not entirely accurate however, as it fails to mention the Germanic remedy for pain
and suffering that later became part of our law. Therefore, it is better to say that in seeking remedy,
one should proceed with the actions set out in Table 1.1.

Table 1.1 Actions to keep in mind when seeking remedy

Form of harm Action taken

Patrimonial loss (damnum iniuria datum) Actio legis Aquiliae

Non-patrimonial harm in the form of pain and suffering or loss in the amenities of The Germanic action for
life associated with bodily injury to the plaintiff and psychiatric injury pain and suffering

Non-patrimonial harm in the form of an injury to a personality right (an iniuria) Actio iniuriarum

PAUSE FOR Union Government (Minister of Railways & Harbours) v Warneke52

REFLECTION Innes J said:53

That being so, it becomes necessary to consider the

fundamental features of this form of action which have a
bearing upon the matter before us. And we are at once
faced with the fact that it was essential to a claim under the
lex Aquilia that there should have been actual damnum in
the sense of loss to the property of the injured person by
the act complained of … . In later Roman law property came
to mean the universitas of the plaintiff’s rights and duties,
and the object of the action was to recover the difference
between that universitas as it was after the act of damage,
and as it would have been if the act had not been
committed … . Any element of attachment or affection for
the thing damaged was rigorously excluded. And this
principle was fully recognised by the law of Holland. As
pointed out by Professor De Villiers …, the compensation
recoverable under the lex Aquilia was only for patrimonial
 damages, that is, loss in respect of property, business, or
prospective gains. He draws attention to the clear cut
distinction between actions of injuria (where intent was of
the essence), and actions founded on culpa alone. In the
former case compensation might be awarded by way of
satisfaction for injured feelings. In the latter, all that could
be claimed was patrimonial damage, which had to be
explicitly and specifically proved. The difference between
the two forms of relief is emphasised by Voet …, who states
that where one and the same act gives ground for both
actions, the receiving of satisfaction for the injuria does not
bar the claim for patrimonial loss resulting from the culpa.
The award of compensation for physical pain caused to a
person injured through negligence, which was recognised
by the law of Holland, constitutes a notable exception to the
rule in question. Professor De Villiers has some interesting
remarks upon this position, which was probably the result
of the influence of Germanic upon Roman law. But however
that may be, there is no warrant for any such exception in
the case of mental distress or wounded feelings causing no
physical injury. Damages calculated on that basis were
wholly outside the scope of the Aquilian procedure.

Even though delict is based on the three pillars outlined in Table 1.1,
there is no need to specify which action one relies upon and to claim
separately under each of them. Common practice is to claim redress in a
single composite action, but only if the requirements of the action
appropriate to the type of interest infringed have been met.

COUNTER The modern law of delict

POINT Neethling, Potgieter and Visser:54

In contrast to the casuistic approach of the Roman law of

delict, the South African law of delict is based … on three
pillars: the actio legis Aquiliae, the actio iniuriarum and the
action for pain and suffering. Unlike the last-mentioned
action which developed in Roman-Dutch law, the first two
remedies had already played an important role in Roman
law.

Van der Walt and Midgley:55

Delict lawyers can no longer focus purely on common-law

principles and treat the Constitution as an imposition, or at
best an adjunct to those principles. Mention was made in
the ‘Preface to the Second Edition’ that the Bill of Rights
would in time become the fourth pillar upon which our Law
of Delict will rest. Recent decisions of both the
Constitutional Court and the Supreme Court of Appeal have
shown that constitutional principles and values are now an
integral part of the fabric of the law of delict and must find
expression in the basic principles of delict.
 Is it correct to say that the Constitution ranks as a fourth pillar? Is the
Constitution a source of law, setting out rules of liability in the same way
as the Roman and Roman-Dutch actions do?
Consider the following extract from Street on Torts, with reference to
English law: 56

Tort law has always protected certain ‘human rights’.

Indeed, tort disputes are by definition about the competing
claims of persons to protected interests and we might well
denominate the more frequently upheld interests as
‘rights’… The enactment of the [United Kingdom] Human
Rights Act 1998 significantly ‘enhanced’ this protection
(most immediately with respect to actions of public
authorities) … Indeed, the Act’s passing prompted
academics and judges almost immediately to rethink the
boundaries and substance of tort law …

Are there parallels with the situation in South Africa following the
adoption of the Constitution? Does the passage in Street on Torts in any
way support the notion that one should consider the Bill of Rights as a
source of delictual rules? Can one say that the modern law of delict is
merely an expression of age-old concepts in a constitutional state?

1.9 Law of delict or law of delicts?

Over the years, it became accepted that delictual liability in our law is based upon a set of general
principles derived from the Roman and Roman-Dutch law actions: the actio legis Aquiliae, the
actio iniuriarum and the Germanic remedy for pain and suffering, with modifications as a result of
English law influence. Neethling, Potgieter and Visser refer to this as a ‘generalising’ approach,
whereby ‘general principles or requirements regulate delictual liability’. 57
The Anglo-American common-law systems are different. While their law also developed out
of various actions, these actions did not lose their identity. As a result, these systems contain a
series of separate torts, each with its own name, set of principles and technical distinctions, for
example, the tort of negligence and the tort of breach of a statutory duty. Neethling, Potgieter and
Visser refer to this as a ‘casuistic’ approach, whereby liability arises only if the requirements of a
specific delict are satisfied.58
The civil law systems in European law also adopt the generalising approach. Their civil codes
contain a few general principles that courts apply to particular factual situations,59 unlike in the
common-law countries, where facts need to fit technical requirements resembling formulae.60 In
such a system, precedents are not as important as in the common law, for one can treat new
situations on principle and do not have to fit them into previously established categories.
The doctrinal untidiness that accompanies the casuistic approach has caused common-law
lawyers to reflect on whether it is better to have a system of tort, instead of a system of torts. This is
not a new debate – over the years great scholars such as Williams, Salmond, Winfield and Pollock
have been involved – but, while the end is not in sight, the casuistic system continues to find
favour. So, for example, while Fleming appears to be critical, he settles for the traditional approach
because of ‘the futility of seeking to impose a spurious unity upon a very complex congeries of
problems’ where different interests with different social values call for different methods of
resolution.61 Cane, on the other hand, sees the law of tort as ‘a system of ethical rules and principles
of personal responsibility for conduct’ and is more partisan.62 The law of tort, he says, has both
backward-looking functions (the resolution of disputes and provision of remedies) and forward-
looking functions (guiding citizens on what behaviour the law allows and prohibits). A system of
discrete and disparate torts fails to provide a coherent ethical foundation for personal responsibility
and the necessary guidance for resolving future disputes.63

COUNTER The South African approach

POINT The following statement, made in Perlman v Zoutendyk,64 has often been
cited as summing up the South African approach:

Roman Dutch Law approaches a new problem in the

continental rather than the English way, because in general
all damage caused unjustifiably (injuria) is actionable,
whether caused intentionally (dolo) or by negligence
(culpa).

Is this statement a valid reflection of our approach? Consider the

following:
• The actio legis Aquiliae, the Germanic remedy for pain and suffering, and the
actio iniuriarum have different requirements, particularly in respect of the fault
element, and there are also instances of strict liability.
• Specific rules for particular types of cases are emerging. Even though they are
all infringements of personality rights, the requirements for liability for assault,
deprivation of liberty, insult and defamation have evolved differently, as have
liability for psychiatric injury and unlawful competition under the Aquilian action.
• The Supreme Court of Appeal appears to favour a view that fault is a primary
driver of the wrongfulness enquiry and that different rules should apply where
harm is caused negligently from those that apply when the same type of harm
is caused intentionally.65
• Should there be uniform general principles of Aquilian liability instead of
different rules for intentional and negligent delicts within the pillars? Doctrinal
fragmentation occurs when peculiar rules are devised for particular situations.
Although some differences are inevitable, it may be preferable to strive for
similar rules for all situations. Since a particular set of facts (for example, a
severe physical assault) can give rise to a claim under any of the three delictual
actions, perhaps one should seek coherence across the broad spectrum of the
law of delict. Additionally, since the same set of facts might infringe separate
rights to dignity, privacy and reputation, should the principles under the actio
iniuriarum not be consistent, so that similar criteria can apply to similar
situations?

Neethling, Potgieter and Visser may have the answer. They state:66

Supplementary to the generalising approach, it is, after all,

a secondary characteristic of our law of delict that specific
forms of delict with their own specific rules have evolved in
order to promote the practical utility of general principles in
given fields of delictual liability … . Here, the casuistry and
detail of the ‘law of torts’ have fulfilled an important and
beneficial role. This blend of the ‘general’ and the ‘specific’
constitutes the hybrid character of our law of delict, in
which Roman-Dutch law and English ‘common law’ have
been merged into a harmonious unity … .

One could perhaps also have regard to criminal law, which is known to
have a set of general principles informing a series of disparate specific
crimes. Is this the direction in which South African law could be heading?
In this book, we adopt the view that a principled approach to the law of delict best provides
consistency and clarity in applying our law. This is not only for the sake of doctrinal tidiness, but
also because a proper framework enables lawyers to predict potential outcomes better and advise
clients with a greater degree of certainty. Figure 1.1 sets out, perhaps provocatively, a conceptual
structure for the law of delict and the interrelationship of the various loss-allocation components.
Notice (a) that the diagram shows that the law of delict is not limited to instances of fault liability
and (b) that other compensation regimes, although relevant, and even crucial to an overall
compensatory framework, fall outside its ambit.

Figure 1.1 A conceptual structure for the law of delict and the interrelationship of the various loss-allocation
components

1.10 Protected interests

A persistent theme throughout this book is that it is the nature of the plaintiff’s interest that has
been infringed, and not the nature and quality of the defendant’s fault, that sets the stage for the
delictual problem-solving process.67 Murphy, for example, notes: 68
In the first edition of this work, Street’s emphasis on the claimant’s interests as opposed to
the defendant’s wrongdoing was perceived as radical, even bizarre … Certainly, no claim in
tort can succeed, however morally reprehensible the defendant’s conduct, unless the court
first recognises some form of harm suffered by the claimant that involves a violation of an
interest sufficient to confer on the claimant a legal right to protection of that interest.69
Although seemingly innocuous, this simple reality is not always recognised and does not
necessarily have universal favour.
However, delict is not only concerned with a plaintiff’s interests, as Cane points out:70
Because of its correlative structure, tort law is concerned with the interests of defendants as
well as those of plaintiffs, just as it is concerned with the conduct of both. Furthermore,
because the rules and principles of tort law are of general application, and because tort law
is a publically enforceable set of ethical rules and principles of personal responsibility, wider
social interests are relevant to defining the scope of tort liability.

So, the law of delict is concerned with protecting the interests of plaintiffs, defendants and society
in general, and where these interests conflict, delict tries to reconcile them in an optimal way. It
does so at a variety of levels. For example:
• A decision to recognise a class of plaintiffs, and to give them title to sue – for example,
allowing juristic persons to sue for defamation where financial prejudice might be suffered – is
the result of a policy decision to protect particular interests of such plaintiffs.
• Societal interests predominate when the nature of liability – fault in the form of negligence or
intention, or strict liability – is chosen for a particular category of claims.
• Societal interests also predominate when courts restrict the ambit of the law of delict by
refusing concurrent claims where relief might be sought in either delict or contract, or delict
and administrative law.

All three categories of interests come into play when a decision is made as to the type of harm that
is actionable in delict, for example, deciding whether:
• Grief and inconvenience should be compensated
• Conduct is wrongful or whether a ground of justification is available to the defendant
• The conditions for factual and legal causation have been satisfied.

The fault requirement focuses more on the defendant’s interests, although the standards that society
sets and expects are inherent in the decision. All three sets of interests feature strongly when
deciding upon the nature of the remedy available (punitive damages, retraction and apology), the
level of compensation to be paid, and how damages should be apportioned.
Almost every milestone along the decision-making process involves a decision that concerns
whose interests, and which interests, ought to predominate. The common theme is that in all
instances, society’s interest is the deciding factor. Sometimes these overlap with those of the
plaintiff, sometimes they fall towards the defendant’s side, while in a few instances the policy
decision is a systemic one made irrespective of the parties’ interests.
The range of interests that are protected is vast, and includes both existing and future interests.
Delict protects one’s person and personality, and one’s physical and non-physical property, such as
intellectual property interests. The range is not limited to those interests that exist in one’s personal
sphere: business, trading, and other economic interests and consumer interests are similarly
protected.
Irrespective of their nature or their origin, the types of interests that the law of delict seeks to
protect from invasion fall into the categories in Table 1.2.

Table 1.2 The types of interests that the law of delict seeks to protect from invasion

Type of interest Method of protection

Property (including physical Provided that they result in some diminution of a person’s patrimony or estate,
and intellectual property, invasions of any corporeal or incorporeal property interests, whether existing or
dependants’ maintenance anticipated, will be protected by the Aquilian action (actio legis Aquiliae). This
 rights, current and future
profits, goodwill, and
inheritances)
means that these interests receive strong protection, similar to bodily interests,
against both negligent and intentional invasions.
Should a domesticated animal cause harm to property, the actio de pauperie
provides that the owner of the animal will be strictly liable for the loss. Plants,
crops and pastures are additionally protected against damage caused by
trespassing domesticated animals, in which event the owner of such animals
would be held strictly liable in terms of the actio de pastu.

Person The importance of protecting persons against invasions of bodily integrity is

reflected in the fact that all the delictual actions focus in some way on protecting
bodily integrity. Compensation for financial expenditure incurred as a result of
bodily harm is given the widest scope for recovery in the fault-liability system. One
can claim under the actio legis Aquiliae in respect of both intentional and negligent
invasions, and also under the actio de pauperie for any such harm caused by
domesticated animals, in which case the owner is strictly liable.

One can also claim under the actio iniuriarum in instances where one has not
suffered financial harm, but only if the invasion of one’s bodily integrity was
intentional. The scope of protection in respect of an assault, for example, is
therefore less when there is no financial impact, i.e. no patrimonial harm. However,
because of the value that society places on people’s liberty, courts give greater
protection in instances involving unlawful deprivation of liberty. (The exact nature
of such protection – whether liability is strict or based on a variation of the intention
element – will be raised later.)

Personality (including
dignity, privacy, identity and
reputation)
Mere invasions, provided that they are done intentionally, may result in
compensation under the actio iniuriarum. The scope of liability under the actio
iniuriarum might be broadened in future, however. There are indications that courts
might favour liability based on negligence in privacy and defamation cases, in
which event the plaintiff’s personality interests would receive greater protection
than before, on a par with those patrimonial interests protected under the actio
legis Aquiliae.

However, where such invasions also result in patrimonial harm, i.e. financial loss,
one can claim additional compensation under the actio legis Aquiliae. In the latter
event, one would be protected against both intentional and negligent invasions.

Psyche (including the entire
spectrum of a person’s
psychological health and
mental tranquility)
This category ranges from injury to a person’s nervous system, through to the
physical pain experienced with physical injury, anxiety and mental distress, to the
reduction in enjoyment of life because of discomfort, inconvenience and
humiliation. Where infringements of such interests result in financial harm, the
actio legis Aquiliae offers the appropriate protection, while the Germanic remedy
for pain and suffering is the appropriate vehicle for compensating pain and
suffering, and loss of a person’s full pleasure of living.

If the injury was caused intentionally, one could also institute a claim under the
actio iniuriarum for invasion of bodily integrity.

Always bear in mind, however, that delict also protects the interests of defendants. Our society
acknowledges that in some instances a person can cause harm to another without having behaved
unlawfully; for example, when acting in self-defence or out of necessity, or when in an emergency
situation. Our society also values free enterprise and freedom of speech, so any compensatory
regime must protect the interests of persons who exercise their rights within lawful bounds.
The law of delict needs to recognise and balance the interests of both the plaintiffs and
defendants, and so, one can also look at these competing interests from another angle: the value that
society ascribes to them and the level of protection that society affords them. In short, the more
important society regards a plaintiff’s particular interest, or the more important that interest is when
compared to the interest of the defendant or other members of society, the easier the rules are for
claiming protection. The most important policy decision in this regard centres on the form of fault
that should be required in protecting particular interests. The nature of the fault depends upon
moral, social and economic considerations, and the value that society places on the competing
interests (for example, dignity versus freedom of expression), but more particularly on the nature of
the interest infringed.71 The policy-making decision goes along the following lines:72
Where society believes that a plaintiff’s interest far outweighs that of the defendant (or that
society’s interest in protecting the plaintiff outweighs its interest in protecting the
defendant), strict liability is imposed. Where the balance is more even, but the plaintiff’s
interest in receiving protection is still considered to be more important than the defendant’s
interest, liability is fault based, and includes both intention and negligence. In instances
where the conflicting interests are closely matched, liability is limited to deliberate or
intentional invasions. In short, the stronger the defendant’s, or society’s interest in
supporting the defendant’s situation, the narrower the fault element becomes.

1.11 Essential elements where liability is based upon fault

In delict one finds five core ‘umbrella’ elements of liability (Figure 1.2), irrespective of the
common-law action that underpins a particular claim. However, in delving deeper one finds that
these concepts have been modified to give expression to the purpose of each particular action, and
the nature of the interests that they are designed to protect. These elements might be expressed
differently to suit the particular action, but they are always there.

Figure 1.2 The five core elements of liability found in delict

The passage also frames the elements of a delict differently from the way in which they are
generally conceived. This is done deliberately, to introduce readers to the thought process required
when problem-solving delictual situations. The process is premised on the fact that the core
element is harm, for without harm, or threatened harm, to some interest, a plaintiff has no cause of
action.73 Our courts have endorsed this approach. In First National Bank of SA Ltd v Duvenhage 74
the Supreme Court of Appeal was favourably disposed towards the idea that harm is the proper
starting point for an enquiry into delictual liability, but the Constitutional Court was more emphatic
in H v Fetal Assessment Centre:75
‘[H]arm-causing conduct’ is a prerequisite for the further enquiry into the other elements of
delict, namely wrongfulness and fault. Without harm-causing conduct there is no conduct
which can be found to be wrongful or committed with the requisite degree of fault.

PAUSE FOR Must the elements of a delict arise contemporaneously?

REFLECTION For example, must the harm and the conduct element arise at the same
time? Consider whether a delict has been committed in the following two
situations:
• Andrew tells Susan that Michael is creditworthy. Two days later, Susan lends
Michael R1 000, to be returned on day 10. On day 10, Michael defaults and
tells her that he has no money and that Andrew knew of his poor financial
 situation all along. Susan wishes to sue Andrew because of the
misrepresentation. Even though the harm occurred 12 days after the conduct,
Susan could institute a claim in delict against Andrew.
• Sbu is a parish priest and Bongi and Lunga are parishioners. Bongi tells Lunga
that Sbu is a mendacious person. Lunga does not know what the word means
and politely smiles at the comment. That evening he looks up the word in a
dictionary and finds that it means ‘lying, untruthful and given to telling lies.’
When the statement was made there was no delict. All the elements were
present only when Lunga understood the meaning and import of the word. Sbu,
therefore, could have a claim against Bongi.

1.12 Essential elements where liability is strict

Strict liability means simply that liability arises independently of any fault on the part of the
defendant, and fault, whether in the form of intention or negligence, is stripped out as an essential
element. Liability in such cases stems from social policy, and does not reflect the notion that
wrongdoers must make good any harm that results from their actions. One can find instances of
strict liability in both the common law and in statutes.
One would naturally think that the requisites for liability should then be those elements that
remain when the fault element is eliminated. In some instances, this has been done, as was the case
when, for a while, media defendants were held strictly liable for defamatory statements that they
had published. These days, media liability is based upon fault. The Consumer Protection Act 68 of
200876 introduces strict liability in respect of certain defectively manufactured products.
There are, however, remnants of strict liability actions at common law, but different criteria
have evolved for justifying liability in a way to compensate for imposing liability without some
moral condemnation of the defendant’s behaviour. So, special rules have been devised for instances
of vicarious liability, where a particular type of relationship between the defendant and the
wrongdoer justifies liability, and liability for harm caused by animals, where ownership serves as a
justification. Under the actio de pauperie and the actio de pastu, owners are held liable, not
because of any fault on their part, but purely because they own the animal, provided that other
elements have also been met. However, these actions do not exclude the possibility of fault-based
liability. In some instances, where one can prove that the owner was at fault, it might be possible,
even more appropriate, to sue under the actio legis Aquiliae instead, or in the alternative. Similarly,
a person might be liable both vicariously and under the lex Aquilia.

1.13 A systematic approach to delictual problem-solving

The elements of delict are used as devices to balance the various countervailing interests of
plaintiffs, defendants and society and to provide a systematic approach for resolving delictual
problems.

1.13.1 Step 1: Identify the parties

The first step is to identify the possible plaintiffs and possible defendants, and how they would be
linked to the action. Imagine that one is in an attorney’s office. The client wishes to sue another
person in delict. The questions that the attorney should ask include:
• Is the client there because of harm suffered personally, or in a representative capacity (for
example, being the breadwinner), or both?
• How many plaintiffs are possible, given the particular set of facts?
• Who are the possible defendants?
• Who (or what, if an animal is involved) committed the delict?
• Were others involved as well?
• Is there someone else (for example, an employer or owner of an animal) who was not involved
in the actual commission of the delict, but who could be held liable nonetheless? In such
instances, bear in mind the principles of vicarious liability, and in the case of animals, claims
under the actio de pauperie or the actio de pastu against the animal’s owner. (An important
practical consideration when selecting possible defendants is not only to look for the person
who committed the delict, but also for the person who has the deepest pockets. There is no
point in suing a person of no means.)

1.13.2 Step 2: Primarily fact-based decisions

• Harm: There can be no delict without some legally recognised harm being present, either
realised harm or, for an interdict, potential harm. Focusing on the harm clarifies who the
plaintiff is and what he or she is seeking compensation for. (There may of course be more than
one plaintiff.) The nature of the harm also determines the nature of the action to use for
seeking a remedy, and the peculiar features of the other elements that must be proven, in
particular, the wrongfulness element and whether fault should take the form of intention or
negligence.
• Conduct: The conduct element identifies the possible defendants and what they are supposed
to have done to infringe the plaintiff’s interests. This identification also serves as a preliminary
step in isolating policy considerations relevant to the wrongfulness enquiry.
• Factual causation: There must be a link between the plaintiff’s harm and the defendant’s
conduct. There cannot be an untargeted claim against someone who was in no way involved in
the situation.

Note that the three elements mentioned are part of what are essentially fact-based enquiries. Courts
determine the issue by looking at the evidence presented and deciding whether the parties have
presented sufficient evidence to prove these elements. However, these elements also have some
normative features.

1.13.3 Step 3: Primarily normative decisions

The following elements are more normative than factual; in other words, the enquiries are more
value-based and, although centred within the factual matrix of each particular case, articulate a
wider societal policy perspective on whether there should be liability. These elements are not
proved directly through evidence. They amount to conclusions of law drawn from the evidence that
has been presented.
• Legal causation: It is not good enough merely to have a factual link between the conduct and
the harm. The link might be so tenuous that society would say that even though there is a
factual connection, it is not strong enough; or that other events had an overriding influence that
rendered the connection too distant, or too remote for liability to arise. So, society requires that
the causal link between the conduct and the harm should be sufficiently strong before liability
should ensue. The focus, then, is on the quality of the connection. Both the factual and
normative (legal) aspects of causation must be satisfied for this element to be present.
• Fault: Although there are exceptions, a central feature of the law of delict is that responsibility
for harm caused will not shift to another unless that person can be blamed in some way for the
violation of interest that occurred. This element has two sub-components: first, determine
whether the defendant has the capacity or mental maturity to be blamed, and then, if the person
can be blamed, look for the nature of the fault, which can take the form of either intention or
 negligence. In some instances, negligence is not sufficient. Society requires that the violation
must have been intentional before any loss-shifting can take place. However, there are also
instances where the opposite is true, and where liability arises even though the person was not
to blame, and so liability is strict. Again, the decision as to whether a person ought to be
blamed involves a value-laden judgement call.
• Wrongfulness: The crux of the wrongfulness enquiry has been formulated in various ways:
whether the plaintiff should be compensated for the loss that the defendant has caused;
whether it is reasonable to impose liability; whether society should allow a claim of this
nature; whether such a claim against the defendant is acceptable; or whether the violation of
the plaintiff’s rights was justified in some way. In this element, policy issues predominate and
it is the primary vehicle through which courts control the scope and ambit of delictual
liability.77 Although it is customary to talk about wrongful conduct, the enquiry does not focus
on the defendant’s conduct alone. The enquiry also looks at the nature of the defendant’s fault
and the consequences, and it is a combination of the nature of the fault, consequences and the
nature of the conduct that determines the policy factors to consider in reaching a conclusion.
Once all five elements have been satisfied (factual and legal causation are actually sub-components
of one element), one can conclude that a delict has been committed, and that circumstances are
such that responsibility for the plaintiff’s harm should shift to the defendant.

1.13.4 Step 4: Determining the remedy

In the same way as a criminal trial moves on to the sentencing stage after conviction, a delictual
enquiry moves on to awarding the appropriate remedy after determining liability. In most instances,
the claim is for damages, the nature and amount of which now have to be determined, but remedies
can also include an interdict or an apology.

1.13.5 Step 5: Apportionment of damages

Where the remedy takes the form of damages, a final step in the process might have to be
undertaken: that of apportioning the damages. The issue here is whether the responsibility for the
plaintiff’s harm should shift to the defendant in its entirety, or whether such responsibility ought to
be shared in some way. In some instances, where there is contributory negligence, fairness
demands that the plaintiff should remain responsible for at least a portion of his or her loss, while in
other instances, the loss might be shared among two or more defendants. It is only when this stage
is finalised that the delictual loss-allocation exercise is complete.
This approach to resolving delictual problems is summarised in Figure 1.3 on the next page.

PAUSE FOR Resolving delictual problems

REFLECTION Use the five problem-solving steps to determine whether the elements of
a delict are present in the following instance:
Katlego is paralysed because he did not receive timeous medical
treatment after being injured in an accident that he caused by driving too
fast. Nomalundi, a medical doctor who was on holiday, had arrived on the
scene but refused to treat Katlego because she did not have any
protective gloves with her. Katlego can prove that had Nomalundi
assisted him immediately, he would not have been paralysed. Also,
Katlego does not suffer from Aids and is not HIV-positive.
1.14 General principles modified for specific actions
The general principles of liability set out so far are refined and expressed differently when
considering specific types of delictual liability. Below, we describe the main requirements for
liability under each of the actions – the actio legis Aquiliae, the Germanic remedy for pain and
suffering, the actio iniuriarum, the actio de pauperie, the actio de pastu and vicarious liability.

1.14.1 The actio legis Aquiliae

• Harm or loss: Patrimonial loss, which is financial harm arising from physical damage to person
or property, or loss that is purely economic
• Conduct: Either a positive act, an omission, or a statement
• Causation: Factual causation, in that the conduct must have been a conditio sine qua non of the
loss, and legal causation, in that the link between the defendant’s conduct and the plaintiff’s
harm must not be too tenuous
• Fault: Blameworthiness in the form of dolus (intention) or culpa (negligence). However, to be
blameworthy, people have to be accountable for their conduct
• Wrongfulness: Wrongfulness deals with the question whether or not the situation in which the
defendant’s culpable conduct caused the harm is legally reprehensible to the extent that
liability should be imposed on the defendant. To determine wrongfulness, the circumstance is
measured against the criterion of reasonableness: if the culpable causing of harm is objectively
unreasonable, or without lawful justification (a valid defence justifies the infliction of harm),
then it is in the circumstances reasonable to impose liability. In determining whether conduct
is wrongful in the circumstances, all the other elements (conduct, harm, causation and fault)
are considered.
Figure 1.3 Steps in resolving delictual problems

1.14.2 The Germanic remedy for pain and suffering

• Harm or loss: Intangible, non-patrimonial harm associated with personal injury to the plaintiff,
for example; actual pain, psychiatric injury, loss of amenities of life and loss of life expectancy
• Conduct: Either a positive act, an omission, or a statement
• Causation: Factual causation, in that the conduct must have been a conditio sine qua non of the
loss, and legal causation, in that the link between the defendant’s conduct and the plaintiff’s
harm must not be too tenuous
• Fault: Blameworthiness in the form of dolus (intention) or culpa (negligence). However, to be
blameworthy, people have to be accountable for their conduct
• Wrongfulness: To determine wrongfulness, the circumstances are measured against the
criterion of reasonableness: if the culpable causing of harm is objectively unreasonable, or
without lawful justification (a valid defence justifies the infliction of harm), then it is in the
circumstances reasonable to impose liability. In determining whether conduct is wrongful in
the circumstances, all the other elements (conduct, harm, causation and fault) are considered.

PAUSE FOR The Germanic remedy for pain and suffering

REFLECTION A father and mother suffer enormous grief because their child has been
killed by a drunken driver. The mother suffers a nervous breakdown after
the funeral and is hospitalised. Would both of them be able to claim
damages for the emotional grief and shock they have experienced?

1.14.3 The actio iniuriarum

• Harm or loss: Non-patrimonial harm in the form of a violation of a personality interest, usually
classified under the headings of corpus (bodily integrity), dignitas (dignity) and fama
(reputation)
• Conduct: Usually statements or positive conduct, seldom an omission
• Causation: Normally not an issue, but may become one in some instances, as in deprivation of
liberty cases
• Fault in the form of intention (animus iniuriandi): This is a controversial element as some
contend that the action has developed to include some instances (such as those involving
deprivation of liberty) where liability is strict, and others (such as defamation involving the
media) where liability is negligence based
• Wrongfulness: To determine wrongfulness, the circumstances are measured against the
criterion of reasonableness: if the culpable causing of harm is objectively unreasonable, or
without lawful justification (a valid defence justifies the infliction of harm), then it is in the
circumstances reasonable to impose liability. In determining whether conduct is wrongful in
the circumstances, all the other elements (conduct, harm, causation and fault) are considered.

COUNTER An action for iniuria

POINT Melius de Villiers notes78 three essential requisites to establish an action
for iniuria. They are:
1. An intention on the part of the offender to produce the effect of his act
2. An overt act, which the person doing is not legally competent to do, and which
at the same time is:
3. An aggression upon the right of another, by which aggression the other is
aggrieved and which constitutes an impairment of the person, dignity or
reputation of the other.

In Delange v Costa79 the Court commented on the elements of the actio

iniuriarum where a plaintiff’s dignity was alleged to have been impaired:

… Logically in an action for injuria one should commence

by enquiring into the existence of the second of these
requisites, viz whether there has been a wrongful overt act.
(It is more common, and probably juristically more correct,
to speak of a ‘wrongful’ rather than an ‘unlawful’ act.) A
 wrongful act, in relation to a verbal or written
communication, would be one of an offensive or insulting
nature. Once the wrongfulness of such act has been
determined animus injuriandi will be presumed … . It would
be open to the defendant to rebut such presumption by
establishing one of the recognised grounds of justification.
If the defendant fails to do so, the plaintiff, in order to
succeed, would have to establish the further requirement
that he suffered an impairment of his dignity. This involves
a consideration of whether the plaintiff’s subjective feelings
have been violated, for the very essence of an injuria is that
the aggrieved person’s dignity must actually have been
impaired. It is not sufficient to show that the wrongful act
was such that it would have impaired the dignity of a
person of ordinary sensitivities. Once all three requisites
have been established the aggrieved person would be
entitled to succeed in an action for damages, subject to the
principle of de minimis non curat lex.

This passage contradicts what we have postulated in this chapter in a

number of ways. First, it reverses the order of the enquiry, as we suggest
that logically one should start with ‘the very essence of an injuria’,
whether the dignity has in fact been subjectively impaired by the
defendant’s conduct. It is only then that one needs to determine
objectively whether the culpable infliction of harm has been wrongful. The
principle of de minimis non curat lex is part of this second aspect of the
enquiry as it is an expression of social policy that indicates that trivial
infringements should not be considered wrongful. Lastly, a ground of
justification does not rebut the presumption of intention: it is a defence
that serves to justify a situation that would otherwise have been
considered wrongful. There are defences, such as mistake, that rebut the
presumption of intention, but these do not justify a person’s fault; they
serve to eliminate fault altogether. One can avoid these conceptual traps
by adhering to the problem-solving approach described in this chapter.

1.14.4 The actio de pauperie

The requirements for liability under the actio de pauperie are:
• Pauperies (harm)
• Conduct on the part of a domesticated animal
• A causal link between the conduct and the harm
• Unnatural conduct, termed contra naturam sui generis, or conduct that stems from inward
excitement or vice (sponte feritate commota), which amounts to a conclusion that society
considers the animal’s conduct to be actionable
• Ownership of the animal at the time of the injury.

A number of defences are normally associated with the actio. These include:
• The plaintiff was to blame
• There was fault on the part of a third person
• There was fault on the part of the controller of the animal
• The conduct on the part of another animal
• Vis maior, outside influence, or an extraneous event not associated with the conduct of any
person
• Volenti non fit iniuria
• The plaintiff (or the injured animal) was not lawfully at the place of injury.

These defences negate any conclusion that the animal’s conduct was contra naturam sui generis.

1.14.5 The actio de pastu

The requirements for liability under the actio de pastu are:
• Harm in the form of damage to plants, crops or pastures through the process of grazing
• Conduct in that a domesticated animal must have trespassed on the plaintiff’s property and
grazed on the plants
• A causal link between the conduct and the harm
• The animal must have acted of its own volition
• Ownership of the animal at the time of the damage.

The defences include:

• The plaintiff was to blame
• Vis maior, outside influence, or an extraneous event not associated with the conduct of any
person
• Volenti non fit iniuria.

1.14.6 Vicarious liability

Vicarious liability arises where one person is indirectly liable, without fault on his or her part, for
the delict of another. Accordingly, this form of liability is strict, and arises in addition to any
personal liability that the wrongdoer incurs. Three elements are usually required:
• A delict: Someone must have committed a delict, in that the delictual elements set out have
been met.
• A relationship: A relationship between the wrongdoer and the defendant, which the law
recognises as being of such a nature as to warrant the imposition of liability. This relationship
is usually an employment relationship, but can also include other relationships that could be
said to be akin to that of employer and employee, for example, mandate and partnership.
• Course and scope: The delictual conduct must fall within the ambit of the defendant’s
instructions, be for the defendant’s benefit, or fall within the risk created by the defendant
when establishing the relationship with the wrongdoer.

So for liability to arise, there should be both a vertical link between the defendant and the
wrongdoer and a horizontal link between the wrongdoer and the plaintiff (Figure 1.4).
Figure 1.4 Vicarious liability

1.15 Conclusion
In this chapter, we lay the foundation for what follows. The issues raised in this chapter are largely
taken for granted when applying principles to factual situations. These are things we all are
supposed to know, but seldom openly confront. Nor do we pertinently assess, in every instance,
whether our application of principles is in line with our inherent assumptions. So, when reading
further, assess whether what follows conforms to, or amplifies what has been expressed in this
chapter, and, as your knowledge of delictual principles deepens, question the assumptions and
conclusions that this chapter has drawn. Although the fundamentals have been proven to stand the
test of time, the principles of delict, after all, are vibrant, living principles and should reflect
contemporary values.

1 Constitution of the Republic of South Africa, 1996 (hereafter ‘the Constitution’).

2 Murphy and Whitting Street on Torts 13 ed (2012) at 4–5.
3 McKerron The Law of Delict: A Treatise on the Principles of Liability for Civil Wrongs in the Law of South Africa 7
ed (1971) at 1.
4 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 26–27.
5 [1932] AC 562; 1932 SC (HL) at 580.
6 Cane The Anatomy of Tort Law (1997) at 51–52.
7 Deakin, Johnston and Markesinis Markesinis and Deakin’s Tort Law 7 ed (2013) at 48–50.
8 Deakin, Johnston and Markesinis (2013) at 49.
9 Linden and Feldthusen Canadian Tort Law 9 ed (2011) at 15.
10 Holmes The Common Law (1923) at 50.
11 Herschel v Mrupe 1954 (3) SA 464 (A) at 494; Telematrix (Pty) Ltd t/a Matrix Vehicle Tracking v Advertising
Standards Authority SA 2006 (1) SA 461 (SCA) para 12. See also Roux v Hattingh 2012 (6) SA 428 (SCA) para 25;
SA Hang & Paragliding Association v Bewick 2015 ([2015] 2 All SA 581 (SCA); 2015 (3) SA 449 (SCA) paras 31
–33; Home Talk Developments (Pty) Ltd v Ekurhuleni Metropolitan Municipality [2017] 3 All SA 382 (SCA) para 1.
12 Van der Merwe and Olivier Die onregmatige daad in die Suid-Afrikaanse reg 6 ed (1989) at 1.
13 Our translation.
14 Boberg (1984) 16 at Notes.
15 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 4 fn 8.
16 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 2.
17 Cane Atiyah’s Accidents, Compensation and the Law 8 ed (2013) at 403–455.
18 Deakin, Johnston and Markesinis (2013) at 43.
19 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC) para 17.
20 Cane (2013) at 461.
21 Cane (2013) at 461–496; Deakin, Johnston and Markesinis (2013) at 51–59.
22 Ison The Forensic Lottery (1967); Elliot and Street Road Accidents (1968); Atiyah Accidents, Compensation and the
Law (1970).
23 Cane (2013) at 459.
24 The Accident Compensation Act of 1972, which has since been replaced by the Accident Compensation Act of 2001.
25 Cane (2013) at 459.
26 Deakin, Johnston and Markesinis (2013) at 3–14.
27 Cane (2013) at 15–16.
28 Report of the Royal Commission on Civil Liability and Compensation for Personal Injury (Cmnd 7054, 1978).
29 The discrepancy seems to lie in the definition: the former figures relate to delictual claims in general, the latter to
personal accident claims only.
30 Van Jaarsveld v Bridges 2010 (4) SA 558 (SCA).
31 Stewart An Introduction to the Scots Law of Delict (1989) at 2.
32 Deakin, Johnston and Markesinis (2013) at 43.
33 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC) and Dikoko v Mokhatla 2006 (6) SA 235 (CC). See also
Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 76 where Yacoob J said that ‘It is not appropriate for civil claims to be brought to
court to punish children.’
34 Linden and Feldthusen Canadian Tort Law 9 ed (2011) at 13–16.
35 Linden and Feldthusen (2011) at 15.
36 Linden and Feldthusen (2011) at 7.
37 Deakin, Johnston and Markesinis (2013) at 44.
38 Cane (1997) at 211.
39 Cane (1997) at 230.
40 Linden and Feldthusen (2011) at 5.
41 Cane (1997) at 231.
42 Cane (1997) at 231.
43 Linden and Feldthusen (2011) at 5.
44 (1996) 71 ALJR 107.
45 (1996) 71 ALJR 107 at 112.
46 Chartaprops 16 (Pty) Ltd v Silberman 2009 (1) SA 265 (SCA) para 37.
47 1999 (2) SA 147 (SCA) at 154.
48 Linden and Feldthusen (2011) at 9.
49 Tsichlas v Touch Line Media (Pty) Ltd 2004 (2) SA 112 (W) (website); Delta Motor Corporation (Pty) Ltd v Van der
Merwe [2004] 4 All SA 365 (SCA), 2004 (6) SA 185 (SCA) (email); Le Roux v Dey (Freedom of Expression Institute
and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577 (CC); 2011 (3) SA 274 (CC) (digital
manipulation circulated via mobile phone); Dutch Reformed Church Vergesig v Sooknunan 2012 (6) SA 201 (GSJ)
(website); Cele v Avusa Media Ltd [2013] 2 All SA 412 (GSJ) (digital manipulation in newspaper); Heroldt v Wills
2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ) (Facebook); Isparta v Richter 2013 (6) SA 529 (GNP)
(Facebook); Ketler Investments CC t/a Ketler Presentations v Internet Service Providers’ Association 2014 (2) SA
569 (GJ) (website); African National Congress v Democratic Alliance 2014 (3) SA 608 (GJ) (SMS message); RM v
RB 2015 (1) SA 270 (KZP) (Facebook).
50 Protea Technology Ltd v Wainer [1997] 3 All SA 594 (W) (electronic monitoring).
51 McKerron (1971) at 10.
52 1911 AD 657.
53 At 665–666. (References omitted.)
54 Neethling and Potgieter (2015) at 8.
55 Van der Walt and Midgley Principles of Delict 4 ed (2016) at ix. (Preface to the third edition).
56 Murphy and Whitting Street on Torts 13 ed (2012) at 5–6.
57 Neethling and Potgieter (2015) a 4.
58 Neethling and Potgieter (2015) at 5.
59 The German BGB para 823, for example, states:
A person who intentionally or negligently unlawfully injures the life, body, health, freedom, property
or another right of another person is liable to make compensation to the other party for the damage
arising from this.
Article 1382 of the French Code Civil states:
 All human conduct of any kind which causes harm to another requires the person by whose fault it
occurred to redress it.
(Translations in Raymond Youngs English, French and German Comparative Law (1998) at 222–223.)
60 Cane (1997) at 9.
61 Fleming The Law of Torts 9 ed (1998) at 8.
62 Cane (1997) at 1.
63 Cane (1997) at 5.
64 Perlman v Zoutendyk 1934 CPD 151 at 155. See also Maisel v Van Naeren 1960 (4) SA 836 (C); Minister of Finance
v EBN Trading (Pty) Ltd 1998 (2) SA 319 (N).
65 See, for example, Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd 2006 (3) SA 138 (SCA) paras
10–12, where the Court emphasised negligence in formulating the wrongfulness criterion. In such instances, the duty
is a duty not to act negligently. Although the Court did not expressly say so, it follows that where fault takes the form
of intention the duty would have to be reformulated, presumably, as a duty not to cause harm intentionally.
66 Neethling and Potgieter (2015) 4–5 fn 14 (references omitted).
67 See also section 1.11 below (‘Essential elements where liability is based upon fault’).
68 Murphy and Whitting (2012) at 5.
69 In an earlier edition (John Murphy Street on Torts 11 ed (2005) at 4) this passage was followed by the following
extract:
The starting point of any analysis of the law of torts must be a consideration of those rights and
interests which the law of tort protects. It is apparent that one of the functions of the law of torts is the
protection of what are popularly known as human rights.
70 Cane (1997) at 66.
71 Midgley ‘Intention Remains the Fault Criterion under the Actio Injuriarum’ (2001) 118(3) SALJ at 433. See also Van
der Walt and Midgley Principles of Delict 4 ed (2016) para 27.
72 Midgley ‘Fault under the actio iniuriarum: Custer’s last stand?’ in Boezaart and de Kock Vita Perit, Labor non
Moritur, Liber Memorialis: PJ Visser (2008) 187 at 201.
73 … a wrong is called a tort only if the harm which has resulted, or is about to result from it, is capable of being
compensated in an action at law for damages, although other remedies may also be available.
(Keeton (Ed) Prosser and Keeton on the Law of Torts 5 ed (1984) at 4.)
74 [2006] 4 All SA 541 (SCA); 2006 (5) SA 319 (SCA) para 1; Knobel ‘Die Samehang Tussen Onregmatigheid en
Skade’ (2005) 68(4) THRHR at 645. See also the comment on Street’s approach to tort, quoted previously (at section
1.10 ‘Protected interests’).
75 2015 (2) BCLR 127 (CC); 2015 (2) SA 193 (CC) para 54. See also para 60.
76 Section 61.
77 Home Talk Developments (Pty) Ltd v Ekurhuleni Metropolitan Municipality [2017] 3 All SA 382 (SCA) para 20.
78 De Villiers (1899) The Roman and Roman-Dutch Law of Injuries: A Translation of Book 47, Title 10, of Voet’s
Commentary on the Pandects at 27.
79 [1989] 2 All SA 267 (A); 1989 (2) SA 857 (A) at 861B–F.
Chapter 2

Delict and the Constitution

2.1 Introduction

2.2 The relationship between the Constitution and the law of delict
2.2.1 The constitutional remedy
2.2.2 Could constitutional damages constitute ‘appropriate relief’?
2.2.3 Does the infringement of a constitutional right amount to a delict?
2.2.4 Do delictual damages constitute ‘appropriate relief’ for the violation of a
constitutional right?

2.3 Direct application and

2.3 Direct application and indirect application of the Constitution

2.3.1 Direct application
2.3.2 Indirect application

2.4 |
Constitutional values and norms
2.4.1 Foundational values that relate to dignity and equality
2.4.2 Foundational values associated with democracy and governance
2.4.3 Foundational values associated with freedom of expression
2.4.4 The values of non-violence and the protection of vulnerable people from
violence and abuse

2.5 The Constitution’s areas of impact on the law of delict

2.5.1 Adjusting the way in which an established legal rule is applied
2.5.2 Changing an existing legal rule
2.5.3 Introduction of a new legal rule
2.5.4 Application of an established legal rule to a new set of facts

2.6 Conclusion
2.1 Introduction
Constitutions play an important role in regulating public power, and so a constitution is often
thought of as a citizen’s shield against abuse of governmental power. The Constitution of the
Republic of South Africa, 1996 (‘the Constitution’), is more than that, however. Not only is it the
supreme law of the country, but it is central to our entire legal system. It determines the validity of
all of the country’s laws,1 including the law of delict, and any delictual principle or application of a
delictual principle that does not conform to constitutional standards is invalid. When looking at the
relationship between the Constitution and the law of delict, two key concepts stand out: the
Constitution as a source of fundamental rights, and the Constitution as a source of fundamental
values. Although both impact on the law of delict, it is the latter concept that has the dominant
influence.
In this chapter, we will examine the relationship between the Constitution and the law of
delict, the theoretical framework underpinning the application of the Bill of Rights to delictual
situations and illustrate how the Constitution has influenced the law of delict. We also examine
some significant questions that arise from the relationship between the Bill of Rights and the law of
delict.

2.2 The relationship between the Constitution and the law

of delict

2.2.1 The constitutional remedy

Generally, a threat to ‘or the infringement’ of a constitutional right warrants a constitutional
remedy in the form of ‘appropriate relief’. 2 The Constitution is silent on the meaning of the term,
but the Constitutional Court has stated that, essentially, ‘appropriate relief’ denotes relief that is
required to effectively protect and enforce the Constitution, for ‘without effective remedies for
breach, the values underlying and the rights entrenched in the Constitution cannot properly be
upheld or enhanced’. 3 Determining what type of remedy may be deemed appropriate depends on
the circumstance of each particular case 4 and the object that ought to be achieved.5

2.2.2 Could constitutional damages constitute ‘appropriate relief’?

Over and above the delictual damages which the plaintiff sought to be awarded in Fose v Minister
of Safety and Security,6 he also attempted to recover constitutional damages – which included ‘an
element of punitive damages’. The Constitutional Court took great care to emphasise the narrow
ambit of the question that fell to be decided in that specific case, namely: in addition to delictual
damages claimed for the assaults, was the plaintiff entitled to recover constitutional damages,
which included an element of punitive damages? 7
Although this question was answered in the negative, the Court held that, in principle,
constitutional damages may be awarded where it is necessary to protect and enforce the Bill of
Rights.8 Despite the Court’s confirmation that constitutional damages could be awarded, awards in
subsequent cases have been infrequent. In Modderfontein Squatters, Greater Benoni City Council v
Modderklip Boerdery (Pty) Ltd (Agri SA and Legal Resources Centre, Amici Curiae); President of
the Republic of South Africa v Modderklip Boerdery (Pty) Ltd (Agri SA and Legal Resources
Centre, Amici Curiae) 9 the Supreme Court of Appeal held that the owner of a farm was entitled to
constitutional damages for the loss of the use of his land during the period in which it had been
occupied by squatters and that the State had failed to provide alternative land for the occupiers. The
Court emphasised that the return of the land was not a feasible option and that, on the particular
facts, no other remedy was apparent.10 The Constitutional Court upheld this decision.
In MEC, Department of Welfare, Eastern Cape v Kate11 the respondent sought payment of an
outstanding amount of a social disability grant that was awarded to her as well as the interest
thereon. The MEC ultimately paid the remaining outstanding R13 000 but disputed liability in
respect of the interest on the R13 000. The High Court awarded her the interest as a measure of
constitutional damages for the department’s unreasonable delay in considering the application,
which had resulted in the respondent’s being deprived of her constitutional right to receive a social
grant during that time and for which she was entitled to be recompensed. The Supreme Court of
Appeal confirmed the award, noting its appropriateness in the particular circumstances of the case.

PAUSE FOR When is it appropriate to award constitutional damages?

REFLECTION In MEC, Department of Welfare, Eastern Cape v Kate,12 while considering
whether constitutional damages ought to be paid to the respondent,
Nugent JA remarked:13

The question that submission raises is not so much

whether the remedy that is now proposed is an appropriate
one to remedy Kate’s loss, but rather whether a
constitutional remedy should be granted at all. No doubt,
the infusion of constitutional normative values into
delictual principles itself plays a role in protecting
constitutional rights, albeit indirectly. And, no doubt,
delictual principles are capable of being extended to
encompass State liability for the breach of constitutional
obligations. But the relief that is permitted by s 38 of the
Constitution is not a remedy of last resort, to be looked to
only when there is no alternative and indirect means of
asserting and vindicating constitutional rights. While that
possibility is a consideration to be borne in mind in
determining whether to grant or to withhold a direct s 38
remedy, it is by no means decisive, for there will be cases
in which the direct assertion and vindication of
constitutional rights are required. Where that is so, the
further question is what form of remedy would be
appropriate to remedy the breach.

However, subsequently, in Minister of Police v Mboweni,14 the Court

seems to have taken a different view. In this case, the deceased, Mr
Mahlati, was assaulted by inmates following his arrest and detention, as
a result of which he eventually died. The mothers of Mahlati’s two
daughters claimed damages from the Minister of Police for the
infringement of their children’s right to parental care as encapsulated by
section 28(1)(b) of the Constitution. The trial court found, among other
things, that a child whose parent had died because of unlawful conduct of
a third party, could claim constitutional damages for infringement of the
right to parental care.
The Supreme Court of Appeal overturned the judgment, however. It
stated, with reference to Fose v Minister of Safety and Security,15 that the
preferred approach in determining whether a court should award
constitutional damages should be:
• A court should first consider whether there is an existing remedy that would
adequately vindicate and protect the constitutional right.
 • If the existing remedy appears to be inadequate, then it should consider
whether such deficiency can be remedied by a development of the common
law. The Court emphasised the flexible nature of the law of delict, which makes
it particularly susceptible for development in line with the spirit, purport and
objects of the Bill of Rights. A necessary consideration in this regard is whether
the infringement of the constitutional right may be appropriately vindicated
through resort to public-law remedies.
• Lastly, in considering whether constitutional damages should be awarded as
appropriate relief, the broader societal implications which such judgment may
have should be taken into account. The Court said:16

[I]n determining a suitable remedy, the courts are obliged to

take into account not only the interests of parties whose
rights are violated, but also the interests of good
government. These competing interests need to be
carefully weighed.

2.2.3 Does the infringement of a constitutional right amount to a delict?

It is conceivable that an infringement of a fundamental right may also amount to an infringement of
a subjective right which the law of delict recognises and seeks to protect.17 For example, when
someone is assaulted, the wrongdoer’s culpable conduct wrongfully causes an infringement of the
victim’s common-law right to personality (his or her corpus, fama and dignitas) as well as the right
to bodily and psychiatric integrity. The victim may institute a variety of common-law remedies
against the wrongdoer, including the actio iniuriarum, the actio legis Aquiliae and the action for
pain and suffering. The wrongdoer’s assault at the same time also violates the victim’s
constitutional rights to dignity, life, freedom and security of the person and privacy.
It should be noted, however, that, even though there might be instances of overlap, a violation
of a constitutional right does not in itself simultaneously constitute a delict. For that to happen, the
elements of a delict should be satisfied independently.

Dendy v University of Witwatersrand18

Dendy had applied, unsuccessfully, for the appointment to a chair of law at the University of
Witwatersrand. He instituted two claims:
• The first was based on alleged procedural irregularities that took place during the process of
making the new appointment, which, he argued, had violated a variety of his constitutional
rights and, as such, infringed his right to dignity in terms of section 10 of the Constitution
and/or the common law.19
• The second claim concerned the University’s alleged failure to furnish him with reasons for
his unsuccessful application and the selection committee’s minutes. This, he argued, also
amounted to a violation of some of his constitutional rights and, accordingly, infringed his
right to dignity in terms of section 10 of the Constitution and/or the common law.20

In other words, the plaintiff argued that the violation of his constitutional rights gave rise to an
infringement of his dignity which, in turn, entitled him to damages as an appropriate remedy.
The plaintiff’s claims were novel and he therefore requested the High Court to develop the
common law so as to afford him a remedy sounding in damages. To achieve this, he
proposed, among other things, that the Court should recognise all violations of constitutional
rights as actionable infringements of dignity.21 The Court responded:22

In effect, what is contended for is the creation of a constitutional delict. There

are substantial reasons not to afford recognition to such a delict. It is desirable
that a clear distinction be drawn between delictual and constitutional wrongs.
 Conceptual difficulties are bound to arise if one were to equate all infringements
of fundamental rights with an ordinary delict. There is the problem of
overlapping and possible conflict between fundamental rights entrenched in the
Constitution and private subjective rights protected by, or legal duties imposed
by, the law of delict. Where the infringement of a fundamental right overlaps with
generally recognised areas of delictual liability, an ordinary delictual claim will
lie at the instance of an aggrieved person. The problem lies with those
infringements of fundamental rights that extend beyond the recognised ambit of
the law of delict and which do not meet the requirements of delictual liability …
To recognise all constitutional violations as infringements of dignity as
suggested by the plaintiff would be to confuse the wider concept of dignity
under the Constitution with the narrower concept of dignitas. Yet a further
reason why all violations of fundamental rights ought not to be regarded as
actionable infringements of dignity, is that in many instances where the violation
of human dignity is offended the primary constitutional breach may be of a more
specific right in respect of which the Constitution affords a particular remedy or
specific protection. In these circumstances dignity serves merely as a flexible
and residual right.

The Supreme Court of Appeal confirmed the High Court decision. The Court found for there
to have been a common-law violation of Dendy’s right to dignity, the violation of his interest
had to be subjectively and objectively insulting and humiliating, and that these requirements
were not present on the facts of the case.23

2.2.4 Do delictual damages constitute ‘appropriate relief’ for the violation

of a constitutional right?
As indicated above, in some instances the remedy against individuals is a constitutional one, but
where there is an overlap between a fundamental right and a private-law (subjective) right that the
law of delict recognises, a delictual remedy is possible. Courts are unlikely, except in extreme cases
in which it might be necessary to note judicial displeasure, to award constitutional damages and
delictual damages on the same facts.24 Where there is an overlap, a delictual remedy usually serves
as a constitutional remedy. For example, an award of damages for unlawful arrest and detention not
only assuages the infringement of a person’s personality right, but it also provides the necessary
relief in respect of the infringement of a person’s fundamental right to freedom and security.25
Where there is no such overlap, courts may award damages as a constitutional remedy, but not in
delict.
Fose v Minister of Safety and Security26
Fose sued the Minister for damages arising out of a series of assaults perpetrated by police
officers who had acted within the course and scope of their employment. More particularly,
the plaintiff instituted the Aquilian action, the actio iniuriarum and the action for pain and
suffering in respect of the past and future medical expenses, contumelia, pain and suffering
and loss in the amenities of life.
The Constitutional Court had to determine whether the assaults, which constituted a
breach of certain constitutional rights,27 could be vindicated by the delictual remedy of
damages. The respondent Minister argued that any person who applied to court for
appropriate relief for an infringement of a fundamental human right under the interim
Constitution could not resort to a delictual remedy, because the applicant was entitled only to
a constitutional remedy.28 The Court held that, in principle, ‘appropriate relief’ was aimed at
protecting and enforcing the Bill of Rights and that there was no reason in principle why the
concept should not also include an award of damages where such an award was necessary
to protect and enforce constitutionally recognised fundamental rights. Specifically, the Court
emphasised that the law of delict was flexible and that, in many cases the common law
 would be broad enough to provide all the relief that would be appropriate for a breach of
constitutional rights.29 The Court found that the delictual damages awarded to the applicant
successfully protected, enforced and vindicated his constitutional rights and that an
additional award of constitutional damages would be inappropriate.

Since Fose, courts have continued to award delictual damages where a plaintiff’s constitutional
rights have been infringed, but typically, this occurs where the plaintiff’s common-law rights have
also been infringed.

2.3 Direct application and indirect application of the

Constitution
The Constitution differentiates between direct and indirect application of the Bill of Rights. The
direct application refers to a scenario in which a plaintiff relies directly on the infringement of a
fundamental right in the Constitution when formulating his or her cause of action. Indirect
application denotes a situation where a plaintiff looks towards the common law to protect the
infringed right, but the latter is then influenced by a fundamental right or an underlying
constitutional value when determining the nature of the common-law right or duty, or in providing
an effective remedy. Direct application is achieved via section 8(2) while indirect application
occurs by virtue of section 39(2) of the Constitution.
The distinction between the direct and indirect application of the Constitution has had little
practical effect when delictual disputes have been adjudicated. Since delictual rules are primarily
common-law based and it is much easier to achieve harmony by developing the common law, the
indirect application predominates. In the main, courts have assessed the law of delict ‘through the
prism of the Constitution and in relation to its values’30 and the approach has been that ‘where the
common law deviates from the spirit, purport and objects of the Bill of Rights, the courts have an
obligation to develop it by removing that deviation’.31

PAUSE FOR Direct and indirect application

REFLECTION Van der Walt and Midgley suggest that the difference between direct and
indirect is predominantly philosophical in nature:32

[I]n following the direct application, one approaches the

issue from the point of view that the Constitution is
supreme overall law, while the indirect approach views the
Constitution and the common law as being complementary,
the role of the Constitution being to nudge and influence
common-law principles rather than to override them.

While their description of the nature of the indirect application might be

apposite, is the authors’ distinction correct? Is the Constitution not in all
instances the ‘supreme overall law’?

2.3.1 Direct application

The Constitution’s direct application to the law of delict is effected primarily through the
application of section 8, but sections 36(1) and 39(3) are also relevant in some instances. The
Constitution explicitly states that its provisions apply to the conduct of natural persons and juristic
persons in appropriate cases,33 and so the Bill of Rights also serves to protect persons not only from
infringements of their rights by the State, but also in respect of infringements by private
individuals.

Khumalo v Holomisa34
This case was decided in terms of the provisions of the interim Constitution. Subsequently the
drafters of the Constitution dealt with some of the issues raised in this case, but certain
statements of principle in this case remain relevant.
The applicant asserted that the constitutional right to freedom of expression set out in
section 16 of the interim Constitution was directly applicable to the dispute between two
private individuals; and, further, that it necessitated the common law of defamation to be
developed so that a plaintiff in a defamation action against a defendant, who is a member of
the mass media, is required to plead and prove that the defamatory statement made by the
defendant was false. The Constitutional Court agreed that the Bill of Rights had so-called
direct horizontal application, but denied the applicant’s request to develop the common law to
give effect to section 16, because the common law relating to defamation was already in line
with constitutional requirements.
The Court provided some guidelines in respect of the potential future direct application of
the Bill of Rights:
• Section 8(2) bound natural and juristic persons to the extent that it is applicable, taking into
account the nature of the right and the nature of any duty imposed by the right. It added that,
once it has been determined that a natural person is bound by a particular provision of the
Bill of Rights, section 8(3) then provides that a court must apply and, if necessary, develop
the common law to the extent that legislation does not give effect to the right.35
• Development of the common law would be necessary if it was found that the common law is
inconsistent with the Constitution. This would be the case where, for example, the common
law rule in question amounts to an unjustifiable limitation of any right contained in the Bill of
Rights. In determining this issue, ‘sight must not be lost of other constitutional values and, in
particular, the value of human dignity’.36
• Rules of the common law may be developed so as to limit a right, as long as that limitation
would be consistent with the provisions of section 8(3)(b).37
• Some rights might not be capable of horizontal application.38

Unfortunately, it is not clear from this judgment exactly how the application of section 8(2)
would differ from the application of section 39(2), which also suggests that courts are required
to develop the common law in order to give effect to the spirit, purport and objectives of the
Bill of Rights.

The question is to what extent a constitutional right is actionable in delict in the sense that (a) it
might form the basis of a plaintiff’s claim, or (b) it should be given effect to when applying a
delictual rule to the facts. A second question is to what extent a constitutional right might protect a
defendant against a delictual claim.
We can reasonably easily find a delictual counterpart for the following fundamental rights:
human dignity (section 10), life (section 11), freedom and security of the person (section 12),
privacy (section 14), freedom of assembly, demonstration, picket and petition (section 17), freedom
of trade, occupation and profession (section 22), environment (section 24), and property (section
25). In these instances, overlaps are likely, and the fundamental rights confirm that an action in the
law of delict conforms to constitutional values. This adds weight to a contention that the legal
convictions of the community would lean towards granting a remedy in delict. Of course, in
particular instances there might be countervailing rights that could point the other way, for
example, dignity (section 10), freedom of religion, belief and opinion (section 15), freedom of
expression (section 16), assembly, demonstration, picket and petition (section 17), and freedom of
association (section 18).
 Some fundamental rights simply do not lend themselves to actions in delict. Political rights,
such as the right to vote (section 19) and the right to citizenship (section 20) do not have private-
law counterparts. Similarly, a person is unlikely to have an action in delict if, for example, that
person’s right to housing (section 26), or rights to health care, food, water and social security
(section 27), is infringed, unless some other fundamental right that has a subjective-right
counterpart is also infringed. Other such examples include the rights to education (section 29),
language and culture (section 30), access to information (section 32), and just administrative action
(section 33).

Dendy v University of the Witwatersrand39

Dendy had applied, unsuccessfully, for a post of professor at the University of the
Witwatersrand. He claimed that the University had infringed his right to dignity in that there
were a number of procedural irregularities in the appointment process and, secondly, that the
University had failed to provide reasons for his lack of success and to furnish him with a copy
of the selection committee’s minutes. This conduct could obviously have infringed Dendy’s
right to fair labour practices (section 23), and his right to information (section 32), but since
there is no private-law counterpart for these rights, his claim in delict was not based on their
infringement. Instead he alleged that the University had infringed his right to dignity. The
Court found that society would not consider such conduct to be insulting and so his claim
failed.40

There are some instances in which we cannot identify an existing private-law right that corresponds
to a fundamental right, but, a subjective right could exist. Unless we can link the fundamental right
to a recognised subjective right – dignity, for example – a person cannot claim damages for
discrimination (a violation of the right to equality (section 9)), or for having views censored (a
violation of freedom of religion, belief and opinion (section 15)), or for a violation of freedom of
expression (section 16) or freedom of association (section 18). The law is not static and, in time,
courts might begin granting damages in delict for the mere infringement of these rights. If this does
happen, the effect would be that courts will recognise new subjective rights.
Remember, however, that fundamental rights are not omnipotent, and that the Constitution
makes provision for their limitation in section 36. So, a fundamental right that coexists with a
subjective right might strengthen the delictual claim, but there may still be other rights or values
that prevent the claim’s success. There might be countervailing rights that must be balanced against
the rights supporting the claim. Also, the traditional grounds for justification continue to play a role
in possibly defeating a plaintiff’s claim. In claims under the actio iniuriarum, there is often a
classic contest between two fundamental rights: on one side either the plaintiff’s right to dignity or
to privacy, and on the other, the defendant’s right to freedom of expression. Neither the plaintiff’s
nor the defendant’s right is more important than the other, and so we cannot say that one will
automatically override the other in every instance. However, the existence of one right (and the
values that underpin that right) naturally restricts the scope of the other. The extent of such
restriction, or even the extent to which one right might override the other, depends on the facts of
each case and what policy considerations apply to those facts. In some instances, policy might
dictate that the plaintiff’s right to dignity should prevail; in other instances, the defendant’s right to
freedom of expression.

2.3.2 Indirect application

The indirect application of the Bill of Rights to the law of delict occurs by virtue of section 39(2) of
the Constitution which requires courts to promote the spirit, purport and objects of the Bill of
Rights when interpreting any legislation or developing the common law.41 When courts apply the
Bill of Rights indirectly to the law of delict they generally do not override the common-law rules or
principles. Instead they respect the common law, but mould the rules and principles to accord with
constitutional rights and values.
Courts have repeatedly emphasised that the Constitution embodies a normative value system
that underpins our law and provides the backdrop against which we must develop the common
law.42 Therefore, the Constitution expresses society’s core values and sets basic criteria against
which we must test laws and conduct. The obligation to test and develop the common law rests
with every court. While they need not actively check constitutionality in every instance, they
should at least be familiar with the obligation, and where circumstances necessitate, ensure
compatibility.43 In S v Thebus 44 the Constitutional Court indicated when these circumstances might
arise:
The first would be when a rule of the common law is inconsistent with a constitutional
provision. Repugnancy of this kind would compel an adaptation of the common law to
resolve the inconsistency. The second possibility arises even when a rule of the common law
is not inconsistent with a specific constitutional provision but may fall short of its spirit,
purport and objects. Then the common law must be adapted so that it grows in harmony
with the ‘objective normative value system’ found in the Constitution.

In Dendy v University of the Witwatersrand 45 the Supreme Court of Appeal considered whether the
law relating to infringements of dignity needed to be developed. The Court accepted that it had an
obligation to ensure that the common law is consistent with the Constitution, but also noted that
any development must be done within the common law’s paradigm.46 The Thebus 47 test requires a
court first to enquire whether the existing common law should be developed. If the answer is no,
then that would be the end of the enquiry. Only if the answer is yes, should we enquire what
development should occur.48 In this instance, the Court concluded that the common law had not
been found wanting and was not in need of reform. Dendy could have vindicated his rights ‘by the
remedies of review or a relatively simple application to compel production of the documentation
and the reasons sought’, but chose not to do so, seeking instead to fashion a novel claim in delict.49

PAUSE FOR The difference between the two instances identified in S v Thebus50 that
REFLECTION would justify constitutional development of the common law is not self-
evident. For example, if a legal rule of the common law is inconsistent
with the Constitution, it must be taken that it also falls short of the spirit,
purport and objectives of the Bill of Rights, thereby placing it squarely in
both categories that the Court identified. Unless courts provide a clear
indication of when a legal rule would ‘fall short’ of the spirit, purport and
objectives of the Bill of Rights while not being inconsistent therewith, the
distinction seems to have little practical meaning.

Each of the rights mentioned in the Bill of Rights expresses in some way an aspect of the country’s
legal culture, its norms and its values. For example, the right to dignity articulates human dignity as
a societal value, as do the rights to privacy, and to freedom and security of the person. The right to
equality expresses both the core value of equality and that of human dignity. However in grey
areas, or in unfamiliar areas, where courts have to exercise discretion or choose between conflicting
courses of conduct, courts articulate constitutional principles and in so doing, express constitutional
values. In this regard, section 39 of the Constitution is clear. When interpreting the Bill of Rights,
‘the values that underlie an open and democratic society based on human dignity, equality and
freedom’ must be upheld, and similarly, when interpreting legislation or when developing the
common law or customary law, ‘the spirit, purport and objects of the Bill of Rights’ must be
promoted. And this, in essence, is the aim of indirectly applying the Constitution: the ordinary
common law is taken as the point of departure and courts then ensure that the constitutional values
permeate and radiate therein, not only in expressing legal principles and rules, but also in applying
them to factual situations. So, the purpose of constitutional values is to influence and guide the
common law, not to override it.51
Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies
Intervening) 52
The Constitutional Court provided clarity regarding the approach that courts must adopt when
requested to develop the common law:
• A court should consider whether the existing common law, having regard to the objectives
set out in section 39(2), requires development in accordance with those objectives. If this
question is answered positively, then the court must concern itself with how such
development is to take place in order to meet the objectives of the Bill of Rights.53
• The ‘obligation of Courts to develop the common law, in the context of s 39(2) objectives, is
not purely discretionary’. They are under a ‘general obligation’ to develop the common law in
an appropriate manner, but this does ‘not mean to suggest that a court must, in each and
every case where the common law is involved, embark on an independent exercise as to
whether the common law is in need of development’.54
• Courts were cautioned against ‘overzealous judicial reform’.55 The legislature remains the
‘major engine for law reform’.56
• The common law must be developed in a manner that not only gives effect to the
constitutional objectives, but also in a way most appropriate for the development of the
common law within its own paradigm.57 This is also in line with the general preference in
accordance with which legal disputes should preferably be decided in terms of the
established common law legal rules and principles, properly interpreted or developed to give
effect to the Bill of Rights, as opposed to a complete overhaul of the existing common-law
legal framework.58

There are various ways in which the courts may practically give effect to section 39(2) when
developing the law of delict by indirectly applying the Bill of Rights and examples of these are
discussed in further detail in the paragraphs below.

2.4 Constitutional values and norms

The Constitution embodies a normative value system that underpins our law and provides the
backdrop against which the common law is to be developed. It expresses society’s core values and
sets basic criteria against which laws and conduct must be tested. So, what are the core values that
must permeate every aspect of our law of delict? From the Constitution itself, and from judgments,
we can generally identify the following categories of norms.59

2.4.1 Foundational values that relate to dignity and equality

Section 1 of the Constitution sets out the foundational values for the country. These include ‘human
dignity, the achievement of equality and the advancement of human rights and freedoms’, and
‘non-racialism and non-sexism’. These norms cannot be divorced from their historical context and
these foundational values serve a transformative purpose to move from a society of inequality,
disrespect and exclusion, to one that respects human rights and freedoms. These values, especially
human dignity, underpin and support a number of sections in the Constitution,60 and have informed
a number of judgments.61 So, for example, equality, pluralism, religious freedom and tolerance have
served to recognise forms of marriage and life partnerships that were previously considered
unacceptable and accordingly, recognise claims arising from such relationships.62

PAUSE FOR Equality as a foundational value

REFLECTION Section 9(1) of the Constitution guarantees a general right to equality and
implies that any rule of common law or any legislation that differentiates
between persons must meet a minimum rationality requirement. In
Prinsloo v Van der Linde,63 a case dealing with legislation creating a
presumption of negligence in certain cases for the causing of veld fires,
Ackerman J explained the application of the minimum rationality standard
as follows:

[T]he constitutional state is expected to act in a rational

manner. It should not regulate in an arbitrary manner or
manifest naked preferences that serve no legitimate
governmental purpose, for that would be inconsistent with
the rule of law and the fundamental premises of the
constitutional state. The purpose of this aspect of equality
is, therefore, to ensure that the state is bound to function in
a rational manner. This has been said to promote the need
for governmental action to relate to a defensible vision of
the public good, as well as to enhance the coherence and
integrity of legislation.

In Harksen v Lane NO64 the Constitutional Court explained the standard

set by section 9(1) as a simple test: Does the provision differentiate
between people or categories of people? If so, does the differentiation
bear a rational connection to a legitimate government purpose? If it does
not, then there is a violation of section 9(1). Even if it does bear a rational
connection, it might nevertheless amount to unfair discrimination under
sections 9(3) or 9(4). The Constitutional Court in Jooste v Score
Supermarket Trading (Pty) Ltd (Minister of Labour intervening),65 a case
dealing with legislation replacing the common-law action by an employee
against his or her employer with a statutory claim, said the following on
the purpose of the rationality test:66

It is clear that the only purpose of rationality review is an

inquiry into whether the differentiation is arbitrary or
irrational, or manifests naked preference and it is irrelevant
to this inquiry whether the scheme chosen by the
legislature could be improved in one respect or another.

On application of the rationality test the courts have decided, for

example, that the limiting of claims for loss of income or loss of support
under the Road Accident Fund Act 56 of 1996 passed the rationality
test,67 whereas the former limit on claims by passengers did not.68

PAUSE FOR Human dignity as a foundational value

REFLECTION In The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici
Curiae) Ngcobo CJ said:69
 The Constitution proclaims human dignity to be one of the
foundational values of our constitutional democracy.
Human dignity is specifically mentioned in section 1 of the
Constitution in order to contradict our racist past. For this
reason, the Constitution holds human dignity up as not
only a human right that is given constitutional recognition,
as with freedom of expression, but also as a fundamental
value upon which the legitimacy of the sovereign state is
based. The Republic was – ‘founded on’ the value of human
dignity, and failure to uphold that value is both a violation
of a constitutional right and a threat to a bedrock principle
that underpins the legitimacy of the state.

2.4.2 Foundational values associated with democracy and governance

Section 1(d) of the Constitution mentions the need for democratic government and ‘to ensure
accountability, responsiveness and openness’. 70 So, where public authorities are involved, courts
should ensure that their decisions promote governmental service delivery that reflects these values.
In Minister of Safety and Security v Van Duivenboden71 the police had information showing that the
person who had shot Van Duivenboden was unfit to possess a firearm, but they failed to investigate
further. In holding the police liable for Van Duivenboden’s injuries, the Court noted that public
functionaries and public authorities were not on the same level as ordinary citizens.72 With
reference to the norm of accountability the Supreme Court of Appeal said:73
When determining whether the law should recognise the existence of a legal duty in any
particular circumstances what is called for is not an intuitive reaction to a collection of
arbitrary factors but rather a balancing against one another of identifiable norms. Where
the conduct of the State, as represented by the persons who perform functions on its behalf,
is in conflict with its constitutional duty to protect rights in the Bill of Rights, in my view, the
norm of accountability must necessarily assume an important role in determining whether a
legal duty ought to be recognised in any particular case. The norm of accountability,
however, need not always translate constitutional duties into private law duties enforceable
by an action for damages, for there will be cases in which other appropriate remedies are
available for holding the State to account. Where the conduct in issue relates to questions of
State policy, or where it affects a broad and indeterminate segment of society, constitutional
accountability might at times be appropriately secured through the political process or
through one of the variety of other remedies that the courts are capable of granting.

A decision like this one, which calls upon police to improve the quality of their services, clearly
promotes the constitutional norm. In Fair Cape Property Developers (Pty) Ltd v Premier of the
Province of the Western Cape 74 the Court indicated that an accountable government would be
promoted if citizens could obtain relief for harm caused to them pursuant to operational decisions.

2.4.3 Foundational values associated with freedom of expression

Freedom of expression has its roots in section 1 of the Constitution, and it could also fall into either
of the categories mentioned in 2.4.1 and 2.4.2. The common law recognised, and substantially
endorsed freedom of expression, and so this is not a new value that was absent in the previous era.75
The place of freedom of expression in our common law is slightly different from other foundational
values, and because of its prominence as a value that supports defences against claims under the
actio iniuriarum, it is categorised separately.
But this special status does not mean that freedom of expression is more important than other
foundational values.76 Although important, there is no hierarchy of rights77 and this freedom ‘must
be construed in the context of the other values enshrined in the Constitution’, particularly, human
dignity, equality and freedom.78 Freedom of expression partially underpins freedom of assembly,
demonstration and petition, and it has always played an important role in justifying injurious
statements. Accordingly, the law recognises that a democracy requires free expression in
parliament and that persons should be entitled to voice opinions on other privileged occasions, that
fair comment is acceptable, and that there should be no liability in respect of true statements made
for the public benefit. In each of these instances, courts are required to balance plaintiffs’
constitutional and common-law entitlements to be protected against violations of their personality,
against similar entitlements that defendants have in respect of freedom of expression. This freedom
dominates in the public domain. While freedom of expression has a role in recognising
parliamentary privilege, it is the core value that determines the role of the media in our society. Not
only are the media the messengers and society’s eyes and ears with regard to societal issues, but
they also provide an important vehicle through which members of society can exercise their roles
as citizens. Although arguments based on freedom of expression have not been sufficient to
convince courts to recognise a special privilege for the media, the value was instrumental in
overruling the controversial strict-liability regime for the media, which placed a prior restraint on
the media and so had a ‘chilling effect’ on their right to free expression.79

2.4.4 The values of non-violence and the protection of vulnerable people

from violence and abuse
These values, especially the protection of women and children from violence and abuse, were
considered in Van Eeden v Minister of Safety and Security 80 and in Carmichele v Minister of Safety
and Security.81 In the former case, a serial rapist escaped from police cells and subsequently
assaulted, raped and robbed Van Eeden. In the latter, Carmichele was attacked by a man with a
record of violence, and who had been granted bail. In light of the State’s constitutional duty to
protect persons from violence and to ensure people’s safety, the claims in both instances succeeded.
Although the majority in Le Roux v Dey 82 did not decide the case on this basis, the rights of
children was prominent in the two minority judgments.83 Yacoob J noted in this case that ‘the
violation of the dignity of a relatively powerless and vulnerable person by a powerful, strong
person in authority’ would be more serious than the other way around.84

2.5 The Constitution’s areas of impact on the law of delict

We have seen so far that the impact of the Constitution can be both direct and indirect. However, it
is its indirect, developmental impact that is likely to be more substantial. The manner of such
influence, and its scope, is multi-faceted, but fall into two broad categories in which:
1. The validity of rules and principles are tested for constitutional compatibility
2. The application of rules and principles are tested for compatibility with constitutional norms
regarding social justice.

Constitutional issues often come to the fore under the wrongfulness enquiry, when courts have to
determine the nature of the community’s legal convictions and give effect to constitutional values
and norms.85
However, this is not the only area where they can apply. In addition:
• We can use the Constitution to test the delict system as a whole, for example, whether a
particular rule – fault or strict liability in particular circumstances, or the test for wrongfulness
– conforms to the rights and values that society ascribes to.
• We can also use the Constitution to test the application of (constitutionally valid) rules to
particular facts. In this latter sense, the outcome of a case must give effect to constitutional
principles and values. So, where we have to make a choice between two legitimate courses of
action, we should select the one that best gives effect to constitutional principles.
• We can use the Constitution to create new delictual rules. For example, we may now recognise
rights that we previously did not consider to be actionable, and we may find new remedies. In
short, the Constitution is relevant to every aspect of the law of delict and we should not
attempt to confine its influence in particular boxes.

2.5.1 Adjusting the way in which an established legal rule is applied

In Carmichele v Minister of Safety and Security 86 the applicant brought a delictual claim for
damages against the Minister of Safety and Security and the Minister of Justice for harm she had
suffered as a result of being assaulted by one Coetzee. Prior to the assault, Coetzee had already
been convicted on charges of housebreaking and indecent assault. He had also been accused of rape
and had appeared earlier before the magistrate’s court on this charge. At this hearing he was
released on his own recognisance on the recommendation of the investigating officer.
Even though members of the public provided the investigating officer with information that
Coetzee posed a significant threat to their safety and security, the officer advised the public
prosecutor that there was no reason to deny Coetzee bail and recommended that he be released on
warning. When Coetzee subsequently appeared before a magistrate on the charge of rape, the
prosecutor did not place before the magistrate any information concerning Coetzee’s previous
convictions, nor did he oppose Coetzee’s release on his own recognisance. Following his release, a
concerned member of the community again approached the police and requested Coetzee’s
detention pending his trial. The police officer in question advised her to discuss the matter with the
public prosecutor who, in turn, advised that nothing could be done unless Coetzee committed
another offence.
Shortly thereafter, Coetzee was re-arrested, but after pleading not guilty on the charge of rape,
he was re-released by the magistrate, pending a decision by the Attorney-General as to whether the
case should be tried in the High Court or the regional court. The Attorney-General, who had been
in possession of the referral documents which reflected the seriousness of the rape and the extent of
Coetzee’s sexual deviation, had not instructed the public prosecutor to oppose bail, with the result
that his release was not opposed.
Following Coetzee’s unopposed re-release, Carmichele was assaulted. She then instituted a
delictual claim against the Minister of Safety and Security and the Minister of Justice, arguing that
the members of the police as well as the public prosecutors had owed her a legal duty to prevent
Coetzee from being released on bail, and that their negligent failure to comply with this duty
enabled him to cause her harm. The High Court rejected the argument and ordered absolution from
the instance on the ground that such failure could not be considered wrongful.87 The appeal to the
Supreme Court of Appeal was dismissed. In accordance with established precedent at the time,
wrongfulness and the concomitant enquiry into the existence of a legal duty fell to be determined
with reference to the boni mores or legal convictions of the community. In denying the existence of
a legal duty on the part of the police and state prosecutors, the High Court and the Supreme Court
of Appeal had followed established precedent.88
Carmichele subsequently appealed to the Constitutional Court. The Court held that the
previous courts had overlooked the demands of the Constitution and that the common law of delict
had to be developed beyond existing precedent. The Constitutional Court stressed the general
obligation that rested on courts to develop the common law to give effect to the constitutional
objectives 89 and referred the case back to the High Court. This court, in its second judgment in the
matter, allowed the plaintiff’s claim.90 Following another appeal to the Supreme Court of Appeal,
the Ministers’ appeal was dismissed and the matter finally came to conclusion.91
Subsequent to the Constitutional Court decision, but before the Carmichele saga came to its
full conclusion, the Supreme Court of Appeal had an opportunity to illustrate the impact of the
Constitution on the law of delict, as per the Constitutional Court’s guidelines in its Carmichele
judgment. In Minister of Safety and Security v Van Duivenboden 92 the plaintiff brought a claim in
delict against the Minister of Safety and Security after he had been shot by his neighbour, who
proceeded to kill his own wife and daughter. It was common cause that, prior to the incident, police
officers had information that the perpetrator, when drunk, habitually threatened to use his firearms
against himself and others. Nonetheless, they had failed to take any steps to initiate an enquiry in
terms of the Arms and Ammunition Act 75 of 1969 empowering the Commissioner of Police to
declare someone unfit to possess a firearm and to seize it.
The Supreme Court of Appeal reiterated the common-law rule that the negligent failure to act
positively in preventing harm was not prima facie wrongful.93 The Court therefore had to consider
whether or not the police officers had owed the plaintiff a legal duty to prevent his harm. The
majority restated the established common-law criterion as developed in Minister van Polisie v
Ewels 94 (the boni mores or legal convictions of the community criterion)95 and noted that
establishing wrongfulness ultimately depended on a judicial determination of whether it would be
reasonable to impose liability on a defendant for the harm arising from the defendant’s culpable
conduct. Such a judicial determination would in turn depend on considerations of public and legal
policy, which, importantly, now also included the norms and values of the society as embodied in
the Constitution.96 In this way, public and legal policy became the gateway for the introduction of
constitutional rights and norms in the application of the traditional rules of the law of delict.97
The Court noted several public and legal policy considerations that might weigh against the
imposition of delictual liability upon the State:
• Public policy considerations, rooted in a laissez faire concept of liberty, suggest that it might
be an unreasonable infringement upon someone’s personal autonomy to expect him or her to
take positive steps in order to avert harm to others.98
• The principle of equality might be infringed upon if liability is imposed on one person where
others might equally be faulted for their failure in preventing the relevant harm.99 • The
apparent utility of allowing the State and its functionaries the freedom to provide public
services without the chilling effect of the threat of litigation were they negligently fail to
prevent harm.100
• The possibility of unlimited liability and its effect on potential policing resources.

These concerns were outweighed by the following considerations:

• The obligation imposed on the State in terms of section 7 of the Constitution not only to
respect but also to ‘protect, promote and fulfil the rights in the Bill of Rights’. 101
• Section 2 of the Constitution demanded that duties imposed by the Constitution on the State
must be fulfilled.102 (The relevant constitutional rights which the Court had in mind may have
included the right to safety and security of the person, the right to life and the right to human
dignity.)
• Section 41(1) of the Constitution expressly stated that all spheres of government and all state
organs within such sphere must provide government that is not only effective, transparent and
coherent, but also accountable.103
Ultimately, the Court attached considerable weight to the constitutional norm of accountability,104
emphasising that, other than imposing delictual liability on the State, there appeared to be no way
of holding the State accountable. The Court found that the relevant police officers’ harm-causing
failure to act was indeed wrongful105 and, having also determined that their failure was negligent106
and caused the victim’s harm,107 it accordingly held that the State was vicariously liable.108
When the Carmichele matter came to the Supreme Court of Appeal for the second time,109 the
Court applied this reasoning in overturning its initial decision, holding that, in view of the
considerations outlined above, the harm caused to the plaintiff by the negligent failure of the police
officers and public prosecutors was indeed wrongful.
Wrongfulness is a normative concept which is ultimately determined by legal and public
policy considerations and the Constitution now informs this enquiry. This avenue of constitutional
development does not require amendment to the established legal rules and principles of the
common law, but does necessitate an adjustment in the way in which these legal rules and
principles have been applied in the past.110 It has been particularly effective within the law of delict
where its significant impact is reflected in subsequent decisions regarding the imposition of
delictual liability.111

2.5.2 Changing an existing legal rule

The common law is developed whenever an existing common-law rule is changed. For instance, in
Du Plessis v Road Accident Fund 112 the appellant and the deceased were in a same-sex union when
the deceased was killed in a motor vehicle accident. On appeal, the primary question was whether
the appellant should be entitled to claim compensation for loss of support from the respondent in
terms of the Road Accident Fund Act. After holding that the deceased owed the appellant a
contractual duty of support, the Constitutional Court had to determine whether the common law
should be developed to recognise the appellant’s right to such support as being worthy of
protection. The Court applied the reasoning developed in Carmichele CC and Van Duivenboden
and noted that regard must be had to the spirit, purport and objects of the Bill of Rights; and
further, that the determination of wrongfulness had to occur in the context of a constitutional state
founded on dignity, equality and freedom and in which the State had positive duties to promote
such values. The constitutional rights relevant to the development of this sphere of the common law
were those contained in sections 9 (equality) and 10 (human dignity) of the Constitution.113
Ultimately, under the indirect influence of the Constitution, the Court opted to develop the common
law and held that the appellant, as a same-sex partner of the deceased in a permanent life
relationship similar in other respects to marriage and in which the deceased had undertaken a
contractual duty of support to him, is entitled to claim damages from the respondent for loss of that
support. Other courts have used this reasoning to extend similar rights to dependent partners,
children, and parents in customary-law and Islamic relationships, gay and heterosexual co-
habitation relationships, adopted children, close family members and ex-husbands.114
In Lee v Minister for Correctional Services 115 the plaintiff contracted tuberculosis during his
incarceration in the Pollsmoor Prison. He alleged that the defendant had failed to implement a
reasonable preventative system to guard against the contagion of the disease, which failure was not
only negligent and wrongful, but also caused his infection. Accordingly, he instituted a delictual
claim for damages and the key legal issue was whether factual causation had been proved.
The established common-law test for determining factual causation is the so-called ‘but-for’
test,116 whereby a plaintiff is required to prove that, but for the culpable act or omission of the
defendant, the harmful consequence in question would probably not have occurred.117 While the
High Court upheld the plaintiff’s claim, the Supreme Court of Appeal dismissed it on the basis that
it had not been shown that the defendant’s wrongful and negligent failure to implement existing
measures was the factual cause of the plaintiff’s infection. The Court noted the peculiar nature of a
tuberculosis infection and the complexities in satisfying the ‘but-for’ test on the particular facts,118
but nonetheless, held that the orthodox application of the established legal rule for determining
factual causation meant that the plaintiff could not be awarded a remedy in delict.119
The Constitutional Court found that the Supreme Court of Appeal’s application of the common
law ‘but-for’ test had been too rigid.120 Although the Court, in a majority judgment, maintained that
it was not developing the common law of delict, it nevertheless repeatedly emphasised the so-called
flexibility of the ‘but-for’ test for factual causation;121 and, by applying the established rule flexibly,
the Court was able to find the necessary factual link for the purposes of holding the defendant
delictually liable.122 Notwithstanding the fact that the application of the so-called flexible approach
enabled the Court to establish the required factual link for the purposes of delictual liability,
Nkabinde J went further and argued that the orthodox ‘but-for’ test could yield the same result.123
This conclusion was reached in the following way: if reasonable conduct on the part of the prison
authorities were to be postulated, as is required in the application of the orthodox ‘but-for’ test, the
general risk of an infection would decrease and, on the basis of inference, this also meant that the
risk of an infection in respect of a specific detainee (in this case, the plaintiff) would similarly
decrease.124 Therefore, it is arguable that the majority of the Court did in fact develop the common
law in holding that factual causation could be established in situations where the defendant’s
conduct increased the risk of harm and the harm in fact occurred.

PAUSE FOR It is arguable that, in departing from the orthodox understanding and
REFLECTION application of the established common-law rule for establishing the
factual connection, by applying the so-called flexible test in
circumstances where the defendant’s conduct had increased the risk of
harm, the majority of the Constitutional Court changed the existing legal
rule of the common law. The implications of this approach are uncertain.
The majority provided no guidelines or criteria for further application of
the so-called flexible approach. The new approach furthermore
undermines the constitutional principle of the rule of law, which is aimed
at providing legal certainty and making it easier to plan one’s affairs, to
give reliable legal advice and may result in an increase in costly, time-
consuming and undesirable litigation.125

2.5.3 Introduction of a new legal rule

In H v Fetal Assessment Centre 126 the plaintiff, a boy with Down’s syndrome, instituted a delictual
claim for the harm he had suffered from the defendant’s alleged failure to warn his pregnant mother
that there was a high risk of him being born with the syndrome. The plaintiff alleged that, if his
mother had been informed of the risk, she would have terminated the pregnancy. He claimed
damages for past and future medical expenses, for disability and for loss of amenities of life. The
defendant excepted to the claim on the basis that it failed to disclose a cause of action, which was
upheld by the High Court. The plaintiff appealed directly to the Constitutional Court.
Our law recognises a parent’s claim in similar circumstances and at issue was whether a child
should also have a claim, which up to that point our law had not recognised. The Constitutional
Court had to consider whether delictual liability should be expanded by recognising the so-called
claim for wrongful life. The Court found that the existing authority barring the claim did not take
sufficient account of the right of a child in section 28(2) of the Constitution, nor other constitutional
rights, and that the common-law elements of delictual liability could accommodate the claim.127 It
accordingly referred the matter back to the High Court to determine whether the claim should be
allowed. If the High Court follows the Constitutional Court’s lead and recognises the new cause of
action, it will amount to the introduction of a new legal rule.

2.5.4 Application of an established legal rule to a new set of facts

In K v Minister of Safety and Security 128 three uniformed, on-duty police officers raped the plaintiff
after offering to give her a lift home. At issue was whether the Minister of Safety and Security, as
employer of the three policemen, could be held vicariously liable for the wrongful and intentional
conduct of these employees. Previously the Supreme Court of Appeal, in line with an earlier
Constitutional Court decision, Phoebus Apollo Aviation CC v Minister of Safety and Security,129
held that the test for vicarious liability is constitutionally consistent and that the application of the
requirements for vicarious liability does not raise a constitutional issue.130 However, on appeal the
Constitutional Court distinguished the Phoebus Apollo judgment. It found that the requirements for
vicarious liability, when applied to a new set of facts, amounted to a development of the common
law.131
Because the Constitutional Court interpreted the situation as a development of the common
law, such development was required to occur in a manner that is consistent with constitutional
rights and values.132 Ultimately, by relying heavily on the plaintiff’s constitutional rights to human
dignity and safety and security, as well as the constitutional norm of accountability, the Court
developed the common-law doctrine of vicarious liability by holding that the police officers were
acting in the course and scope of their employment when they intentionally committed a violent
crime.

COUNTER The Constitutional Court’s line of reasoning means in practice that the
POINT application of any common-law rule or principle to a new set of facts will
present constitutional issues and call for the development of the common
law. This may place a heavy load on the already over-burdened courts to
hear an inordinate number of cases dealing with the constitutional
development of the common law, which in turn could be prejudicial to the
administration of justice. In addition, it could also undermine legal
certainty pertaining to the existence and application of established
common-law rules, which, as mentioned above, may contribute to the
erosion of the constitutional principle of the rule of law.133

2.6 Conclusion
While its role is not always fully appreciated, the Constitution is core to every aspect of the law of
delict and delictual problems cannot be resolved without the full understanding and acceptance that
delict law functions as part and parcel of a constitutional state. The Constitution serves both an
oversight and a supportive role. Delictual rules that run contrary to constitutional rights and values
must be made to comply, while those that are compatible with the normative matrix that the
Constitution provides are supported and strengthened. Therefore, the violation of a delictual interest
‘in the context of the violation of other constitutional rights would ordinarily be regarded as more
serious than otherwise’,134 but similarly, a violation might not be as egregious if the Constitution
favours a plaintiff’s interest as well.

Carmichele v Minister of Safety and Security135

The facts of this case are set out in the text above.136 Because the case arguably illustrates
most dramatically the Constitution’s impact on the law of delict, we conclude this chapter by
noting some of the principles that this case highlights.
The High Court, and subsequently the Supreme Court of Appeal,137 dismissed the claim on
the grounds that neither the police nor prosecutors had acted wrongfully. These decisions
reflected the accepted delictual principles that applied at the time for determining the
circumstances in which there is a duty to act.
The issue in the Constitutional Court was whether to develop the law in light of
Carmichele’s contention that her rights to equality, life, human dignity, freedom of security
and privacy had been violated, as well as the constitutional provisions that relate to the
functions of police. In particular, she alleged that the State had a duty to protect women
against violent crime and sexual abuse.
• The Constitutional Court reiterated that the Constitution is the supreme law and that the Bill
of Rights applies to all law, and when developing the common law, every court must promote
the spirit, purport and objects of the Bill of Rights, and where deviations are found to exist,
remove that deviation.138 The High Court and the Supreme Court had been requested to
develop the common law and so these principles applied. The constitutional obligation to
develop the common law is not purely a discretionary one, but a general one:139
We say a ‘general obligation’ because we do not mean to suggest that a court must,
in each and every case where the common law is involved, embark on an
independent exercise as to whether the common law is in need of development and,
if so, how it is to be developed under section 39(2). At the same time there might be
circumstances where a court is obliged to raise the matter on its own and require
full argument from the parties.

The Court further noted:140

It was implicit in the applicant’s case that the common law had to be developed
beyond existing precedent. In such a situation there are two stages to the inquiry a
court is obliged to undertake. They cannot be hermetically separated from one
another. The first stage is to consider whether the existing common law, having
regard to the section 39(2) objectives, requires development in accordance with
these objectives. This inquiry requires a reconsideration of the common law in the
light of section 39(2). If this inquiry leads to a positive answer, the second stage
concerns itself with how such development is to take place in order to meet the
section 39(2) objectives.
• After noting the common-law test for wrongfulness in delict, and that it required courts to
weigh and strike a balance between the interests of parties and the conflicting interests of
the community, the Court commented:141
This is a proportionality exercise with liability depending upon the interplay of
various factors. Proportionality is consistent with the Bill of Rights, but that
exercise must now be carried out in accordance with the ‘spirit, purport and objects
of the Bill of Rights’ and the relevant factors must be weighed in the context of a
constitutional state founded on dignity, equality and freedom and in which
government has positive duties to promote and uphold such values.
• The State is bound (directly) not to perform any act that infringes the rights to life, human
dignity and freedom and security of the person, and ‘(i)n some circumstances there would
also be a positive component which obliges the state and its organs to provide appropriate
protection to everyone through laws and structures designed to afford such protection’.142
The Court stressed that the Constitution ‘is not merely a formal document regulating public
power’, but ‘also embodies, like the German Constitution, an objective, normative value
system’ that provides a matrix within which the common law must be developed.143
• The Court also provided an important caveat:
 Not only must the common law be developed in a way which meets the section 39(2)
objectives, but it must be done in a way most appropriate for the development of the
common law within its own paradigm.144
• After referring to the principles that it set out earlier in Minister of Safety and Security v Van
Duivenboden,145 the Supreme Court of Appeal146 concluded that:
[S]omeone in the position of the plaintiff has no other effective remedy against the
State, an action for damages is the norm unless public policy considerations point
in the other direction.147

The Court continued:148

Did the State owe a duty to the plaintiff? The answer lies in the recognition of the
general norm of accountability: the State is liable for the failure to perform the
duties imposed upon it by the Constitution unless it can be shown that there is
compelling reason to deviate from that norm. In Van Eeden,149 it is suggested that
such a deviation might be warranted where it would not be in the public interest to
inhibit the police (and by parity of reasoning the prosecution) in the proper
performance of their duty. A deviation was not, however, considered to be
necessary in that case.
Nor is there reason in this case to depart from the general principle that the State
will be liable for its failure to comply with its Constitutional duty to protect the
plaintiff. On the contrary, the plaintiff is pre-eminently a person who required the
State’s protection. It was known by Klein, Hugo and Louw that Coetzee resided in
Noetzie with his mother. Noetzie is a small hamlet with a few houses. Coetzee’s
mother worked for Gosling in the house where the attack on the plaintiff occurred.
She regularly visited the house. She knew Coetzee. The attack took place within
four months after his release after the attack on Eurona Terblanche. The plaintiff
was not simply a member of the public whom the State had a duty to protect. She
was a member of a class of people whom the State would have foreseen as being
potential victims of another attack by Coetzee. Proximity, while not an independent
requirement for wrongfulness, must surely reinforce the claim that the State should
be held liable for a culpable failure to comply with its duties. And foreseeability of
harm is another factor to be taken into account in determining wrongfulness. The
greater the foreseeability, the greater the possibility of a legal duty to prevent harm
existing. This can be compared to the development in English law in relation to the
tort known as misfeasance by a public officer. An element of this tort is, in our
terms, dolus directus or eventualis: if a public officer knows that his unlawful
conduct will probably injure another or a class of persons, the State may be liable
for the consequences. (The question of foreseeability arises also, of course, when
determining negligence: but it may in appropriate cases play a role in determining
whether the defendant should be held liable for failure to perform a duty.)

1 Section 2.
2 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC) paras 19 and 69.
3 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC) para 69.
4 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC) para 19.
5 Dendy v University of the Witwatersrand 2005 (5) SA 357 (W) para 19.
6 1997 (3) SA 786 para 1.
7 Para 66.
8 Para 60.
9 2004 (6) SA 40 (SCA). See also President of the Republic of South Africa v Modderklip Boerdery (Pty) Ltd (Agri SA
and Legal Resources Centre, Amici Curiae) 2005 (5) SA 3 (CC).
10 Paras 42–43.
11 2006 (4) SA 478 (SCA).
12 2006 (4) SA 478 (SCA).
13 Para 27.
14 2014 (6) SA 256 (SCA).
15 1997 (3) SA 786 (CC) para 1.
16 Minister of Police v Mboweni 2014 (6) SA 256 (SCA) para 25.
17 Typically, personal rights, personality rights, immaterial property rights, real rights.
18 2005 (5) SA 357 (W); 2007 (5) SA 382 (SCA).
19 The plaintiff contended that his constitutional rights to equality, dignity, freedom of conscience, freedom of
expression, freedom of association, fair labour practices and just administrative action had been violated.
20 Dendy contended that his constitutional rights to equality, dignity, fair labour practices, access to information and just
administrative action had been violated.
21 Para 15.
22 Paras 23–24.
23 Dendy v University of the Witwatersrand 2007 (5) SA 382 (SCA) paras 17–19.
24 Fose v Minister of Safety and Security 1997 (3) SA 786 (CC). Also see President of the Republic of South Africa v
Modderklip Boerdery (Pty) Ltd (Agri SA and Legal Resources Centre, Amici Curiae) 2005 (5) SA 3 (CC); MEC,
Department of Welfare Eastern Cape v Kate 2006 (4) SA 478 (SCA); Dikoko v Mokhatla 2006 (6) SA 235 (CC);
Zealand v Minister of Justice and Constitutional Development 2008 (4) SA 458 (CC); 2008 (2) SACR 1 (CC).
25 Section 12 of the Constitution.
26 1997 (3) SA 786 para 1.
27 For example, the rights relating to life, human dignity and freedom and security of the person.
28 Law Society of South Africa v Minister of Transport 2011 (1) SA 400 (CC) para 73.
29 Para 73.
30 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 31.
31 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC)
para 33.
32 Van der Walt and Midgley Principles of the Law of Delict 4 ed (2016) para 19.
33 Section 8(2).
34 2002 (5) SA 401 (CC) paras 30–31.
35 Para 31.
36 Para 41.
37 Para 31.
38 Para 32.
39 2007 (5) SA 382 (SCA).
40 Paras 17–19.
41 Section 39(2) of the Constitution.
42 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC)
para 54; Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) para 17; Van Eeden v Minister of
Safety and Security (Women’s Legal Centre Trust as Amicus Curiae) 2003 (1) SA 389 (SCA) para 12.
43 S v Thebus 2003 (6) SA 505 (CC) paras 34 and 39.
44 2003 (6) SA 505 (CC) para 28.
45 2007 (5) SA 382 (SCA).
46 Para 22. See also S v Thebus 2003 (6) SA 505 (CC) para 45.
47 S v Thebus 2003 (6) SA 505 (CC).
48 Para 23.
49 Para 24.
50 2003 (6) SA 505 (CC) para 28.
51 Van der Walt and Midgley (2016) paras 21 and 25.
52 2001 (4) SA 938 (CC).
53 Para 40.
54 Para 39.
55 Para 55.
56 Para 36.
57 Para 55.
58 Currie and De Waal Bill of Rights Handbook 6 ed (2013) at 56–66.
59 See in general, Van der Walt and Midgley (2016) paras 22–24 and 25. There are other ways of grouping these values,
of course. We have merely chosen a way that suits us here.
60 S v Mamabolo (E TV and Others Intervening) 2001 (3) SA 409 (CC) para 41.
 61 Including Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA
938 (CC); Van Eeden v Minister of Safety and Security (Women’s Legal Centre Trust as Amicus Curiae) 2003 (1) SA
389 (SCA); Dendy v University of the Witwatersrand 2007 (5) SA 382 (SCA); NM v Smith (Freedom of Expression
Institute as Amicus Curiae) 2007 (5) SA 250 (CC).
62 Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA
421 (SCA); 1999 (4) SA 1319 (SCA); Du Plessis v Road Accident Fund 2003 (11) BCLR 1220 (SCA); 2004 (1) SA
359 (SCA); Langemaat v Minister of Safety and Security 1998 (3) SA 312 (T).
63 1997 (3) SA 1012 (CC); 1997 (6) BCLR 759 (CC).
64 1998 (1) SA 300 (CC) para 53.
65 1999 (2) SA 1 (CC).
66 Para 17.
67 Law Society of South Africa v Minister for Transport 2011 (1) SA 400 (CC) paras 34–35.
68 Mvumvu v Minister of Transport [2011] 1 All SA 90 (WCC).
69 2011 (4) SA 191 (CC); 2011 (8) BCLR 816 (CC) para 143.
70 These values are repeated in section 195 in respect of public administration.
71 2002 (6) SA 431 (SCA).
72 Para 19.
73 Para 21. See also Van Eeden v Minister of Safety and Security (Women’s Legal Centre Trust as Amicus Curiae) 2003
(1) SA 389 (SCA) and Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening)
2001 (4) SA 938 (CC).
74 [2003] 2 All SA 465 (SCA).
75 Du Plessis v De Klerk 1996 (3) SA 850 (CC) para 58; Argus Printing & Publishing Co Ltd v Esselen’s Estate 1994 (2)
SA 1 (A) at 25B–E; Hix Networking Technologies v System Publishers (Pty) Ltd 1997 (1) SA 391 (A) at 400.
76 S v Mamabolo (E TV and Others Intervening) 2001 (3) SA 409 (CC) para 41:
With us the right to freedom of expression cannot be said automatically to trump the right to human
dignity. The right to dignity is at least as worthy of protection as the right to freedom of expression.
How these two rights are to be balanced, in principle and in any particular set of circumstances, is not
a question that can or should be addressed here. What is clear though and must be stated, is that
freedom of expression does not enjoy superior status in our law. (See also para 37.)
77 The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae) 2011 (4) SA 191 (CC); 2011 (8) BCLR
816 (CC) para 148.
78 Khumalo v Holomisa 2002 (5) SA 401 (CC) para 25.
79 National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) at 1209–1211; Holomisa v Argus Newspapers Ltd 1996 (2)
SA 588 (W). Freedom of expression also featured in The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others,
Amici Curiae) 2011 (4) SA 191 (CC) ; 2011 (8) BCLR 816 (CC) paras 78 and 141–153, and Le Roux v Dey (Freedom
of Expression Institute and Restorative Justice Centre as Amicus Curiae) 2011 (3) SA 274 (CC) para 47.
80 2003 (1) SA 389 (SCA) para 13.
81 2001 (4) SA 938 (CC). See also Hoffmann v South African Airways 2001 (1) SA 1 (CC).
82 2011 (6) BCLR 577 (CC); 2011 (3) SA 274 (CC).
83 See, for example, paras 32–34, 46, 48–50, 53–57 and 210–215.
84 Para 46.
85 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amicus Curiae) 2011 (6) BCLR
577 (CC); 2011 (3) SA 274 (CC) para 180.
86 2001 (4) SA 938 (CC).
87 Carmichele v Minister of Safety and Security and Minister of Justice (11 November 1997, CPD, unreported).
88 Carmichele v Minister of Safety and Security 2001 (1) SA 489 (SCA) at 494–497.
89 2001 (4) SA 938 (CC) para 39.
90 Carmichele v Minister of Safety and Security 2003 (2) SA 656 (C).
91 Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA).
92 2002 (6) SA 431 (SCA).
93 Paras 12–13.
94 1975 (3) SA 590 (A).
95 2002 (6) SA 431 (SCA) paras 12–13.
96 2002 (6) SA 431 (SCA) paras 16–18.
97 Brand ‘Influence of the Constitution on the Law of Delict’ (2014) 27(1) Advocate 42 at 42–43.
98 2002 (6) SA 431 (SCA) paras 16–18.
99 Para 19.
100 Paras 19–20.
101 Para 20.
102 Para 20.
103 Para 21.
104 See also section 41(1) of the Constitution, which expressly provides that all spheres of government and all organs of
state within such sphere must provide government that is not only effective, transparent, and coherent, but also
accountable.
105 Para 22.
106 Para 23.
107 Paras 24–30.
108 Para 22.
109 Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA).
110 Brand (2014) at 43.
111 See, for example, Van Eeden v Minister of Safety and Security (Women’s Legal Centre Trust as Amicus Curiae) 2003
(1) SA 389 (SCA); Minister of Safety and Security v Hamilton 2004 (2) SA 216 (SCA), Minister of Finance v Gore
NO 2007 (1) SA 111 (SCA); Dendy v University of the Witwatersrand 2007 (5) SA 382 (SCA); NM v Smith (Freedom
of Expression Institute as Amicus Curiae) 2007 (5) SA 250 (CC), Minister of Correctional Services v Lee 2012 (3) SA
617 (SCA).
112 2003 (11) BCLR 1220 (SCA); 2004 (1) SA 359 (SCA).
113 In particular sections 9(4) and 9(5), which provided that no person may unfairly discriminate against anyone on the
ground of sexual orientation and that discrimination on this ground is presumed to be unfair.
114 Fosi v RAF 2008 (3) SA 560 (C); MB v NB 2010 (3) SA 220 (GSJ); Verheem v RAF 2012 (2) SA 409 (GNP); Paixão v
Road Accident Fund 2012 (6) SA 377 (SCA); Mnguni v RAF 2015 JDR 1723 (GP); Osman v Road Accident Fund
2015 (6) SA 74 (GP); Seleka v RAF 2016 (4) SA 445 (GP). For more detail, see Chapter 23.
115 2013 (2) SA 144 (CC).
116 Mashongwa v Passenger Rail Agency of South Africa 2016 (3) SA 528 (CC) para 65; South African Hang and
Paragliding Association v Bewick 2015 (3) SA 449 (SCA) para 34.
117 Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA) para 46; Minister of Police v Skosana 1977(1) SA 31
(A) at 35.
118 Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA) paras 1–10, 51–55 and 62–67; Lee v Minister for
Correctional Services 2013 (2) SA 144 (CC) paras 82–87.
119 Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA) paras 44, 46–47, 56 and 61–67.
120 Lee v Minister for Correctional Services 2013 (2) SA 144 (CC) paras 44–46.
121 Paras 44–50.
122 Paras 41, 43, 45, 47, 49, 50, 63 and 73.
123 Paras 58–70.
124 Paras 58–70.
125 See Wessels ‘Alternatiewe benaderings ten opsigte van feitelike kousaliteit in die deliktereg’ (2013) 10(3) Litnet
Akademies (Regte); Price ‘Factual Causation after Lee’ (2014) 131(3) SALJ at 491.
126 2015 (2) SA 193 (CC).
127 Paras 52 and 53.
128 K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC) at 432.
129 2003 (2) SA 34 (CC).
130 2005 (3) SA 179 (SCA) para 8.
131 Paras 16–20.
132 K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC) at 432.
133 See also Fagan ‘Reconsidering Carmichele’ (2008) 125(4) SALJ 659 at 672, who maintains that the Constitutional
Court has misinterpreted section 39(2) of the Constitution. He submits that, properly interpreted, it does not impose a
duty on courts to develop the common law whenever that would promote human rights. He argues that it only obliges
a court which has decided to develop the common law for other reasons to ensure that this independently justified
development also promotes human rights, and the values underlying it.
134 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 46.
135 2001 (4) SA 938 (CC).
136 See section 2.5.1.
137 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC).
138 Para 33.
139 Para 39.
140 Para 40.
141 Para 43.
142 Para 44.
143 Para 54.
144 Para 55.
145 2002 (6) SA 431 (SCA) paras 21–22.
146 Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA).
147 Para 38.
148 Paras 43–44.
149 Van Eeden v Minister of Safety and Security (Women’s Legal Centre Trust as Amicus Curiae) 2003 (1) SA 389 (SCA).
 Chapter 3

Delict in a multi-cultural society

3.1 Introduction

3.2 The constitutional window

3.3 The influence of African culture and values

3.4 The influence of Muslim and Hindu culture and values

3.5 Specific instances

3.5.1 Dependants’ action
3.5.2 Defamation
|
3.5.3 Remedies

3.6 Conclusion

3.1 Introduction
At the outset of this book we claim that the South African law of delict is a hybrid system,
grounded in Roman law, yet reflecting a variety of influences from other legal systems. We note
further that society’s legal convictions, or boni mores must express society’s views on what it
considers acceptable behaviour, and what it considers not. Some would argue, however, that the
claim is not entirely accurate, in that our law of delict reflects by and large imported legal cultures
and values without recognising and assimilating the rules and values of the country’s indigenous
peoples. The truth is that the customary law of delict continues to operate as a distinct system
which ‘lives side by side with the common law’ 1 and its impact on the common law has been
minimal. Others will point out that, similarly, Islamic law, applicable to a well-defined sector of
our society, is not afforded sufficient recognition.
The purpose of this chapter is to document and explore the extent to which the Constitution of
the Republic of SA, 1996 has stimulated internal transformation of our law of delict towards a
situation where its principles and rules could be considered a true amalgam that expresses the
values of the country’s entire population.2 Our aim is not to traverse specific customary-law delicts.
Instead, the focus in this chapter is on how, if at all, principles and values of African, Muslim and
Hindu cultures have come to influence and enrich our common-law principles of delict.

3.2 The constitutional window

The Constitution has created a window of opportunity for moulding an inclusive, plural South
African legal culture,3 which embraces and respects the dignity of all its citizens.4 To be legitimate,
the country’s legal culture (including its boni mores) must therefore, in appropriate circumstances,
recognise and reflect some of the values that inform the cultural practices of all its peoples.5 Yet, in
reality, references to indigenous customs and cultural values when applying delictual principles
have been scant.6
The Constitution obliges courts to apply customary law 7 provided that it is consistent with the
Bill of Rights and, further, that it has not been changed by legislation that deals specifically with
customary law.8 This means that for those people who structure their lives according to traditional
culture,9 either completely or only in respect of certain transactions,10 customary law 11 serves as a
separate legal system, to be applied to them in the first instance. However, being culture dependent,
there is no uniform system of customary law in our plural society and there are no fixed criteria for
determining when customary law is applicable to a particular situation.12 Regarding delict
specifically, ‘reference to the way parties live and thus their overall cultural orientation, has had a
strong influence on choice of law’. 13
Although they are also distinct systems, Islamic law and customs and Hindu law and customs
are not similarly recognised as systems that receive priority in specific instances.
Their role, as with ‘living’ 14 or ‘unofficial’ African customs,15 is to inform and enrich the normative
base upon which judges make discretionary value judgements in applying and developing the
common law of delict.
The constitutional gateway for multi-cultural values to become embedded in the law of delict
is two-pronged:
• Section 9(3) protects persons from discrimination on cultural grounds.
• Sections 30 and 31 entrench the right of cultural communities to practise and enjoy their
culture.

This means that an ‘ethos of tolerance, pluralism and religious freedom’, different from that which
in the past ‘informed the determination of the boni mores of the community’ is now required.16
Customary law and other cultural values must now be considered when courts develop the law 17
and there is increasing evidence that courts are willing to consider living customary law and other
unofficial laws in resolving disputes.18

Fosi v RAF19
As part of an extensive discussion on the role of customary law, the Court said:20

We now live in a constitutional democracy. Customary law should not only be

tolerated (as was the position in the past) but it must be recognised, applied and
married to the existing Roman-Dutch legal system currently in place in this
country.

It took the promulgation of an interim Constitution (The Constitution of the

Republic of South Africa Act 200 of 1993) that customary law became a matter of
constitutional importance in the legal history of this country. It was at this stage
 that it became apparent that customary law was now being treated as a
foundation of the South African legal system virtually on the same terms as
Roman-Dutch law. The position presently is that s 211(3) of the Constitution of
the Republic of South Africa, 1996, determines that all courts in South Africa
must apply customary law where appropriate, subject to the Constitution and
legislation that deals in particular with customary law. The Constitution is the
supreme law in this country. Finally, full recognition has been given to
customary law. The courts are obligated to apply it in disputes where applicable.
Full recognition and the obligatory application of customary law in instances
where it is indeed applicable comes with an added obligation to the
administrators of justice (magistrates and judges) to actively engage in the
development of customary law. I am thus constitutionally enjoined to develop
customary law and bring it to the same level reached by common law. The
plaintiff in this matter is an African (black) person. The deceased was a black
person. I fail to see why I should not apply customary law that governed them.

I have shown above that customarily the child who is financially able to do so is
under an obligation to maintain his needy parent. There is no reason, in my view,
why consideration should not be given to this portion of customary law in the
determination of liability of the Road Accident Fund towards a parent who has
lost a child in a motor vehicle accident caused by the negligent driving thereof. I
hold therefore that, even on this consideration, the Road Accident Fund cannot
escape liability towards the plaintiff in this matter.

PAUSE FOR Should cultural values be imposed on others?

REFLECTION If the South African boni mores (the societal values that underpin legal
policy decisions) are to reflect cultural values, would this not amount to
cultural practices being imposed upon persons who do not adhere to
such practices? On the other hand, could it not be argued that this is
exactly what has been happening all along in that a Western culture has
been imposed on indigenous people? Should any particular cultural
dominance be curtailed? How?

3.3 The influence of African culture and values

According to Mqeke,21 African legal tradition exhibits the following characteristics: customary and
religious foundations, communal traditions, collective responsibility, flexibility and reconciliation,
integration of law, religion and morality and symbolism in making the law an effective instrument
of social control. There is no clear distinction between delict and criminal law.22 An essential
feature of delictual obligations in customary law is that they exist in a group context and are
manifested in group rights and group liability.23
Customary law delicts cover:
• A variety of sexual wrongs (seduction,24 impregnation of unmarried women,25 adultery,26 as
well as intercourse with an ukungena partner,27 a widow, or a former customary law wife) 28
• Abduction of brides-to-be (ukuthwala) 29 and its associated bopha 30 claim
• Damage to movable or immovable property 31
• Assault 32
• Theft 33
• Defamation, in limited instances.34
The nature of these delicts and the circumstances in which they apply are fully discussed
elsewhere.35 Suffice it to say here that some of these delicts might have to be reconsidered for
compatibility with constitutional principles and values, in the same way as the common law
regarding adultery and breach of promise have been re-examined; 36 and the principles applicable to
these delicts have not been considered when courts deal with corresponding situations in terms of
the common law.

PAUSE FOR Should customary law delicts influence the general principles of
REFLECTION common law?
Should customary law delicts, although part of a separate legal regime,
influence common-law general principles in overlapping circumstances?
Consider, for example:
• Passing references have been made to the fact that a kraalhead’s liability is an
indication that vicarious liability is found in both common law and customary
law.37
• Both systems accept the concept of young persons being doli incapax, in that
household heads do not incur liability where children who caused harm lacked
the ability to distinguish between right and wrong.38

Both customary law and common law recognise liability for damage
caused by animals to persons and to property, also damage to crops.39
Should the principles applicable to the actio de pauperie and the actio de
pastu be modified in certain circumstances?

When courts apply or develop delictual principles, the role of customary law is still in its infancy.
Nonetheless, as will be seen in the rest of this chapter, the value of ubuntu-botho has already been
used to good effect.

TERMINOLOGY Ubuntu-botho
Mokgoro J described the concept in S v Makwanyane 1995 (3) SA
391 (CC), 1995 (6) BCLR 665 (CC) para 308:

Generally, ubuntu translates as ‘humaneness’. In its most

fundamental sense, it translates as personhood and
‘morality’. Metaphorically, it expresses itself in umuntu
ngumuntu ngabantu, describing the significance of group
solidarity on survival issues so central to the survival of
communities. While it envelops the key values of group
solidarity, compassion, respect, human dignity, conformity
to basic norms and collective unity, in its fundamental
sense it denotes humanity and morality. Its spirit
emphasises respect for human dignity, marking a shift from
confrontation to conciliation. In South Africa ubuntu has
become a notion with particular resonance in the building
of a democracy. It is part of our rainbow heritage, though it
might have operated and still operates differently in diverse
community settings. In the Western cultural heritage,
respect and the value for life, manifested in the all-
embracing concepts of ‘humanity’ and ‘menswaardigheid’
are also highly priced. It is values like these that [Section 35
of the Constitution] requires to be promoted. They give
 meaning and texture to the principles of a society based on
freedom and equality.

Botho is the Sesotho and Setswana equivalent of the Nguni

languages’ expression ubuntu. It bears the same essential
meaning.

3.4 The influence of Muslim and Hindu culture and values

Islamic law ‘comprises the ethos of religion, the precepts of ethics, the tenets of the law and the
bases of other disciplines’ moulded into a system that governs ‘all aspects of human life’;40 and
similarly, Hindu law is a culture-specific system of law that interlinks ‘religion, social and moral
rules, ethics, justice and the principles of law’. 41 Thus far the only area in delict in which Muslim
and Hindu culture and values have been given pertinent effect has been in respect of claims for loss
of support.

3.5 Specific instances

3.5.1 Dependants’ action

The dependants’ action is the area in which the most significant transformative development has
taken place and is the only sphere in which African, Muslim and Hindu cultures have been
integrated into mainstream delictual principles.
The common-law position was that customary unions and Islamic marriages, being potentially
polygamous, were not valid marriages 42 and so could not give rise to a legally recognised duty of
support in respect of partners to the relationship and between parents and children.43 Amod v
Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) 44
provided the watershed. In that case, the discriminatory effect of the common-law provision led to
the development of the law.45
Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality
Intervening)46
A widow who had been married in terms of Islamic law claimed damages for loss of support.
The discriminatory effect of the common law on dependants of persons married in terms of
Islamic law was a key reason for recognising the widow’s claim.47 The Court noted:48

[I]t was common cause that the Islamic marriage between the appellant and the
deceased was a de facto monogamous marriage; that it was contracted
according to the tenets of a major religion; and that it involved ‘a very public
ceremony, special formalities and onerous obligations for both parents in terms
of the relevant rules of Islamic law applicable’. The insistence that the duty of
support which such a serious de facto monogamous marriage imposes on the
husband is not worthy of protection can only be justified on the basis that the
only duty of support which the law will protect in such circumstances is a duty
flowing from a marriage solemnised and recognised by one faith or philosophy
to the exclusion of others. This is an untenable basis for the determination of the
boni mores of society. It is inconsistent with the new ethos of tolerance,
pluralism and religious freedom which had consolidated itself in the community
even before the formal adoption of the interim Constitution on 22 December
1993.
 I have no doubt that the boni mores of the community at the time when the
cause of action arose in the present proceedings would not support a
conclusion which denies to a duty of support arising from a de facto
monogamous marriage solemnly entered into in accordance with the Muslim
faith any recognition in the common law for the purposes of the dependant’s
action; but which affords to the same duty of support arising from a similarly
solemnised marriage in accordance with the Christian faith full recognition in
the same common law for the same purpose; and which even affords to
polygamous marriages solemnised in accordance with African customary law
exactly the same protection for the same purpose, (by virtue of the provisions of
section 31 of the Black Laws Amendment Act 76 of 1963 which reverses the
consequences of the Fondo judgment in respect of customary marriages). The
inequality, arbitrariness, intolerance and inequity inherent in such a conclusion
would be inconsistent with the new ethos which prevailed on 25 July 1993 when
the cause of action in the present matter commenced. The boni mores of the
community would at that time support the approach which gave to the duty of
support following on a de facto monogamous marriage in terms of the Islamic
faith the same protection of the common law for the purposes of the
dependant’s action, as would be accorded to a monogamous marriage
solemnised in terms of the Christian faith.

The Court emphasised that its decision was based on the de facto monogamous nature of
the marriage in the case before it. It left open whether the same result would have obtained ‘if
the deceased had been party to a plurality of continuing unions’.49 That issue is still to be
decided, but the Constitutional Court has suggested that it might be similarly unfairly
discriminatory if the right were not extended to polygynous Muslim marriages.50

In line with this approach, courts have not shied away of granting a remedy to dependent partners,
children, and parents in customary-law 51 and Islamic 52 relationships. Significantly, however, courts
have relied on African, Muslim and Hindu norms to extend the dependants’ action to co-habitation
relationships (gay and heterosexual), adopted children, close family members and ex-husbands.53
Underpinning this development is the contention that society’s culture and morality should
determine whether there is a ‘duty worthy of protection’ 54 which can arise voluntarily out of a
sense of duty.55 In JT v RAF 56 the Court noted:
A duty of support between de facto family members is one of those areas in which the law
gives expression to the moral views of society.57

Fosi v RAF58
A parent sued for loss of support of a deceased child, one of the issues being whether the
child had a legal duty to support and maintain the parent. The Court said:59

African law obligates a child who is financially able to do so to provide

maintenance to his/her needy parents. When an African (black) provides support
and education to his/her son/daughter, he/she is not only under a duty to do so
on the strength of the South African legal system, but custom also obliges such
a parent. In fact, in African tradition to bring up a child is to make for oneself an
investment in that when the child becomes a grown-up and is able to participate
in the labour market, that child will never simply forget about where he came
from. That child, without being told to do so, will make a determination (taking
into account the amount he/she earns, her travelling to and from work, food to
sustain himself and personal clothing, etc) of how much he must send home to
the parents on a monthly basis. This duty is inborn and the African child does
not have to be told by anybody to honour that obligation. In fact, that is the trend
 in almost all black families in rural areas including the so-called urban black
communities. In each family there would invariably be one or two sons or
daughters who is/are employed. Those children in employment provide their
individual parental home with hope in life in that they monthly and without fail
send money to their parents so that basic necessities of life are afforded by the
latter. It is for this reason that the plaintiff was puzzled on being asked in cross-
examination why the deceased sent her money. Her answer was rather telling:
‘Because the deceased knew where he was coming from’. The duty of a child to
support a needy and deserving parent is well known in indigenous/customary
law. It is observed by such children. There is always an expectation on the part
of a parent that his child will honour this duty.

In African law it is most certainly an actionable wrong on the part of the child
who is financially able not to provide support to his needy and deserving
parents. Quite apart from it being an actionable wrong, failure to maintain one’s
parents by a child who is financially able to do so is, in black traditional law,
contrary to public policy (contra bonos mores). The parent can successfully
proceed civilly against such a child in traditional courts. It is also a morally
reprehensible act to fail to maintain one’s own parents who are in need of such
maintenance. If the parents were to decide not to lodge a complaint before the
tribal court, but opt somehow to alert members of the immediate family about
this predicament, such a child would be ostracised and be looked down upon as
a person who has no ubuntu. The latter scenario is rather rare because as stated
above every African child is born with this duty consciousness never to forget
his/her roots. It is unacceptable to African traditional law that the death of a child
who is employed and who is conscious of his duty to support and sustain his
parent, should not entitle the parent who has lost such support as a result of the
untimely death of such a child consequent upon any wrongful act on the part of
anybody including an accident caused by a negligently driven motor vehicle (as
in the instant matter) to claim that support.

Osman v Road Accident Fund60

The deceased’s mother claimed damages from the Road Accident Fund on the basis that she
was indigent and that, while alive, her son had supported her. (Even though her husband was
still alive, they were separated and he did not support her.) Her son had been married and
she lived in the same house as he did. At issue in this instance, was whether the Court
should extend the common law so as to allow the mother’s claim. In the course of doing so,
the Court said:

There can be no doubt that in certain cultures such as Muslim or Hindu cultures,
amongst others, there is a similar duty upon children [as in customary law]61 to
support their parents … .

In these communities the family is not restricted to the nuclear family but rather
to the extended family. It is not uncommon for grand-parents or even an aunt
who is single to live with the family. The deceased lived with his mother and wife
in the same home. The plaintiff’s evidence was that she never worked at all. It is
clear that she was dependent upon him to the same extent as a child, had there
been one. In these societies there are hardly any old age homes or places where
old people can retire. This is not because these communities cannot afford to
build such institutions but rather because the societal mores scorns upon
children who do not take care of their aged parents.62

As in African culture and tradition there is a moral and social duty in Muslim and
Hindu cultures as well, which is family orientated in the sense of the extended
 family looking after its elders. Elderly parents often reside with one or other
child who supports them and caters for their wellbeing. Old age homes are
almost non-existent in these cultures. The social mores of such societies dictate
that parents and the elderly are cared for and where this is not done there is a
social stigma associated with it.63

The Court stated explicitly that it took it upon itself to extend the common law. While it is not clear
from the judgment, one can assume that Ms Osman and her son lived within a Muslim and/or
Hindu community and shared those cultural values. Should this judgment be read as extending the
common law to litigants from that community only? Or does the extension apply to all litigants,
irrespective of their cultural background? How would the values of equality and legal pluralism
require us to interpret this judgment? (Compare the views of Dlodlo J in Fosi v RAF, 64 quoted in
section 3.2 above, which appears to restrict that judgment to a particular sector of society only.)

JT v RAF65
A grandmother had adopted her teenage granddaughter when she was seven years old.
However, her biological father maintained a father-daughter relationship with her and
voluntarily continued to support her until his death in a motor vehicle accident. On being
sued, the Road Accident Fund contested liability on the basis that the deceased’s legal
obligation to support his child had been extinguished when the child was adopted. After an
extensive review of relevant case law, the Court said:66

It seems to me that these cases demonstrate that the common law has been
developed to recognise that a duty of support can arise, in a given case, from
the fact-specific circumstances of a proven relationship from which it is shown
that a binding duty of support was assumed by one person in favour of another.
Moreover, a culturally imbedded notion of ‘family’, constituted as being a
network of relationships of reciprocal nurture and support, informs the common
law’s appetite to embrace, as worthy of protection, the assumption of duties of
support and the reciprocal right to claim support, by persons who are in
relationships akin to that of a family. This norm is not parochial but rather is
likely to be universal, it certainly is consonant both with norms derived from the
Roman–Dutch tradition, as alluded to by Cachalia JA in Paixão v RAF supra67
and, no less, from norms derived from African tradition, not least of all as
exemplified by the spirit of Ubuntu, as mentioned by Dlodlo J in Fosi v RAF
supra.68

Society’s morality is a strong theme throughout this judgment as the Court sought to
determine whether there was a right worthy of protection. It is also clear that customary
values played a central role in the Court’s conclusion that the ‘common law ought to be
developed to embrace’ the norm that a duty of support exists between de facto family
members’.69

3.5.2 Defamation
In Mogale v Seima70 the Supreme Court of Appeal noted:
[I]ndigenous law also does not in general allow damages claims for defamation unless
allegations of witchcraft are involved.71

As a result, the opportunities for influencing the general principles of defamation were limited,
where they did occur, courts for many years followed a restrictive approach.
It has long been the law that the defamatory nature of material is assessed through the eyes of
right-thinking persons generally, which meant that courts would not have regard to the views and
opinions of a particular segment of society.72 So, in Omarjee v Post Newspapers (Pty) Ltd 73 the
Court could not take into account that comments which were not defamatory to women generally
could be defamatory in respect of the two Muslim women plaintiffs; and in HRH King Zwelithini of
KwaZulu v Mervis 74 the Court did not entertain the view that Zulu people might regard the term
‘isoka’ (ladies man) to be defamatory when referring to the King. In both these instances cultural
values were not considered in resolving the issues.
A sectoral approach does not suit a heterogeneous society 75 and so in Mohamed v Jassiem 76
the Appellate Division modified the applicable rule:77
[W]here the statement complained of is alleged to be defamatory only in the eyes of a
particular segment of society … it is the reaction of ‘right-thinking members of that segment
of society’ which becomes the yardstick rather than that of ‘right-thinking’ members of
society generally.

The rule change did not come about through any consideration of cultural values, but rather through
the recognition that the relevant norms of Muslim society were not anti-social or contra bonos
mores.78 The change to the common-law rule now allows courts greater freedom to consider
sectoral values in determining whether material ought to be considered defamatory.

3.5.3 Remedies
Although not expressly stated in the judgment, tolerance of defamatory conduct in customary law
appears to have been one of the factors which influenced the Court in Mogale v Seima79 to reduce
the damages award. There appears to be no other reported instance where this issue was considered.
By far the most significant influence of customary values on the common-law principles of
delict relate to the remedy of retraction and apology.
In Dikoko v Mokhatla 80 Sachs J said:81
In present day terms [ubuntu-botho] has an enduring and creative character, representing
the element of human solidarity that binds together liberty and equality to create an
affirmative and mutually supportive triad of central constitutional values. It feeds
pervasively into and enriches the fundamental rights enshrined in the Constitution.

In that case, Mokgoro J and Sachs J relied on the restorative nature inherent in ubuntu-botho to
advocate for retraction and apology to be recognised as an appropriate remedy for defamation.82 In
particular, Mokgoro J noted: 83
The primary purpose of a compensatory measure, after all, is to restore the dignity of a
plaintiff who has suffered the damage and not to punish a defendant. A remedy based on the
idea of ubuntu or botho could go much further in restoring human dignity than an imposed
monetary award in which the size of the victory is measured by the quantum ordered and
the parties are further estranged rather than brought together by the legal process. It could
indeed give better appreciation and sensitise a defendant as to the hurtful impact of his or
her unlawful actions, similar to the emerging idea of restorative justice in our sentencing
laws.

Although these were minority views, they did stimulate similar sentiments in subsequent
judgments. In Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as
Amici Curiae) 84 the Constitutional Court confirmed that retraction and apology is an appropriate
remedy in certain instances. The Court noted that in this regard customary law and tradition, and
Roman-Dutch law had similar roots which with ‘mature reflection and consideration’ could be
melded into a single system of law.85 It suggested that aspects of restorative justice could be
considered when the amount of damages is measured, but that this should be done on a case-by-
case basis.86

PAUSE FOR Should ubuntu-botho play a more prominent role in the law of
REFLECTION delict?
Keep and Midgley suggest:87

While the emphasis thus far has been on using ubuntu-

botho values in finding an appropriate remedy, there is also
scope for the philosophy to permeate other areas in the law
of delict – the wrongfulness element, for example, an area
where norms are prominent and in which the Constitution
has already had substantial influence. In future one might
possibly find ubuntu-botho being used, not only in cases
involving bodily integrity, dignity, privacy and reputation,
but also to reinforce responsibility – of the state and of
private persons – where vulnerable and marginalised
sectors of society are harmed.

They also ponder a more global role for ubuntu-botho:88

[I]f an emphasis on African values can shape and enhance

human rights jurisprudence in South Africa, and give
dignity to ideas that emanate from Africa without offending
or eroding contributions of other societies, why, in the
spirit of ubuntu-botho, should African values also not
contribute more in the international domain? Already
restorative criminal justice expressing values similar to
ubuntu-botho has been espoused in a number of countries
… and there are many similarities between ubuntu-botho
and the ‘poldermodel’ of The Netherlands ... Such
harmonisation might not be as daunting as it may at first
seem.

The authors believe that ubuntu-botho has a significant transformative

role to play within the South Afican law of delict. Is this view realistic?

3.6 Conclusion
We have not been concerned, in this chapter, with the nature of legal pluralism.89 Our aim has been,
more narrowly, to point to some of the areas in which the common law has evolved, or is capable
of evolving, towards a law of delict that is more inclusive of cultural nuances than in the past. Our
contention is that, apart from the dependants’ action and to a lesser extent, retraction and apology in
defamation cases, progress has been piecemeal and slow; and that courts have not had much cause
to grapple with these issues. The door is not closed, however, and there is room for the emergence
of a more inclusive and legitimate legal culture. We agree with Nhlapo when he suggests:
Essentially, the proposal is to grasp more boldly the merits of infusing African values into
South Africa’s legal and moral arena. This necessitates in the first instance an
acknowledgement of the existence of African values, which display a different emphasis
 from the Western world view, and an acceptance that these values do have a positive
contribution to make in creating the new South African society.90

PAUSE FOR A decolonised and afro-centric law of delict?

REFLECTION Student protests since 2015 have not been only about free tertiary
education. In addition, students have demanded decolonised and afro-
centric education, among other things. Nhlapo argues:91

[I]n 20 years of democracy both the legislature and the

judiciary have too often opted for the easy path, in contrast
with the more difficult path that requires us to take
seriously the Constitution’s attempt to infuse African
values into the common culture of the new South Africa,
including its legal culture.

Our law of delict is steeped in the Roman-Dutch tradition. While there are
signs of some afro-centricism, has the infusion of a human rights culture
contributed to decolonising the law of delict? What else could realistically
be done?

1 Gumede (born Shange) v President of the Republic of South Africa 2009 (3) SA 152 (CC); 2009 (3) BCLR 243 (CC)
para 22. See also Mthembu v Letsela 1997 (2) SA 936 (T) at 936B–C.
2 This is in line with Nhlapo’s view that courts and lawmakers should accept that customary values ‘have a positive
contribution to make in creating the new South African society’ (Nhlapo ‘Customary law in post-apartheid South
Africa: constitutional confrontations in culture, gender and “living law”’(2017) 33(1) SAJHR 1 at 22.
3 Keep and Midgley ‘The Emerging Role of Ubuntu-botho in Developing a Consensual South African Legal
Culture’ (2007) 29 at 29–30.
4 The Investigating Directorate: Serious Economic Offences v Hyundai Motor Distributors (Pty) Ltd In Re Hyundai
Motor Distributors (Pty) Ltd v Smit No 2001 (1) SA 545 (CC); 2000 (10) BCLR 1079 (CC) para 21. See also Amod v
Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA 421
(SCA); 1999 (4) SA 1319 (SCA) para 22.
5 Keep and Midgley (2007), at 29 and 56, argue that there is a ‘real need in South Africa for legitimating the legal
system and to develop a legal culture that expresses values originating in African societies’ and that a ‘break from past
domination of one school of thought over another needs to be emphasised’.
6 Van der Walt and Midgley Principles of Delict 4 ed (2016) paras 23 and 24.
7 Van der Walt and Midgley (2016) point out (para 23 fn 1) that a distinction is drawn between ‘official customary
law’ (primarily the law found in the written sources of customary law) and ‘living customary law’ (the law as
evidenced by current practices in communities: Bhe v Magistrate, Khayelitsha (Commission for Gender Equality as
Amicus Curiae); Shibi v Sithole; South African Human Rights Commission v President of the Republic of South Africa
2005 (1) SA 580 (CC), 2005 (1) BCLR 1 (CC) paras 87 and 152). See, generally, Himonga and Nhlapo (Eds) African
Customary Law in South Africa (2014) at 23–39. Although the traditions and practices of various cultural
communities will differ, there remain sufficient similarities and communalities to speak of an African legal tradition
(Mqeke Customary Law and the New Millennium (2003) at 16).
8 Section 211(3) of the Constitution of the Republic of SA, 1996. See also sections 31(2) and 39(3). Section 6 of the
Recognition of Customary Marriages Act 120 of 1998, which gives a customary law wife the right to sue and be sued,
is an example of legislation that has altered customary law. Although the Act has also ‘virtually abolished the custom
of ukuthwala’ (Mqeke (2003) at 115), it is uncertain whether the right to claim compensation has similarly fallen
away.
9 Mthembu v Letsela 1997 (2) SA 936 (T) at 944.
10 For example, where persons who overtly live a modern urban lifestyle participate in the payment of lobolo and its
associated practices.
11 Defined in section 1 of the Recognition of Customary Marriage Act 120 of 1998 as ‘customs and usages traditionally
observed among the indigenous African peoples of South Africa and which form part of the culture of those peoples’.
See also Gumede (born Shange) v President of the Republic of South Africa 2009 (3) SA 152 (CC); 2009 (3) BCLR
243 (CC) para 23.
12 Himonga and Nhlapo (2014) discuss some applicable factors at 83–87. See also Bennett Customary Law in South
Africa (2004) at 51–57.
13 Bennett (2004) at 55. See also Ramothata v Makhothe 1934 NAC (N&T) at 74.
14 ‘Living customary law is the law that is actually applied by indigenous people. This law often conflicts with the
official customary law that is applied by the State courts or entrenched in legislation. The Constitutional Court
increasingly takes cognisance of living customary law.’ (Rautenbach and Bekker (Eds) Introduction to Legal
Pluralism in South Africa 4 ed (2014) 5 at fn 3. See also page 14.) See, generally, Himonga and Nhlapo (2014) at 23
–39; Nhlapo ‘Customary law in post-apartheid South Africa: constitutional confrontations in culture, gender and
“living law”’ (2017) 33(1) SAJHR at 1–24.
15 ‘Living customary law is the law that is actually applied by indigenous people. This law often conflicts with the
official customary law that is applied by the State courts or entrenched in legislation. The Constitutional Court
increasingly takes cognisance of living customary law.’ (Rautenbach and Bekker (2014) at 5 fn 3). See, generally,
Himonga and Nhlapo (2014) at 23–39; Nhlapo (2017) 33(1) SAJHR at 1–24.
16 Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA
421 (SCA); 1999 (4) SA 1319 (SCA) paras 20 and 21.
17 S v Makwanyane 1995 (3) SA 391 (CC), 1995 (6) BCLR 665 (CC) paras 365–383; Mthembu v Letsela 1997 (2) SA
936 (T) at 944; Bennett Customary Law in South Africa (2004) at 43.
18 Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA
421 (SCA); 1999 (4) SA 1319 (SCA); Daniels v Campbell NO 2004 (7) BCLR 735 (CC); Hassam v Jacobs NO 2009
(5) SA 572 (CC); Rautenbach and Bekker (2014) at 5 and 14.
19 2008 (3) SA 560 (C).
20 Paras 23–25.
21 Mqeke (2003) at 16–27.
22 Rautenbach and Bekker (2014) at 157.
23 Mqeke Basic Approaches to Problem Solving in Customary Law: A Study of Conciliation and Consensus among the
Cape Nguni (1997) at 157–159; Mqeke (2003) at 114–117; Himonga and Nhlapo (2014) at 197.
24 Defloration of a virgin, according to Mqeke, is an infringement of a patrimonial right, for ‘an unmarried girl
represents to her family head the value of the lobolo he expects to receive for her’ (Mqeke (1997) at 159 fn 10; Mqeke
(2003) at 116). See also Rautenbach and Bekker (2014) at 161–162.
25 Rautenbach and Bekker (2014) at 162.
26 ‘In customary law, adultery can contrary to the South African common law, in the first place only be committed with
the wife, not the husband of a customary marriage.’ (Rautenbach and Bekker (2014) at 163–165); ‘… the violation of
the family head’s right which “custom has translated into pecuniary value’’ ’ (Mqeke (1997) at 159 fn 11; Mqeke
(2003) at 116).
27 Ukungena is a custom ‘according to which a widow enters into a union with a brother or half-brother of the deceased
in order to raise seed for the deceased’ (Himonga and Nhlapo (2014) at 294). See also Rautenbach and Bekker (2014)
at 165.
28 Rautenbach and Bekker (2014) 163 at 165; Mqeke (1997) at 159; Mqeke (2003) at 116.
29 The ‘forcible seizure of an unmarried girl and her abduction without her parent’s consent for the purpose of marrying
her’ (Mqeke (1997) at 158 fn 8; Mqeke (2003) at 115). See also Olivier, Church, Mqeke, Bekker, Mwambene,
Rautenbach and Du Plessis LAWSA (2009) Indigenous Law Vol 32 (2) paras 181–182; Rautenbach and Bekker (2014)
at 165–166.
30 When marriage does not materialise after thwala, a beast is paid to assuage the thwalaed girl’s wounded feelings
(Mqeke (1997) at 158 fn 9).
31 Rautenbach and Bekker (2014) at 167. ‘A distinction is made between damage caused by animals and that caused by
humans’ (LAWSA (2009) Vol 32(2) para 174).
32 Rautenbach and Bekker (2014) at 168.
33 Rautenbach and Bekker (2014) at 167–168.
34 Originally, defamation was not considered to be a delict, except for allegations of witchcraft (LAWSA paras 78 and
121–124); Mogale v Seima 2008 (5) SA 637 (SCA) para 9. The KwaZulu and Natal Codes of Zulu Law widened the
scope, however (Himonga and Nhlapo (2014) at 199–200). See LAWSA (2009) Vol 32(2) para 183; Rautenbach and
Bekker (2014) at 166.
35 For example, Olivier, Church, Mqeke, Bekker, Mwambene, Rautenbach and Du Plessis LAWSA (2009) Vol 32(2);
Himonga and Nhlapo (2014) at 197–210; Rautenbach and Bekker (2014) at 157–169.
36 Van Jaarsveld v Bridges 2010 (4) SA 558 (SCA) (breach of promise); DE v RH 2015 (5) SA 83 (CC) at 18 (adultery).
See also Himonga and Nhlapo (2014) at 206 for views regarding the constitutionality of the customary law delict of
seduction.
37 K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC) para 24.
38 Skenjana v Geca 6 (1928) NAC 4; Bennett (2004) at 325.
39 LAWSA (2009) Vol 32(2) para 174ff; Rautenbach and Bekker (2014) at 167.
40 Rautenbach and Bekker (2014) at 351.
41 Rautenbach and Bekker at 259.
42 Ngqobela v Sihele (1892–1893) 10 SC 346 at 352, SANTAM Bpk v Fondo 1960 (2) SA 467 (A) and Nkabinde v SA
Motor & General Insurance Co Ltd 1961 (1) SA 302 (N) (in respect of customary unions); and Seedat’s Executors v
The Master (Natal) 1917 AD 302 and Ismail v Ismail 1983 (1) SA 1006 (A) (in respect of Islamic marriages).
43 Subsequent statutory intervention (section 31 of the Black Laws Amendment Act 76 of 1963) addressed the negative
consequences in respect of customary marriages.
44 [1999] 4 All SA 421 (SCA); 1999 (4) SA 1319 (SCA).
45 Para 23.
46 [1999] 4 All SA 421 (SCA); 1999 (4) SA 1319 (SCA).
47 Paras 20 and 23.
48 Paras 20 and 23.
49 Para 24.
50 Hassam v Jacobs NO 2009 (11) BCLR 1148 (CC); 2009 (5) SA 572 (CC) para 39.
51 Zimnat Insurance Co Ltd v Chawanda 1991 (2) SA 825 (ZS); Fosi v RAF 2008 (3) SA 560 (C); Mnguni v RAF 2015
JDR 1723 (GP); Seleka v RAF 2016 (4) SA 445 (GP).
52 Osman v Road Accident Fund 2015 (6) SA 74 (GP).
53 Metiso v Padongelukfonds 2001 (3) SA 1142 (T); Du Plessis v Road Accident Fund 2003 (11) BCLR 1220 (SCA);
2004 (1) SA 359 (SCA); MB v NB 2010 (3) SA 220 (GSJ); Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA);
2012 (6) SA 377 (SCA); Verheem v RAF 2012 (2) SA 409 (GNP).
54 Du Plessis v Road Accident Fund 2003 (11) BCLR 1220 (SCA); 2004 (1) SA 359 (SCA) para 17; Paixão v Road
Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA) para 12; Verheem v RAF 2012 (2) SA 409 (GNP)
para 12; JT v RAF 2015 (1) SA 609 (GJ) paras 19–22, 24, 26 and 29.
55 JT v RAF 2015 (1) SA 609 (GJ) paras 26 and 29; Jacobs v RAF 2010 (3) SA 263 (SE) para 7.
56 2015 (1) SA 609 (GJ).
57 JT v RAF 2015 (1) SA 609 (GJ) para 29. See also Mnguni v RAF 2015 JDR 1723 (GP) at 11.
58 2008 (3) SA 560 (C). See also Seleka v RAF 2016 (4) SA 445 (GP) at 454 where the Court followed Fosi v RAF and
held that in Tswana law and custom both sons and daughters had a duty to maintain parents.
59 Paras 16–17.
60 2015 (6) SA 74 (GP).
61 See Fosi v RAF 2008 (3) SA 560 (C).
62 Paras 20–21.
63 Para 24.
64 2008 (3) SA 560 (C) at 24.
65 2015 (1) SA 609 (GJ).
66 Para 26. See also para 17.
67 Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA).
68 Fosi v RAF 2008 (3) SA 560 (C).
69 Paras 29 and 30.
70 2008 (5) SA 637 (SCA).
71 Para 9; Mqeke ‘The Customary Law of Defamation of Character with Specific Reference to the Law of Xhosa-
speaking Peoples of Transkei and Ciskei’ (1981) 44(3) THRHR at 425; Burchell The Law of Defamation in South
Africa (1985) at 22–23.
72 G A Fichardt Ltd v The Friend Newspapers Ltd 1916 AD 1 at 9; Conroy v Nichol 1951 (1) SA 653 (A) at 662;
Mohamed v Jassiem 1996 (1) SA 673 (A) at 704.
73 1967 (2) PH J33 (D).
74 1978 (2) SA 521 (W) at 528.
75 Mohamed v Jassiem 1996 (1) SA 673 (A) at 674; Van der Walt and Midgley (2016) para 106.
76 1996 (1) SA 673 (A).
77 At 704B–C.
78 At 703D.
79 2008 (5) SA 637 (SCA).
80 2006 (6) SA 235 (CC).
81 Para 113. See also Port Elizabeth Municipality v Various Occupiers 2005 (1) SA 217 (CC); 2004 (12) BCLR 1268
(CC) para 37.
82 Paras 68–69 and 112–121.
83 Para 68.
84 2011 (3) SA 274 (CC); 2011 (6) BCLR 577 (CC) paras 199–203.
85 Para 200.
86 Para 202. But see The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae) 2011 (4) SA 191 (CC);
2011 (8) BCLR 816 (CC) para 132 where the Court commented as follows on Le Roux v Dey (Freedom of Expression
Institute and Restorative Justice Centre as Amicus Curiae) 2011 (3) SA 274 (CC); 2011 (6) BCLR 577 (CC):
 There this Court found that ordering an apology was an appropriate measure of restorative justice in
a case involving ruptured personal relationships, where the defendants actionably impaired the
dignity of the plaintiff.
87 Keep and Midgley (2007) at 47.
88 Keep and Midgley (2007) at 48 fn 121.
89 See, for example, Rautenbach and Bekker (2014) at 5–16.
90 Nhlapo (2017) 33(1) SAJHR 1 at 22.
91 Nhlapo (2017) 33(1) SAJHR 1 at 23.
 Chapter 4

Harm

4.1 Introduction

4.2 Patrimonial and non-patrimonial harm

4.3 Pain and suffering

4.4 Infringement of a personality interest

4.4.1 Bodily integrity
4.4.2 Dignity
4.4.3 Privacy
4.4.4 Identity |
4.4.5 Reputation

4.5 Personality rights and constitutional rights

4.6 Conclusion

Figure 4.1 Harm as an element of a delict

4.1 Introduction
For a delict to arise, there must be some actual or potential harm. By pursuing a delictual remedy,
plaintiffs seek compensation or reparation for damage to, or the loss or harm resulting from a
violation of their interests. When seeking an interdict, plaintiffs wish to prevent someone from
threatening their interests. So, the harm element is the cornerstone of the law of delict, and the
fundamental point of departure.1 At the outset, therefore, one needs to determine whether the
plaintiff has an interest that the law of delict protects and, if so, whether that interest has been
violated in a negative way.

TERMINOLOGY Harm, loss, damage and damages

This element is described in various ways in literature and case
law, for example, damage, damages, loss, harm, right, interest,
and injury. This inconsistent use of terminology creates conceptual
difficulties and might lead to confusion between the element of a
delict (harm, loss or damage) and the delictual remedy (damages).
Damages are what plaintiffs claim when they have suffered
damage, harm or loss. To avoid possible confusion, we will not
use the word ‘damage’ to describe the element of a delict. Instead,
we will use the words ‘harm’ or ‘loss’ interchangeably, according to
the context. We will continue to describe the remedy as
‘damages’. Occasionally, we will use the word ‘damage’ to signify
the factual disturbance of an interest to property, for example, ‘he
damaged the motor vehicle’, or ‘the harm element was satisfied
because there was damage to the plaintiff’s motor vehicle’.

On hearing that a person wishes to seek compensation, one’s first reaction should be to ask, ‘What
for?’ This question is important, for it directs one’s mind towards the appropriate delictual action:
• If the harm is patrimonial, compensation will follow only if the elements of the lex Aquilia are
present.
• If one has experienced pain and suffering, the elements of the Germanic remedy must be met.
• For a violation of a personality interest, the elements of the actio iniuriarum must be satisfied.

In this way, one can determine the route to follow and which requirements need to be met for any
action to be successful.
The various actions are not mutually exclusive. It is possible for a person to suffer various
forms of harm at the same time, which means that a person can simultaneously claim remedies
under more than one action. This commonly occurs when a person is injured in an accident: the
person suffers patrimonial loss when he or she is hospitalised and has to pay the medical bills; and
also suffers harm in the form of pain and suffering in respect of the pain experienced and loss of
amenities of life, if any. An assault may similarly give rise to patrimonial loss and pain and
suffering; and in some instances, it might also constitute an iniuria.2
In Chapter 1 we indicated that one can broadly divide the elements of a delict into two
categories: those that are mainly factual and those that are mainly normative. The harm element
falls into the first category and the plaintiff, therefore, has to produce evidence to prove that he or
she has suffered harm. However, this element is not solely factual. There are normative or policy
components to it. Not all factual violations of a person’s interests, whether patrimonial or non-
patrimonial, will necessarily be actionable. In some instances, the law says: ‘You have indeed
suffered harm, but the law of delict does not recognise that as one of the types of harm for which a
person can be compensated. You might have a remedy elsewhere, but not in delict.’ The decision
whether or not to recognise the type of harm is based on policy considerations, and so has strong
normative features.
The law is not static, however. Policy and generally accepted standards of behaviour vary as
society grows and the decisions that reflected appropriate standards in the past may not necessarily
reflect the standards of today. For example, for many years the Aquilian action compensated loss
associated with physical injury to person or property, but not psychiatric injury. Psychiatric injury
was recognised as a form of harm only towards the end of the nineteenth century, and initially only
where there had been a real apprehension of physical danger to the person who suffered the shock.
In time that requirement was watered down, first to remove the personal danger requirement 3 and
later to include ‘hearsay cases’. 4 It is also only recently that post-traumatic stress disorder was
considered worthy of being accepted as a form of psychiatric injury.5

Fourie v Naranjo6
A woman witnessed a dog (Bruno) attacking her domestic worker and savaging her husband
who had come to the worker’s assistance. For some time thereafter she could not sleep,
‘being haunted by the picture in her mind of Bruno with blood and pieces of flesh in his
mouth’.7 She developed a short-term stutter and the incident also affected her ability to drive
a motor vehicle. She had to obtain professional help. Her emotional shock claim was
instituted in terms of the actio de pauperie, which in the past had been used only for claims
where domesticated animals caused physical injury to plaintiffs. The defendant contended
that she was not entitled to do so.
The Court found no case law restricting pauperies to damages resulting from physical
injuries. On the contrary, in addition to claims for physical injury, past plaintiffs could claim
‘for subsequent physical disorders caused by the nervous shock’; and where death ensued,
dependants could claim loss of support.8 Given these developments, the Court extended the
scope of the actio de pauperie to include claims resulting from emotional shock.

Conversely, policy changes do not always favour the extension of liability. Courts continue to
refuse to recognise loss of comfort and support of a loved one, or grief associated with the loss of a
loved one, as actionable harm.9 Similarly, courts do not recognise mental distress or inconvenience
that is not associated with bodily injury.10 This does not mean that these forms of harm might not be
recognised in the future, however. For example, for some time pain and suffering had to be
associated with a person’s own physical injury. There are signs, however, that pain and suffering
associated with another’s injury might now be worthy of compensation.
Sometimes policy justifications for actions that could have been brought in the past no longer
exist, as happened with the action for breach of promise. In Van Jaarsveld v Bridges 11 Harms DP
said:
Courts have not only the right but also the duty to develop the common law, taking into
account the interests of justice and at the same time to promote the spirit, purport and
objects of the Bill of Rights. In this regard courts have regard to the prevailing mores and
public policy considerations … . I do believe that the time has arrived to recognise that the
historic approach to engagements is outdated and does not recognise the mores of our time,
and that public policy considerations require that our courts must reassess the law relating
to breach of promise.

The Court found that breach of promise did not in itself give rise to a claim and that an action can
lie only if the elements of the actio iniuriarum are met, which means that the breach of promise has
to be accompanied by some insulting behaviour.12
From the discussion in this section we can see that the harm element, as with all other elements
of a delict, serves not only to fix liability, but also to control its extent. It is for this reason that the
enquiry into harm is both factual and normative. As Neethling, Potgieter and Visser point out:
 … only harm in respect of legally recognised patrimonial and non-patrimonial interests of a
person qualifies as (harm).13

PAUSE FOR Recognising harm

REFLECTION What is the difference between the policy decision not to recognise harm
and the policy decision made when the wrongfulness of an interest
infringement is determined?
When considering whether harm is actionable, the question to
consider is whether the person has suffered harm that is recognised in
delict. For example, if parents are claiming pain and suffering because
their child is lying in a coma, and the situation is unbearable for them, the
law would say: ‘The law of delict does not recognise that type of harm,
because it is not associated with any bodily injury to the people who wish
to claim.’ However, the child’s claim for pain and suffering is actionable.
When considering wrongfulness, the issue is whether the harm or
violation of an interest (which the law has already considered to be
actionable harm) has occurred in a legally reprehensible way. In other
words, should the law entertain and allow the plaintiff’s claim in this
instance? In some instances, the bodily injury that a plaintiff has suffered,
which is actionable harm, is also caused wrongfully. In other instances,
that same category of actionable harm is not wrongful, for example, if the
defendant had acted in self-defence. In such a case, the plaintiff would
have proved that the harm element, but the action fails because the
wrongfulness requirement was not met.

Union Government (Minister of Railways and Harbours) v Warneke14

A woman died in a train accident, and her husband sued the Minister for loss of comfort and
companionship, and the loss of his wife’s assistance in caring for and bringing up their
children.
The Court found that the lex Aquilia served to compensate only in respect of calculable
pecuniary harm. It could not attach such an economic value to the loss of his wife’s ‘comfort
and society’, and so the husband’s claim failed in that regard. De Villiers CJ said:15

The loss of the comfort and society of a wife does not appear to me to be a
pecuniary loss at all. It is a deprivation, which in many cases transcends, in the
grief, distress and discomfort which it occasions, any other loss which a man
can sustain, but it is not a loss which, for purposes of compensation, as distinct
from retribution, is capable of being calculated in money.

However, the extent to which the husband could show that his deceased spouse had
provided assistance in caring for, and supporting and educating their children, he could claim
compensation for any expenditure incurred to provide care and assistance similar to that
which his wife had given him during her lifetime. Innes J said:16

[T]here is nothing inconsistent with the principles of our law in allowing a

husband who can show that his pecuniary expenditure in connection with the
maintenance of his children has been directly and necessarily increased owing
to the death of this wife, to claim damages against the person who has
negligently caused her death. No Roman-Dutch authority goes so far, but one is
led to that conclusion by giving effect to principles well recognised by them all.
And there is no direct authority to the contrary.
 This case involved two types of claim: one patrimonial and one non-patrimonial. The Court
recognised the patrimonial claim, but did not consider the non-patrimonial harm to be
actionable under the lex Aquilia. The Aquilian action was confined ‘to cases in which a
calculable pecuniary loss has been actually sustained’.
The Court also considered whether the actio iniuriarum applied, but found that since there
was no injury done to, or dishonour brought upon the husband, there was no cause to satisfy
any injured feelings.
In this judgment the normative component of the harm element comes clearly to the fore.
The Court had to decide whether the two types of harm were actionable in the absence of
any previous authority.
Innes J also said:17

It is possible that the plaintiff may prove that after making allowance for the fact
that he no longer has to support his wife, the arrangements necessitated to
replace her supervision and assistance in the upbringing of the children entail a
pecuniary loss.

So, in considering whether the husband had sustained patrimonial harm, both the savings
and the additional expenses arising from the wife’s death should be considered. If the savings
exceed the expenses, he would have suffered no harm (which is what in fact happened in
Santam Insurance Co Ltd v Fourie).18

• The first quotation above contains an indication of the purpose of the lex Aquilia. Could the
statement also apply to the law of delict in general?
• It is implied in the text that the law might recognise inconvenience related to actionable harm.
How would it do this? Is there any reason why a person should not be able to sue in delict for
either the inconvenience of not having a vehicle while his or her damaged vehicle is being
repaired, or the inconvenience associated with having to change holiday plans?

PAUSE FOR The interplay between harm (‘damage’) and the remedy (‘damages’)
REFLECTION It is not always easy to confine delictual concepts into clear, separate
categories, and there are often some overlaps or similarities in approach.
For example, the element (harm) and the remedy (damages) are related.
The harm element of a delict has been satisfied if:
• Harm was suffered (factual aspect)
• The harm is legally recognised, or actionable (normative aspect).

At this stage, when the element is being considered, only the question of
whether actionable harm has occurred is important, not the extent of
harm. It is only once all the elements of a delict are present that one
enquires into the remedy and when the extent of the harm, commonly
referred to as the ‘quantum of damages’, becomes important. In
borderline instances, where the first bullet point is not obviously satisfied,
one would use the sum-formula approach to determine whether harm has
occurred. (See section 4.2 below.) The sum-formula is also a
fundamental device for determining quantum. The fact that the same
methodology is used does not make the concepts the same; it simply
shows that there is consistency in approach.

4.2 Patrimonial and non-patrimonial harm

It is important to establish that the plaintiff’s harm also constitutes actionable harm. In the law of
delict, harm falls into one of two broad categories: patrimonial harm and non-patrimonial harm.
The latter is further divided into two categories: pain and suffering, and infringement of personality
interests.
In short, patrimonial harm arises when a person’s patrimony is affected in a negative way. The
concept of patrimony, or financial estate, is associated with what the Romans called the universitas,
which includes not only a person’s current assets and liabilities, but also financial aspects
associated with rights and obligations that might arise in the future. To ascertain whether there has
been harm, one must consider the value of the plaintiff’s estate as a whole prior to the incident that
precipitated the claim and compare it to the plaintiff’s estate after the event. If there is a negative
impact, harm has occurred. This is known as the ‘sum-formula’ approach. Exactly the same
approach is used to calculate the extent of the diminution of the estate and the amount of damages
to award. However, in practice this theoretical explanation to ascertain whether patrimonial harm
has occurred is seldom used. For example, a doctor’s bill, or the cost of repairing a damaged motor
vehicle, is an expense that did not form part of the estate prior to the event and clearly now renders
the plaintiff poorer.
Patrimonial harm falls into three broad categories:
• Financial loss associated with personal injury
• Financial loss associated with damage to property
• Financial loss that is not associated with the plaintiff’s personal injury or damage to property
and is accordingly purely economic.

Jowell v Bramwell-Jones19
Jowell was a capital beneficiary of a testamentary trust established by his late father, Dr
Jowell. Among the defendants were a stockbroker, a chartered accountant, an attorney and a
supplier of financial services. The purpose of the trust was to provide for Dr Jowell’s wife until
her death, whereupon the capital would devolve upon the beneficiaries. Mrs Jowell sought
and acted upon the defendants’ advice on a financial scheme. No-one consulted the
beneficiaries in this regard. The effect of the scheme was to decrease the value of the
plaintiff’s vested right by approximately R10m. Jowell then sued the defendants, who
excepted on the basis that the claim was for prospective loss alone. At issue was whether a
plaintiff could sue for loss that might arise in the future without having incurred any actual loss
by the time the action is brought.
Scott JA said the following about the harm element:20

The element of damage21 or loss is fundamental to the Aquilian action and the
right of action is incomplete until damage is caused to the plaintiff by reason of
the defendant’s wrongful conduct (see Oslo Land Corporation Ltd v The Union
Government 1938 AD 584 at 590; Evins v Shield Insurance Co Ltd 1980 (2) SA
814 (A) at 838H–839C). This applies no less to claims arising from pure
economic loss than it does to claims arising from bodily injury or damage to
property (see Siman & Co (Pty) Ltd v Barclays National Bank Ltd 1984 (2) SA 888
(A) at 911B–D). Whether a plaintiff has suffered damage or not is a fact which,
like any other element of his cause of action and subject to what is said below,
must be established on a balance of probabilities. Once the damage or loss is
established, a court will do its best to quantify that loss even if this involves a
degree of guesswork. (See Turkstra Ltd v Richards 1926 TPD 276 at 282–283.)
However, a distinction must be drawn between accrued or past damage, or loss
on the one hand and prospective damage or loss on the other, the latter being
damage or loss which has not yet materialised. Delictual actions, which include
claims for prospective loss, are not uncommon, particularly in the case of
actions arising out of bodily injuries where the prospective loss is inevitably
 accompanied by some accrued or past loss. When dealing with such claims,
however, the courts have not required the plaintiff to prove on a preponderance
of probability that such a loss will occur or arise; instead they have made a
contingency allowance for the possibility of the loss. (See Blyth v Van den
Heever 1980 (1) SA 191 (A) at 225E–226B where Corbett JA cites with approval a
passage in the judgment of Colman in Burger v Union National South British
Insurance Company 1975 (4) SA 72 (W) at 75D–G.) The underlying reason for
such an approach is probably the ‘once and for all’ rule, which compels a
plaintiff who has suffered accrued or past damage to institute action in order to
avoid the running of prescription; in other words he is precluded from waiting to
see if the prospective loss will occur. In Coetzee v SA Railways & Harbours 1933
CPD 565, it was held that a person cannot sue solely for prospective damages.
Gardiner JP, with whom Watermeyer J concurred, expressed himself at 576 as
follows:

The cases, as far as I have ascertained, go only to this extent, that if a

person sues for accrued damages, he must also claim prospective
damages, or forfeit them. But I know of no case which goes so far as to
say that a person, who has as yet sustained no damage, can sue for
damages which may possibly be sustained in the future. Prospective
damages may be awarded as ancillary to accrued damages, but they
have no separate, independent force as ground of action.

This approach has been the subject of some criticism. Boberg, The Law of Delict at 488, contends
that there is no reason why a person cannot sue solely for prospective loss provided he can establish
the future loss on a balance of probabilities, although not necessarily the quantum of his claim.22
The advantage of the approach adopted in the Coetzee case is of course the certainty it provides. If
an action for loss which is prospective is completed only when the loss actually occurs, prescription
will not commence to run until that date and a plaintiff will generally be in a position to quantify
his claim. To the extent there may be additional prospective loss the court will make a contingency
allowance for it. On the other hand, if the completion of an action for prospective loss entitling a
person to sue is to depend not upon the loss occurring but upon whether what will happen in the
future can be established on a balance of probabilities, it seems to me that the inevitable uncertainty
associated with such an approach is likely to prove impractical and result in hardship to a plaintiff
particularly in so far as the running of prescription is concerned. However, it is unnecessary to
finally decide the point. As indicated above, the allegations contained in the particulars of claim are
incapable of supporting evidence that would discharge the burden of proving on a balance of
probabilities that there will be a loss on the termination of the trust, nor could such allegations
reasonably have been made. Moreover, the argument advanced by counsel on both sides proceeded
on the premise that some form of past or accrued loss was an essential element of the appellant’s
cause of action.
The Court concluded23 that Jowell’s right to his share of the assets of the trust was postponed
until Mrs Jowell’s death and that Jowell had suffered ‘no past or accrued loss’. The action was
accordingly premature because the harm element had not been satisfied and, ‘whether or not he will
indeed suffer a loss will only be known on some future date’.
• The Court’s use of the phrase ‘damage or loss’ illustrates the terminological issue raised in
section 4.1 above. The element is ‘harm or loss’, not ‘damage or loss’.
• Harm as an element must be proven as a fact, on a balance of probabilities, otherwise a court
cannot entertain a delictual claim. Must the quantum of the claim also be proven as a fact?
How would a court assess quantum? Quantification of the harm is a separate issue that follows
after the harm element has been established. Is it clear from this extract when to make such
quantification? When would be the best time to do this?
• The law distinguishes between past or accrued loss and prospective loss, and recognises that a
person can claim for prospective harm. Why then, did the Court deny Jowell’s claim?
• What is the rationale for entertaining claims for prospective or future harm? Does this rationale
apply to all cases that involve prospective harm?
• The Court did not decide whether someone should be able to claim for future harm that could
be proven to arise on a balance of probabilities. The point was deliberately left open. If a court
is to decide this point in the future, which way would you advise the Court to decide?

PAUSE FOR Should the following constitute legally recognised harm?

• An heir’s expectation of an inheritance that did not materialise because the
REFLECTION
person who drew up the will did not have it properly attested. Has there been a
change in the beneficiary’s patrimony? Has the loss accrued (has the asset
vested in the beneficiary’s universitas? Is this not an instance of lucrum
cessans – where the deprived benefit would certainly have accrued but for the
failure to adhere to proper processes? Does the decision relate to the factual or
normative component of the harm concept?
• A shareholder owns 2,5% of a company’s shares but performs all the
company’s business functions, claiming all his own expenses from the
company bank account as if it were his own personal account. He is
permanently disabled in a motor collision, rendering him unable to perform his
duties, save his managerial functions. The company now employs another
person to do so. Can he claim past and future medical expenses? Past loss of
earnings? Loss of earnings capacity (in respect of future earnings)? Pain and
suffering? Would your conclusions be different were the shareholding to be
100% and not 2,5%? Do these decisions relate to the factual or normative
component of the harm concept?
• Police are called out to investigate a hoax call and incur unnecessary costs in
the form of fuel costs, motor vehicle wear-and-tear costs and forensic-
examination costs, and also pay wages in respect of the time wasted in
investigating the hoax. Does this relate to the factual or normative component
of the harm concept?

The definition of patrimonial harm also establishes what non-patrimonial harm is. Non-patrimonial
harm entails all forms of harm that one cannot measure in monetary terms and so cannot include in
a person’s universitas. Pain, inconvenience, shock and insulting behaviour are clearly not
patrimonial, for these do not reduce the monetary value of a person’s estate. However, expenses
incurred for medical treatment to diminish the pain, for example, or to pay for a psychiatrist’s
services, would constitute patrimonial loss.
Where the plaintiff has suffered patrimonial loss, the remedy lies in the lex Aquilia. The
Germanic remedy for pain and suffering and the actio iniuriarum provide remedies for non-
patrimonial harm.

4.3 Pain and suffering

Pain and suffering is a term of art. It consists of two concepts: pain and suffering in a more literal
sense, and loss of amenities of life. This category of harm must be linked to some physical injury to
the plaintiff. A plaintiff cannot claim pain and suffering for the pain experienced by seeing harm
caused to someone or something else, for example, seeing one’s child lying in a coma or one’s
beloved pet being put out of its misery. This harm and the remedial action associated with it, is
therefore to some extent a parasite concept. The harm must be associated with physical injury to the
plaintiff, and not someone else.
 The first component, pain and suffering, relates to pain that has been experienced and pain that
will continue to be experienced in the future. It has to be experienced subjectively. If a person has
not felt any pain, he or she has not suffered any harm, and also cannot claim future pain that might
be experienced in the absence of existing pain. However, pain and suffering is not limited to
physical pain alone. The concept includes nervous shock, psychological harm, mental anguish or
distress, and also the fear, anxiety or trauma associated with the injury or situation. These, too,
must be experienced subjectively.
The second component, loss of amenities of life, has been crisply defined as ‘a diminution in
the full pleasure of living’ 24 and loss of ‘enjoyment of life’. 25 In Administrator-General, South
West Africa v Kriel 26 Hoexter JA commented as follows:
The amenities of life may further be described, I consider, as those satisfactions in one’s
everyday existence which flow from the blessings of an unclouded mind, a healthy body, and
sound limbs. The amenities of life derive from such simple but vital functions and faculties
as the ability to walk and run; the ability to sit or stand unaided; the ability to read and
write unaided; the ability to bath, dress and feed oneself unaided; and the ability to exercise
control over one’s bladder and bowels. Upon all such powers individual human self-
sufficiency, happiness and dignity are undoubtedly highly dependent.

This concept also includes general inconvenience, discomfort, loss of life expectancy and the
humiliation that arises from any disability or disfigurement. Not all of these need necessarily be
experienced, and so the loss of amenities of life has both a subjective and an objective flavour.

Sigournay v Gilbanks27
The plaintiff, a keen sportsman, had sustained severe bodily injuries. He suffered brain
damage, lost his sight in one eye and fractured several bones. His brain injuries restricted his
movement. For some time after the accident, he had been in a coma.
In the course of assessing an appropriate amount of damages, Schreiner JA said:28

It seems to me that Henochsberg J failed to keep clear the distinction between

pain actually experienced, though possibly subsequently forgotten, and what
would have been pain but for anaesthesia. Whether that anaesthesia is induced
intentionally by drugs, as when an operation is performed, or is the chance
result of a head injury can make no difference.

The fact is that most of what might have been excruciating pain was not pain for
the plaintiff. And compensation under this heading is given for pain, not for the
seriousness of the injuries or the risk to the plaintiff’s life. Injuries may leave
after-effects and may cause mental anxiety but they are not themselves pain.
The learned Judge, in my view, misdirected himself on this part of the enquiry
with the result that he thought that the plaintiff, who certainly did suffer severely,
should be treated as having suffered more pain than he actually did.

These comments were made during the assessment of damages stage, where the Court was trying to
assess the extent of the pain, whether slight or extensive. To summarise:
• The important issue here is that the plaintiff had subjectively suffered some pain. The issue of
its severity, or its duration, is a question that is relevant to the assessment of the quantum of
damages.
• It does not matter if the plaintiff cannot later recall that he or she had been in pain, provided
that he or she had experienced it.29 However, loss of amenities of life does not need to be
experienced.30
• The passage distinguishes between pain and mental anxiety, which forms part of the second
aspect of pain and suffering, namely loss of amenities of life.
In this instance, the amenities of life that were diminished included:
• He ‘now had to content himself with watching instead of partaking in games’.
• While watching sport, he had on occasion broken down emotionally.
• There was a remote chance that he might suffer a more serious breakdown later, when he
realised fully the extent of his disability.
• His prospects of marriage had been lessened.
• He had lost sight in his right eye.
• He walked awkwardly, and for limited distances only.

Other forms of loss of amenities of life include lengthy periods of hospitalisation,31 being restricted
to a wheelchair, being deprived of ‘participating in sport, listening to music, learning a vocation,
getting married and having children’,32 and having to be tended to, or supervised, for the rest of
one’s life.33

PAUSE FOR Pain and suffering

1. A person is lying in a comatose state. Is she experiencing harm that could be
REFLECTION compensated for under the Germanic remedy? In considering the answer,
refer to Botha v Minister of Transport 1956 (4) SA 375 (W), Sigournay v
Gilbanks 1960 (2) SA 552 (A), Gerke NO v Parity Insurance Co Ltd 1966 (3)
SA 484 (W), Southern Insurance Association Ltd v Bailey NO 1984 (1) SA 98
(A), and Reyneke v Mutual and Federal Insurance Co Ltd 1991 (3) SA 412
(W).
2. Assume that the person referred to above is a 10-year-old girl. Can her father
claim damages for pain and suffering? What principles inform your decision?
3. When a person experiences shock, is the nature of the harm patrimonial, non-
patrimonial, or both? Would a person be able to claim damages under the lex
Aquilia as well as for pain and suffering?
4. Can a person who has suffered psychiatric injury claim pain and suffering as
well?

To obtain a remedy for pain and suffering, the elements of this Germanic remedy must be met.34

4.4 Infringement of a personality interest

Personality interests are traditionally associated with being human, and until the 1980s one could
safely say that artificial or juristic persons, such as companies and universities, had no rights of
personality. However, towards the end of the twentieth century, a trend emerged recognising that
juristic persons could also have personality rights, albeit in a less extensive way than natural
persons. When the Bill of Rights was adopted, this situation was confirmed. Artificial persons,
therefore, have (artificial) personality rights insofar as they are able to have such rights. This
simply means that juristic persons cannot have the more subjective personality rights that are
associated with a person’s physical existence, or with human feelings, but they can have the more
objective personality interests, such as privacy or reputation.
From the beginning, our common law recognised the three classic personality interests: corpus,
dignitas and fama (bodily integrity, dignity and reputation). Any factual violation of any of these
interests would constitute an iniuria, the harm element required under the actio iniuriarum.

TERMINOLOGY Iniuria
The term ‘iniuria’ has three possible meanings:
 • Anything contrary to justice and equity, that is, wrongfulness (dictionary
meaning); covering all delictual situations, irrespective of whether the
harm is patrimonial or non-patrimonial
• The wrongful and intentional impairment of a person’s personality right:
bodily integrity (corpus), dignity (dignitas) or reputation (fama); covering
only those situations that fall within the actio iniuriarum
• Impairment of dignity in the form of an insult where contumelia is
present; covering only those situations that fall within the narrow
meaning of the dignity concept.

4.4.1 Bodily integrity

Bodily integrity is normally violated by an assault or deprivation of liberty, such as arrest or
imprisonment, but seduction also falls into this category. Neethling, Potgieter and Visser note35 that
a right to corpus also includes psychological integrity and so the corpus violation occurs when a
person suffers a psychiatric injury or nervous shock. A violation of bodily integrity may, but not
necessarily, also be insulting and therefore also an infringement of dignity.

4.4.2 Dignity
Dignity is a complex concept. The first point to clarify is its scope. At one stage, and sometimes
still now, dignity was used both in a narrow sense, denoting self-esteem, and in an umbrella sense,
covering a variety of associated personality interests. To violate a plaintiff’s self-esteem (dignity in
the narrow sense), a defendant’s behaviour must have had some degrading or insulting effect –
what the Romans called contumelia – for otherwise the plaintiff’s feelings would not have been
wounded. The insult must be established as a fact, and it is sufficient to show that the plaintiff
subjectively felt insulted by the defendant’s behaviour. The test, therefore, is subjective and focuses
on what the plaintiff actually experienced. From this one can see that a juristic person, such as a
company, would not be capable of having this form of dignity. A juristic person cannot have
personal feelings to offend and so cannot suffer contumelia.

PAUSE FOR Dignity

REFLECTION Consider the following questions:
• Is dignity a purely subjective phenomenon?
• Is it entirely concerned with what one thinks of oneself?
• Is there not also an outward component to the concept?
For example, one can describe a person as a dignified person. The aura
that this person has is not reputation, but an outward manifestation of
inner dignity. When this aspect is violated, it surely does not require a
subjective insult in the form of contumelia, and it also does not require
that there be publication to anyone. However, if the plaintiff is not
offended in any way, how would one assess whether there had been a
violation of dignity?
Consider also what Melius de Villiers said in his book, Injuries:36

It is clear from the Digest that the word dignitas must be

understood in a wide sense, and not as merely equivalent
to the elevated public position of the Roman citizen.
Injuries against dignity evidently comprise all those injuries
which are not aggressions upon either the person or the
reputation; in fact, all such indignities are violations of the
 respect due to a free man, as such (vide Huschke, Gaius p.
152).

Originally, the law classified personality rights as corpus, dignitas and fama. Given this constraint,
personality interests like privacy and identity were for a long time grouped together under the
dignity umbrella and were not recognised as distinct personality interests. Of course, this line of
thinking accorded with the common-law position at the time.37 However, research has since shown
that privacy and identity should be treated as separate, self-standing personality interests, and that
they should not be subsumed under the dignity umbrella. Courts have responded accordingly,
recognising privacy and identity as separate, self-standing personality rights. This is a step forward.
Not only does it indicate the dynamic nature of delict and its ability to respond to changes in
thinking and needs in society, but it also frees the law of trying to apply rules that have evolved in
respect of one set of interests to other interests that are not entirely comparable. For insult, for
example, there must be some degrading or insulting behaviour (contumelia), but this aspect is not a
necessary element when a person’s privacy or identity is violated. The separate recognition of these
interests should rid the law of convoluted thinking.

4.4.3 Privacy
The development of the law of privacy illustrates the tension that arises between a system that
seeks to develop within the constraints set by its roots and precedent, and one that wishes to
develop outside those historical parameters. The right to privacy was not specifically recognised in
the actio iniuriarum, and so any attempt to protect a person’s privacy had to be located within the
dignity concept. Even towards the end of the twentieth century one finds, for example, the
Appellate Division declaring: ‘The actio iniuriarum protects a person’s dignitas and dignitas
embraces privacy.’38
What is ‘privacy’? Famously, Warren and Brandeis said in 1890 that it is the right ‘to be let
alone’.39 Privacy protects a person’s interest to enjoy personal space as well as peace and tranquility
away from the public and the glare of publicity. Every individual has the right to decide what to
keep private and the extent to which to interact with others. Hence, the personal sphere is
determined subjectively. Whether that determination is a reasonable one, especially in communal
interactions, is a question more appropriate to the wrongfulness element.
The factual disturbance of privacy occurs when personal space and peaceful existence are
violated, usually in one of two ways: intrusion or disclosure. However, the right is not infringed
until a court determines that the violation occurred in a legally unacceptable way. Van der Walt and
Midgley40 list the following examples of factual violations of privacy:
Examples of intrusion are searches and seizures by police and other authorities, entry into
private premises, the reading of private documents, listening in to private conversations,
monitoring and interception of electronic mail, including bombardment of unwanted
material, exposure to unwanted broadcasts (acoustic privacy), peeping at someone,
shadowing a person, uninvited obtaining of information from a person, for example, filming
a person’s activities or, obtaining blood samples from a person. Disclosures can take the
form of a disclosure of private facts obtained by means of intrusion, disclosure of private
facts generally, disclosure of confidential facts, unwanted publication of photographs, and
the unauthorised extended publication of private information in circumstances where the
privacy expectation boundaries are circumscribed.
 PAUSE FOR Privacy
1. Is there a difference between personal facts and private facts? Does the law of
REFLECTION
privacy cover both categories?
2. ‘As a corporate entity, Thint does not bear human dignity and thus its rights of
privacy are much attenuated compared with those of human beings’ (Thint
Holdings (Southern Africa) (Pty) Ltd v National Director of Public Prosecutions;
Zuma v National Director of Public Prosecutions).41 To what extent would a
juristic person have a right to privacy? Can a person intrude on a company’s
personal space? Can a person disclose private information belonging to a
company, or a university, for example? (In considering these questions, also
refer to section 8 of the Constitution, Financial Mail (Pty) Ltd v Sage Holdings
Ltd,42 Investigating Directorate: Serious Economic Offences v Hyundai Motor
Distributers (Pty) Ltd: In re Hyundai Motor Distributors (Pty) Ltd v Smit NO 43
and South African Broadcasting Corporation v Avusa Ltd.)44
3. Does an organ of state have privacy?
4. Van der Walt and Midgley45 contend that, with electronic communications, one
should distinguish monitoring and disclosing information about emails from
monitoring and disclosing content of emails; with the former being acceptable
and the latter usually not. Would you agree? In which instances, if any, would
monitoring and disclosing of content be acceptable? (In considering these
questions, also refer to section 8 of the Constitution, the Electronic
Communications and Transactions Act 25 of 2002; the Regulation of
Interception of Communications and Provision of Communication-related
Information Act 70 of 2002 and Klein v Attorney-General, WLD,46 Protea
Technology Ltd v Wainer )47
5. Do the decisions in 1–4 above relate to the factual or normative component of
the harm concept?

O’Keefe v Argus Printing and Publishing Co Ltd48

The plaintiff had agreed to her photograph being published as part of a news article.
However, the photograph was used in an advertisement for a gun manufacturer. The plaintiff
claimed that this violated her personality interests, and in particular, violated her dignity. The
Court phrased the issue as follows:

Can it reasonably be said (in the light of modern conditions)49 that a person has
been subjected to offensive, degrading and humiliating treatment if his
photograph and name have, without his consent, been used in the press for
advertising purposes?50

After noting that insult was not necessarily ‘the essence of an iniuria’, the Court then
answered the question in the affirmative, holding that:

[T]o use a person’s photograph and name, without his consent, for advertising
purposes may reasonably constitute offensive conduct on the part of the user.51

• When referring to ‘the essence of an iniuria’, was the Court using the term in the wider or
narrower sense?
• What personality rights were involved in this instance?
• Compare this case with Grütter’s case, to which we refer in the next section. If O’Keefe were
to be decided today, would the Court’s reasoning be different? Why?
• Is misappropriation of image for advertising purposes a violation of identity, privacy or
dignity?

4.4.4 Identity
Neethling, Potgieter and Visser 52 explain identity as follows:
 Identity is that uniqueness which identifies each person as a particular individual and as
such distinguishes him from others. Identity manifests itself in various indicia by which the
person involved can be recognised: ie, facets of his personality which are distinctive of or
peculiar to him, eg his life history, his character, his name, his creditworthiness, his voice,
his handwriting, his outward shape, etc. Identity is thus infringed if indicia thereof are used
in a way that does not reflect the person’s true (own) personality image.

The right to identity, therefore, is a right to a person’s image and aspects associated with it. The
integrity of a person’s image is similar to a person’s bodily integrity and a factual violation of
identity occurs when aspects associated with a person’s particular image are used outside the sphere
or scope of that image. Therefore, the violation occurs as a fact, and the necessary harm is suffered.
However, one still needs to determine whether the violation occurred in a legally acceptable or
legally unacceptable way. Only then can one say that a person’s personality interest has been
infringed. Points to note are:
• A factual violation of a person’s identity constitutes sufficient harm for there to be a violation
of a personality interest.
• As with cases that involve bodily integrity, the violation of identity may, but not necessarily,
be accompanied by contumelious conduct. In the latter instance, there would be an additional
infringement of dignity.
• The factual violation of identity can be assessed objectively.

Grütter v Lombard53
Grütter and Lombard had been partners in an attorneys firm called ‘Grütter and Lombard’.
When their partnership terminated, Grütter began practising with another attorney as ‘Grütter
and Grobbelaar’, while Lombard continued to practise under the old partnership name.
Grütter asked Lombard not to continue using his name, but Lombard refused to do so. Grütter
then sought a court order, which the Supreme Court of Appeal granted. In doing so, the Court
recognised identity as part of a variety of personal rights that ‘are included in the concept of
dignitas in the context of the actio injuriarum’ and concluded:54

The interest that a person has in preserving his or her identity against
unauthorised exploitation seems to me to be qualitatively indistinguishable and
equally encompassed by that protectable ‘variety of personal rights’.55

• When referring to ‘the concept of dignitas’, was the Court using the term in the wider or
narrower sense?
• What personality rights were involved in this instance?
• Compare this case with O’Keefe’s case. Has Grütter changed the common law in any way?
• The Court talked about ‘dignitas in the context of the actio injuriarum’. Are there any other
contexts in which we could use this concept? Would the concept then be used in a different
sense?
• In the previous text, we said that courts are now recognising identity as a self-standing,
separate right. Would you agree with this statement?

In summary, irrespective of the way in which one views the concept, dignitas is more than the
narrow iniuria, so one can sometimes assess it purely subjectively, and at other times less so, and
even objectively. While juristic persons would not be able to claim for injured feelings, or for
wrongful arrest, their identity and privacy interests are surely capable of protection, for these
interests are not closely associated with the physical and emotional aspects of being human.

4.4.5 Reputation
The last right that falls within the original trilogy of personality rights, reputation, is violated when
one person lowers the public esteem in which another is held. For this interest to be violated, there
must have been some publication of the defamatory material to someone other than the plaintiff and
the defendant. As a result of what is said or done, a person, or people, would now think less of the
plaintiff than previously. Naturally, for a person’s public esteem to be violated, others must become
aware of what was said or done, so publication of defamatory material is the core aspect of any
violation of reputation. The effect of such publication must be that it lowers a person’s esteem in
the estimation of right-thinking persons generally.56 Examples include:
• Imputations against moral character
• Imputations that arouse hatred, contempt and ridicule
• Impairments that cause shunning and avoiding
• Impairments of professional or business reputation.57
Other common violations of reputation include malicious proceedings, such as malicious
prosecution and malicious attachment of goods.

PAUSE FOR Protecting rights

REFLECTION What rights do wrongful arrest, malicious arrest, malicious prosecution
and malicious attachment of goods protect? Is the classification of these
rights logical?

PAUSE FOR The violation of personality interest

REFLECTION One assesses the violation of a personality interest subjectively, or from
the plaintiff’s point of view. It is sufficient merely to show that the violation
occurred, for example, that the plaintiff was assaulted or arrested, that
the plaintiff felt insulted, or that his or her reputation was lowered. Once
established, this satisfies the harm element for the actio iniuriarum. One
does not assess any other aspect, such as whether the harm was minor
or substantial, or whether the plaintiff was over-sensitive in the
circumstances, under this heading. These issues become relevant when
one assesses the wrongfulness element.
In Delange v Costa 58 Delange felt deeply hurt by remarks that Costa
had made in a letter to him. (Note that because the remarks were not
published to a third party, there was no infringement of reputation.) The
Court noted that Delange needed to show that his dignity was impaired
by these remarks:

This involves a consideration of whether the plaintiff’s

subjective feelings have been violated, for the very essence
of an injuria is that the aggrieved person’s dignity must
actually have been impaired. It is not sufficient to show that
the wrongful act was such that it would have impaired the
dignity of a person of ordinary sensitivities.

(Note that the latter criterion, ‘whether the conduct would have impaired
the dignity of a person of ordinary sensibilities’, is an objective
wrongfulness criterion, which must be established in addition to the harm
criterion. So, the harm element is assessed subjectively and the
wrongfulness element objectively.)
 PAUSE FOR Digital manipulation
REFLECTION In Le Roux v Dey (Freedom of Expression Institute and Restorative
Justice Centre as Amici Curiae)59 two schoolchildren created a computer
image in which the faces of their school’s principal and deputy principal
(Dey) were superimposed onto an image of two naked men sitting in a
sexually suggestive posture. In Cele v Avusa Media Ltd60 a newspaper
published two articles and a digitally altered photograph of Cele, who at
the time was the Member of the Executive Council in KwaZulu-Natal
responsible for Transport, Community Safety and Liaison. In the
photograph Cele’s head was superimposed onto an image of a person
dressed as a gun-carrying sheriff from the Wild West. The caption read:
‘Police must aim for the head.’
Both Dey and Cele sued for defamation and infringement of dignity.
Are infringements of dignity and reputation the only possible forms of
harm arising from the manipulation of the images?

The remedy where harm takes the form of a personality interest is the actio iniuriarum.

4.5 Personality rights and constitutional rights

The second point that requires clarification (the first point was the scope of the dignity concept)
relates to the relationship between the common-law concept of dignity and dignity as set out in the
Constitution. Note that the Constitution recognises the rights to dignity and privacy, but not the
right to reputation. Also, the right to identity is not pertinently recognised either separately in the
Bill of Rights or as part of the right to dignity. So, at face value, the common-law concepts do not
correspond with their constitutional equivalents and some reconciliation needs to be made.
The obvious question to ask is whether the Constitution recognises reputation as a human
right, even though it is not mentioned in the Bill of Rights. The answer is that it does, but it sees
reputation as part of a person’s human right to dignity. Unlike the common law, the Constitution
does not distinguish between a person’s personal feelings and self-esteem, and a person’s esteem in
the eyes of others. The constitutional right to dignity is seen to cover both situations.
Does this then mean that there is a conflict between the Constitution and the common law?
Courts have consistently held no.61 Although many common-law rights and interests easily translate
into constitutional rights,62 the common law does not have to mirror or replicate concepts found in
the Constitution. All that is required is that the common law should not be incompatible with the
Constitution. The common-law concept of fama, or reputation, does not violate the constitutional
right and value of dignity.63 In fact, it gives effect to and protects human dignity. Courts, therefore,
say that there is no problem with the common-law approach.

PAUSE FOR Khumalo v Holomisa64

REFLECTION In Khumalo v Holomisa65 O’Regan J commented on the interplay between
the delictual and constitutional rights:66

In the context of the actio injuriarum, our common law has

separated the causes of action for claims for injuries to
reputation (fama) and dignitas. Dignitas concerns the
individual’s own sense of self-worth, but included in the
concept are a variety of personal rights including, for
 example, privacy. In our new constitutional order, no sharp
line can be drawn between these injuries to personality
rights. The value of human dignity in our Constitution is not
only concerned with an individual’s sense of self-worth, but
constitutes an affirmation of the worth of human beings in
our society. It includes the intrinsic worth of human beings
shared by all people as well as the individual reputation of
each person built upon his or her own individual
achievements. The value of human dignity in our
Constitution therefore values both the personal sense of
self-worth as well as the public’s estimation of the worth or
value of an individual. It should also be noted that there is a
close link between human dignity and privacy in our
constitutional order. The right to privacy, entrenched in
section 14 of the Constitution, recognises that human
beings have a right to a sphere of intimacy and autonomy
that should be protected from invasion. This right serves to
foster human dignity. No sharp lines then can be drawn
between reputation, dignitas and privacy in giving effect to
the value of human dignity in our Constitution. No
argument was addressed to this court on the relevance of
the right to privacy to this case and I shall not consider it
further.

Note that the justice is referring to dignity as a value, not dignity as a

right. Does that make a difference?
• What are the similarities and differences between the constitutional rights to
bodily integrity, dignity, privacy and reputation, and those found in the law of
delict?
• What effect is the Constitution likely to have on the way personality rights in the
law of delict will be defined in future?

4.6 Conclusion
The harm element is the cornerstone of delictual problem-solving. Once the nature of the harm is
identified, it is possible to identify the nature of the enquiry and elements that need to be proven.
However, like all other elements, the harm element serves not only to fix liability, but also to limit
it. This is where the normative aspects of the element come in, and they serve to restrict delictual
claims only to those where harm is actionable, provided that one remembers that no delictual
element exists in isolation. From this chapter, one can see that there is an interplay between the
harm and wrongfulness elements. There is a similar interaction between the way in which one
determines harm and assesses damages. So, for conceptual clarity, it is always important to
remember where one is along the problem-solving route towards the intended destination.
Figure 4.2 The harm element

1 First National Bank of South Africa Ltd v Duvenhage 2006 (5) SA 319 (SCA); [2006] 4 All SA 541 (SCA) para 1. See
also H v Fetal Assessment Centre 2015 (2) SA 193 (CC); 2015 (2) BCLR 127 (CC) paras 54 and 60.
2 Bennett v Minister of Police 1980 (3) SA 24 (C).
3 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A).
4 Barnard v Santam Bpk 1999 (1) SA 202 (SCA).
5 RAF v Sauls 2002 (2) SA 55 (SCA); Minister of Safety and Security v Sibili [2003] 4 All SA 451 (Tk).
6 [2007] 4 All SA 1152 (C); 2008 (1) SA 192 (C). See also section 4.3 below.
7 Para 22.
8 Para 23.
9 Union Government (Minister of Railways and Harbours) v Warneke 1911 AD 657; Jooste v Botha 2000 (2) SA 199
(T).
10 Edouard v Administrator, Natal 1989 (2) SA 368 (D) at 391; Kellerman v SA Transport Services 1993 (4) SA 872 (C)
at 876–877.
11 2010 (4) SA 558 (SCA) para 3 (footnotes omitted).
12 Similarly, the Constitutional Court has ruled that adultery can no longer be considered wrongful. In the past adultery
was viewed as being automatically insulting towards the innocent spouse (see, for example, Wiese v Moolman 2009
(3) SA 122 (T)), but in DE v RH 2015 (5) SA 83 (CC) the Court noted that public attitude towards adultery has
softened (para 52) and that the potential infringement of dignity had to be balanced against the rights of the adulterous
spouse and the third party to privacy, freedom of association and security of person (para 62).
13 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 222.
14 1911 AD 657.
15 At 662.
16 At 669.
17 At 669.
18 1997 (1) SA 611 (A).
19 2000 (3) SA 274 (SCA).
20 Jowell v Bramwell-Jones; 2000 (3) SA 274 (SCA); [2000] 2 All SA 161 (A) para 22.
21 The element is ‘harm or loss’, not ‘damage or loss’. See section 4.1 above.
22 See also Corbett The Quantum of Damages Vol 1, 4 ed (Gauntlett) at 9 where the same criticism is made.
23 Paras 24 and 25.
24 H West & Son Ltd v Shephard [1963] 2 All ER 625 (HL) at 636G–H.
25 Sigournay v Gilbanks 1960 (2) SA 552 (A) at 572B.
26 1998 (3) SA 275 (A) at 288E–F.
27 1960 (2) SA 552 (A).
28 At 571D–G.
29 Sigournay v Gilbanks 1960 (2) SA 552 (A) at 571B–C; Botha v Minister of Transport 1956 (4) SA 375 (W); Gerke
NO v Parity Insurance Co Ltd 1966 (3) SA 484 (W).
30 Reyneke v Mutual and Federal Insurance Co Ltd 1991 (3) SA 412 (W) at 426B.
31 Sandler v Wholesale Coal Suppliers Ltd 1941 AD 194 at 199.
32 Reyneke v Mutual and Federal Insurance Co Ltd 1991 (3) SA 412 (W) at 429C–D.
33 Southern Insurance Association Ltd v Bailey NO 1984 (1) SA 98 (A) at 120A.
34 See Chapter 25 below.
35 Neethling and Potgieter (2015) at 345 and 346.
36 De Villiers The Roman and Roman-Dutch Law of Injuries: A Translation of Book 47, Title 10 of Voet’s Commentary
on the Pandects (1899) 24 fn 19.
37 O’Keefe v Argus Printing and Publishing Co Ltd 1954 (3) SA 244 (C).
38 Jansen van Vuuren v Kruger 1993 (4) SA 842 (A) at 849 E–F.
39 Warren and Brandeis ‘The Right to Privacy’ (1890–1891) 4(5) Harvard Law Review 193 at 195.
40 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 102 (footnotes omitted).
41 2009 (1) SA 141 (CC) para 76.
42 1993 (2) SA 451 (A).
43 2001 (1) SA 545 (CC).
44 2010 (1) SA 280 (GSJ) at 295.
45 Van der Walt and Midgley (2016) para 102.
46 1995 (3) SA 848 (W); [1995] 4 All SA 205 (W).
47 [1997] 3 All SA 594 (W); 1997 (9) BCLR 1225 (W).
48 1954 (3) SA 244 (C).
49 The Court added this aspect elsewhere in the judgment at 249A.
50 At 248A–B.
51 At 249A.
52 Neethling and Potgieter (2015) at 373–374 (footnotes omitted).
53 Grütter v Lombard 2007 (4) SA 89 (SCA).
54 Para 12.
55 A phrase used by O’Regan J in Khumalo v Holomisa 2002 (5) SA 401 (CC) para 27.
56 Sindani v Van der Merwe 2002 (2) SA 32 (SCA) para 15.
57 Burchell The Law of Defamation in South Africa (1985) at 103–126.
58 1989 (2) SA 857 (A) at 861D–E.
59 2011 (6) BCLR 577 (CC); 2011 (3) SA 274 (CC).
60 [2013] 2 All SA 412 (GSJ).
61 Khumalo v Holomisa 2002 (5) SA 401 (CC) para 28; The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others,
Amici Curiae) 2011 (4) SA 191 (CC); 2011 (8) BCLR 816 (CC) para 79.
62 H v Fetal Assessment Centre 2015 (2) SA 193 (CC); 2015 (2) BCLR 127 (CC) paras 78–79.
63 The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae) 2011 (4) SA 191 (CC); 2011 (8) BCLR
816 (CC) para 79.
64 2002 (5) SA 401 (CC).
65 2002 (5) SA 401 (CC).
66 Para 27.
 Chapter 5

Conduct

5.1 Introduction

5.2 Human conduct

5.3 Voluntary conduct and the defence of automatism

5.4 Commission and omission

5.5 Animal behaviour

|

5.1 Introduction
In the same way that there cannot be a delict without harm being present, there cannot be a delict
unless one can link the harm suffered to the conduct of a person, or the behaviour of an animal. It is
this conduct or behaviour that sets the delictual sequence of events and consequences in motion. As
a general rule, delictual liability is based on voluntary human conduct, and this is certainly true for
any liability based upon fault and some instances of strict liability. However, in a diagram in
Chapter 1,1 we suggested that damages claims based on animal behaviour should also fall within
the concept of delict, but as we describe below, liability in such instances rests on a different
principle.
The conduct element is another one of the factual elements referred to in Chapter 1 which a
plaintiff needs to prove by producing evidence. The element requires the plaintiff to show that there
was some overt behaviour linked to the harm. Mere thoughts, without being manifested in some
way, cannot create delictual consequences. The overt behaviour that could constitute and satisfy the
conduct element includes:
• A positive physical act, such as driving your vehicle into a neighbour’s wall; or manufacturing
a product; or enticing a rival’s staff member to join your business (a commission)
• A positive statement or comment, in writing or orally (which is also a commission)
• A failure to do or say something, such as merely looking on while a person is drowning
without attempting a rescue; failing to repair potholes in a road or to institute safety measures
where these are required; or failing to warn someone of the harmful effects of, for example, a
tablet (an omission).
The distinctions between positive conduct and omissions, and between positive conduct and
statements, are important because the nature of the conduct is a crucial consideration when courts
determine whether they should consider the causing of harmful consequences to be wrongful, and
so to attract liability. For reasons of policy, courts are more readily inclined to hold defendants
liable in respect of consequences flowing from positive acts, and less so in cases of omissions and
statements. For example, while it is usually socially and economically unacceptable to cause others
harm, it would be unduly restrictive, and therefore undesirable, to enforce a wide and general duty
to prevent harm to others. It follows that there is no general duty to prevent harm to others. In law,
you are generally not compelled to be your brother’s keeper. The same restrictive approach applies
to conduct in an oral or written statement. This is because the harmful effects of the spoken or
written word may be widespread and indeterminate. Therefore, courts impose liability for the
harmful effects of an omission or statement only in special circumstances that indicate a duty to
prevent harm. The existence of such a legal duty, in respect of positive acts, omissions and
statements, is essentially a question of wrongfulness.

5.2 Human conduct

The conduct that forms the subject of a delictual enquiry is, as a rule, that of a human being, and
usually, defendants are natural persons. However, juristic persons (for example, a company or a
university) may also be sued in delict. In such instances, it is the conduct of the juristic person’s
office bearers or organs that is attributed to that entity. In other words, a juristic person acts through
its organs, and so, for example, a decision of a board of directors becomes a decision of the
company.2
Human conduct will also be present where objects or animals are used as instruments to cause
harm, for example, where a person uses a knife to stab someone, a motor vehicle strikes a
pedestrian, a tape recording contains a defamatory statement, or where a defendant incited a dog to
bite the plaintiff.3

5.3 Voluntary conduct and the defence of automatism

The conduct must be voluntary, that is, subject to the actor’s will and control. This implies that the
person should have the ability to direct muscular activity or to prevent such activity. In most
definitions of voluntary conduct, the particular act or omission is referred to as being ‘subject’ or
‘susceptible’ to the control of the actor’s conscious will. Essentially, the voluntary nature of a
defendant’s conduct refers to the extent to which the defendant can make a decision to act or to
refrain from acting.
The term voluntary, as it is used in this context, has a special and restricted meaning to
determine the ability in law to act. It does not necessarily correspond to a recognised psychological
condition and the term does not refer to what the defendant intended to do or should reasonably
have done. These are questions that concern intent or negligence, which we deal with under the
enquiry into fault. Even an infant or a mentally disturbed person can act voluntarily, based on the
mental capacity to direct muscular activity. However, such people lack a different kind of
‘voluntariness’ – accountability, or the capacity for fault – and therefore courts will not hold them
liable. Accountability requires a mental capacity more advanced than simply being able to direct
muscular activity, namely the ability to distinguish between right and wrong, and to act
accordingly.

PAUSE FOR A person’s capacity to act versus a person’s capacity to be at fault

REFLECTION There is a distinction between a person’s capacity to act and one’s
capacity to be at fault. In the first instance, the question is whether the
person is able to control his or her muscular movements towards a
particular end. If the person is unable to do so, then the conduct element
has not been satisfied. In the second instance, the question is whether
the person has the ability to behave responsibly, that is, whether he or
she is able to distinguish between right and wrong, and to act
accordingly. If the person cannot distinguish between right and wrong, or
does not have the mental ability or maturity to resist the impulse to
commit the wrong, then the person is not at fault. In other words, one
cannot be blamed for acting because one lacks accountability. In the first
instance, there is no conduct. In the second, there is conduct, but no
fault.

The defence of automatism occurs when a defendant asserts that he or she behaved involuntarily or
mechanically, or lacked the capacity to act voluntarily. Such incapacity is usually linked to a
known physical or mental condition that can render a person unable to direct muscular activity, but
involuntary conduct may result from other instances as well.
Conduct will not be voluntary in the following instances:
• Compulsion (vis absoluta or absolute force): Snyman4 cites the example of when someone is
slicing an orange with a knife and a stronger person grabs the hand holding the knife and
forces the knife into the chest of someone else. In such circumstances, we cannot say that the
person who was slicing the orange acted in stabbing the third person.
• Reflex muscular movements: A person who damages an article during a sneezing fit,5 or who
kicks over an expensive vase when tapped with a ruler on the knee, causing it to jerk upwards,
does not act voluntarily.
• Unconscious state: Included in this category is conduct occurring during sleep,6 an epileptic
fit,7 extreme intoxication8 or an extreme emotional state.9

The defence will not succeed in the following instances:

• Impulsive or spontaneous acts: These are not purely reflexive acts and are, therefore, regarded
as voluntary; for example, where a driver loses control of a vehicle when reacting to a bee
sting or to a burning match falling into his lap.10
• Intentional prior conduct: If the state of automatism is deliberately induced to cause harm to
another (actio in libera causa);11 for example, where a person deliberately gets drunk to work
up courage to assault another person,12 liability is founded on voluntary conduct that
intentionally induces a state of automatism. The defendant intentionally sets in motion a causal
chain of events that results in a harmful consequence.
• Negligent prior conduct: When a defendant voluntarily engages in a course of conduct that
leads up to a state of automatism, a court will impose liability if the defendant could
reasonably have foreseen the possibility of causing harm in the state of automatism. So a court
will impose liability where a person negligently allows a state of unconsciousness to develop,
for example, falling asleep while driving, or failing to prevent a reasonably foreseeable health-
related condition. In Wessels v Hall and Pickles (Coastal) (Pty) Ltd 13 the defendant, a diabetic,
suffered a hypoglycaemic attack and became comatose while driving a vehicle. This resulted
 in an accident. The defendant was aware of his diabetic condition and of the possibility of
sudden attacks and knew what precautionary measures (eating a mid-morning snack) were
necessary. He was held to have been negligent for having failed to take reasonable precautions
before driving his car.

The enquiry as to whether an act is voluntary is a subjective one – it concerns the defendant’s
capacity to direct muscular activity at the relevant time. On the other hand, the enquiry into
negligence is objective – it focuses on whether the defendant’s voluntary harm-causing conduct
conformed to the standard that can be expected of a reasonable person. The issue of voluntariness,
therefore, precedes and differs in principle from the issue of negligence. Where a defendant
voluntarily engages in a course of conduct that leads to a state of automatism, the enquiry into
negligence should focus on the course of conduct that preceded the state of automatism. Courts will
impose liability if the possibility of causing harm in the state of automatism was reasonably
foreseeable. For instance, a person who has forewarning of an impending heart attack will be
negligent if he or she nonetheless continues driving a vehicle.
When automatism is claimed, courts will scrutinise this defence carefully, because it is
difficult to verify the allegation of, for example, a sudden ‘black-out’. They will take into account
the circumstances at the time of the voluntary conduct preceding the state of automatism to
determine whether the state of unconsciousness and any possible harm to others were, at that time,
reasonably foreseeable and preventable. In doing so, courts do not merely rely on the defendant’s
evidence, but form an opinion using all the evidence presented to them and assessing the
probabilities of different scenarios arising from that evidence.

Molefe v Mahaeng14
The appellant (plaintiff) was involved in a motor collision with the respondent (defendant). The
accident occurred when two vehicles approached one another from opposite directions. The
defendant’s car swerved on to the incorrect side of the road, where it collided with the
plaintiff’s vehicle. The defendant gave evidence to the effect that he was overcome by a
sudden, unforeseen and uncontrollable black-out; that is, he raised the defence of
automatism. He attributed this to a fall in a supermarket earlier on the morning of the
accident. There was evidence that he had been treated at a hospital after the fall, and also
that he appeared to have been in a daze after the accident. When the matter came before the
Appellate Division, the Court held that defences based on automatism had to be scrutinised
with great care but, as is the norm, the plaintiff still had the onus to establish the defendant’s
conduct, which included proof of a voluntary act.
The Court held that the plaintiff had not discharged the onus of proving that the
defendant’s conduct had been voluntary and that the defendant did not suffer a black-out.
Despite shortcomings in the defendant’s testimony, there was sufficient evidence showing
that the defendant had had a fall shortly before the collision; that he had lost consciousness;
that he had received medical treatment. The black-out had come upon the defendant
unexpectedly and there was no evidence indicating that he had appreciated the possibility of
blacking-out; or that a reasonable person in his position would have been aware of such a
possibility. The plaintiff had, therefore, failed to prove, ‘as a matter of probability’ that the
defendant had been driving on the wrong side of the road ‘due to his voluntary act’.15

5.4 Commission and omission

As pointed out earlier, a positive act is called a commission, while a failure to act is called an
omission. It is not always easy to distinguish between commissions and omissions, especially
where the activity is continuous. So, the same conduct might at the same time constitute both a
positive act and an omission. The following examples indicate this overlap and the difficulty it
poses:
• Failing to stop at a stop sign and colliding with an oncoming vehicle constitutes positive
conduct (the act of driving), but some people might also think that it is an omission (failing to
stop). However, the failure to stop is usually treated as deficient (negligent) positive conduct.
In such an instance, the failure to stop is not an omission, but an act of negligent driving, in the
same way as driving without keeping a proper look-out (that is, failing to be observant) is not
an omission, but a way of saying that the person was driving negligently.
• Consider the case where a woman asks a policeman to take her home and the policeman
instead takes her to a secluded place and rapes her. The act of rape constitutes positive
conduct, but, simultaneously, the policeman’s failure to comply with his duty to protect
citizens amounts to an omission.16 In this instance, the rape is not a deficient way of executing
the policeman’s duty to protect the woman, but involves the breach of an entirely separate
obligation. The failure to protect the woman cannot be expressed as constituting deficient
(overlapping) positive conduct (the woman’s rape).

5.5 Animal behaviour

A delict can also be committed where animals cause harm without human conduct being involved.
For example, a pig owner may be liable if the animals stray onto neighbouring land and destroy
crops belonging to a neighbour (actio de pastu), or a dog owner may be liable if the dog bites a
passer-by (actio de pauperie). In these cases, the owner’s liability is based on a legal relationship
concerning the animal (ownership) and not on human conduct. Liability for the event that caused
harm is transferred, or imputed, to the owner. This situation is analogous to vicarious liability,
where liability for the delict of an employee is transferred, or imputed, to the employer.

Figure 5.1 Conduct in the law of delict

1 Figure 1.1.
2 A company or other legal entity may also be vicariously liable for the act of an employee or of a member of its senior
management. In such an instance, however, it is not the company that has acted, but its employee or manager, and
liability is not based on the company’s conduct.
3 Jooste NO v Minister of Police 1975 (1) SA 349 (E).
4 Snyman Criminal Law 6 ed (2014) at 54–55.
5 Snyman (2014) at 55.
6 R v Dlamini 1955 (1) SA 120 (T); R v Ngang 1960 (3) SA 363 (T).
7 R v Victor 1943 TPD 77; R v Mkize 1959 (2) SA 260 (N).
8 S v Chretien 1981 (1) SA 1097 (A) at 1104.
9 S v Arnold 1985 (3) SA 256 (C).
10 S v Smit 1963 (4) SA 824 (GW); S v Crockart 1971 (2) SA 496 (RA); S v Erwin 1974 (3) SA 438 (C).
11 The rule in full reads ‘actio non in se, sed tamen in sua causa libera’ meaning literally ‘an act that was not in his
power (at the time) but its (original) cause was in his power’.
12 S v Baartman 1983 (4) SA 395 (NC).
13 1985 (4) SA 153 (C).
14 1999 (1) SA 562 (SCA).
15 At 569F–G.
16 K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC).
 Chapter 6

Factual causation

6.1 Introduction

6.2 Determining factual causation and the conditio sine qua non test

6.3 Critique of the conditio sine qua non test

6.3.1 The elimination/substitution process of reasoning is clumsy and circuitous
6.3.2 The conditio sine qua non test provides no answer in cases of multiple (or
cumulative) causes
6.3.3 The conditio sine qua non test is not a true test for determining factual
causation

6.4 Alternatives to the conditio sine qua non approach

6.4.1 Material contribution
6.4.2 Common sense
6.4.3 Human experience and knowledge
6.4.4 Increasing risk and creating opportunities for occurrence of harm

6.5 Alternative tests for multiple, cumulative and successive causes

6.6 The appropriate test for factual causation

6.7 Conclusion

6.1 Introduction
To find a defendant delictually liable, there has to be a causal connection between the harm that the
plaintiff suffered and the defendant’s conduct. In other words, the defendant’s conduct must have
caused the plaintiff’s harm or loss. Without a causal connection between the harm and the
defendant’s conduct, there can be no delict.1
 TERMINOLOGY Cause
In Minister of Police v Skosana,2 Corbett JA cited Prosser’s
definition of ‘cause’:3

A cause is a necessary antecedent: in a very real and

practical sense, the term embraces all things which have so
far contributed to the result that without them it would not
have occurred. It covers not only positive acts and active
physical forces, but also pre-existing passive conditions
which have played a material part in bringing about the
event. In particular, it covers the defendant’s omissions as
well as his acts.

A conclusion as to whether a causal link exists between the wrongdoer’s conduct and the harm is
drawn from the facts, the evidence before the court, and the relevant probabilities in the
circumstances. The important question for purposes of this chapter is the approach to and the
manner in which one should draw this conclusion. In other words, what is the test for factual
causation? This enquiry consists of two parts, which was expressed in Minister of Police v Skosana
in the following terms:4
Causation in the law of delict gives rise to two rather distinct problems. The first is a factual
one and relates to the question as to whether the negligent act or omission in question caused
or materially contributed to … the harm giving rise to the claim. If it did not, then no legal
liability can arise and cadit quaestio. If it did, then the second problem becomes relevant, viz
whether the negligent act or omission is linked to the harm sufficiently closely or directly for
legal liability to ensue or whether, as it is said, the harm is too remote. This is basically a
juridical problem in which considerations of legal policy may play a part.

This two-fold enquiry represents the two fundamental components of the element of causation:
factual causation and legal causation.5

PAUSE FOR The two-fold enquiry

1. Factual causation requires a factual enquiry. This enquiry entails the following:
REFLECTION
◆ Did the defendant’s wrongful conduct cause, or materially
contribute to, the harm sustained by the plaintiff (Minister of
Police v Skosana 6)?
◆ If the answer is no, the defendant is not liable and there is no
delict.
◆ If the answer is yes, move to the second question/component.
2. Legal causation entails a juridical enquiry where legal policy and other normative
issues play a role. The enquiry entails the following:
◆ Is the factual link strong enough?
◆ Is the harm sufficiently closely connected to the conduct?
◆ Should the law confirm that the defendant caused the harm, or
should liability be limited?

In Chapter 1 we stated that one can divide the elements of a delict broadly into factual issues and
normative issues. The causation element has components that fit into both these categories. The
first component, factual causation, comprises primarily a factual enquiry. The second component,
legal causation, falls in the category of normative issues. This chapter focuses only on factual
causation. The enquiry into legal causation requires value judgements and policy considerations,
which we discuss in the next chapter.

6.2 Determining factual causation and the conditio sine qua

non test
In South African courts, the main tool for determining factual causation is the conditio sine qua non
theory (or ‘but-for’ test).7 According to this test, the defendant’s conduct must have been a
necessary condition (a conditio sine qua non) for the plaintiff’s harm to occur. The basis of the
conditio sine qua non test is that every event is the result of many conditions or factors that are
jointly sufficient to produce (or cause) that event. Therefore, the defendant’s conduct can have
caused the harmful consequence only if it was a necessary condition (conditio sine qua non) for
that consequence to occur. To determine this, the so-called ‘but-for’ test is used as a tool for
inferential reasoning: a defendant’s conduct (either an act or an omission) is a necessary condition
of the plaintiff’s loss if, but for the particular conduct, the harmful consequence would not have
occurred.8 The plaintiff bears the onus of proving on a preponderance of probabilities that the
defendant’s conduct in fact caused the harm.9

TERMINOLOGY The ‘but-for’ test

In International Shipping Co (Pty) Ltd v Bentley 10 the Appellate
Division articulated the test for factual causation as follows:

The enquiry as to factual causation is generally conducted

by applying the so-called ‘but-for’ test, which is designed to
determine whether a postulated cause can be identified as
a causa sine qua non of the loss in question. In order to
apply this test one must make a hypothetical enquiry as to
what probably would have happened but for the wrongful
conduct of the defendant. This enquiry may involve the
mental elimination of the wrongful conduct and the
substitution of a hypothetical course of lawful conduct and
the posing of the question as to whether upon such a
hypothesis the plaintiff’s loss would have ensued or not. If
it would in any event have ensued, then the wrongful
conduct was not a cause of the plaintiff’s loss; aliter, if it
would not so have ensued. If the wrongful act is shown in
this way not to be a causa sine qua non of the loss
suffered, then no legal liability can arise.

Applying the ‘but-for’ test requires a particular process of reasoning, which differs according to
whether the conduct is a positive act or an omission.
• For positive conduct, one applies a process of mental elimination. In this process, one mentally
eliminates the defendant’s act from the conditions that lead to the harm to determine whether
the outcome would have been the same. If the harm would not have occurred, the defendant’s
conduct was in fact a necessary condition (conditio sine qua non) of the harm.11
• For omissions, one applies a process of hypothetical substitution. In this process, one ‘thinks
in’ a (hypothetical) lawful course of conduct as a substitute for the omission. If the
hypothetical act probably would have prevented the loss from occurring, the wrongful
omission was a necessary condition (conditio sine qua non) of the loss occurring and,
therefore, was the cause of the loss. Hypothetical substitution involves a retrospective analysis
of what probably would have occurred, based on the evidence, and what could be expected to
occur in the ordinary course of human endeavour.12
S v Van As13
The appellants were police officers who had arrested a man called Makwena. While the
officers were arresting Makwena, five young children who were in his company disappeared.
They were scantily dressed and of poor physique. The following morning, two of the children
were found dead from exposure to cold and rain. The police officers who had failed to search
for the children were accused of culpable homicide. The State contended that the
circumstances of Makwena’s arrest had placed the appellant policemen in a relationship with
the children that placed a legal duty on them to take reasonable precautions to ensure the
children’s safety, which they had failed to do.
The question before the Court was whether there was a causal relationship between the
officers’ failure to look for the children and the death of two of the children.
The Court concluded that the State had failed to prove a causal link between the officers’
omission and the children’s death. In considering whether the children would probably have
been found and saved from death if the police had searched for them, the Court described
the (hypothetical) course of action that a reasonable person in the position of the police
officers would have taken. The Court decided that, on the evidence before the Court, a
proper search would not have saved the children from death. As a result, there was no
factual link between the omission and the death of the children.

Minister of Police v Skosana14

Whilst under the influence of alcohol, the respondent’s husband, Timothy Skosana, was
involved in a car accident and sustained injuries. He was taken to the police station and, after
being processed, taken to the district surgeon who performed a clinical examination on him.
At that point, the district surgeon did not observe any serious injuries. When Timothy’s cell
door was opened the next morning at 07:45, he complained of severe abdominal pain. Two
hours later, a constable walked with him to the rooms of the district surgeon, who examined
him and diagnosed him with an ‘acute abdomen’. The doctor wrote a note for the hospital and
instructed the constable to arrange to take Timothy there. After a two-hour delay in
summoning an ambulance, Timothy arrived at the hospital at 12:30. The doctors there found
that his condition was serious. He died shortly after an operation that had started at 16:25.
Timothy’s widow and children claimed damages from the Minister of Police based on the
officers’ negligent failure to provide him with prompt medical help. Corbett JA expressed the
applicable test as follows:15

The negligent delay in furnishing the deceased with medical aid and treatment,
for which Davel and Mahela were responsible, can only be regarded as having
caused or materially contributed to his death if the deceased would have
survived but for the delay. This is the crucial question and it necessarily
involves a hypothetical enquiry into what would have happened had the delay
not occurred.

Using a hypothetical chain of events based on prompt and efficient conduct by the police, the
Court decided, by a majority, that the probable outcome would be that Timothy Skosana
would not have died. The police officers’ omission was therefore a necessary condition – and
thus a factual cause – of Timothy’s death.
 mCubed International (Pty) Ltd v Singer NNO16
The respondents were trustees of the Leon John Singer Family Trust who wanted to invest
R40m overseas through various investment companies. They wished to invest R10m of the
R40m through the appellants (mCubed) by converting the R10m into US dollars. Employees
of mCubed proposed an investment structure and the funds were invested overseas.
However, the investment structure proposed by mCubed later proved to be unlawful and
when the trustees learned about this, they terminated the investment through the early
withdrawal of the investment funds. However, they could not recover all the money invested
as there had been an unexpected strengthening of the rand against the dollar. The trustees
then instituted legal proceedings against mCubed to recover the balance of the invested
funds.
The trustees alleged that, but for the misrepresentations made by mCubed, they would not
have invested the money through mCubed and they would have retained the money in rand.
They also argued that, but for the misrepresentations made by mCubed, they would have
terminated the investment immediately and would not have suffered the loss.
The Court confirmed the ‘but-for’ test as being the appropriate test to determine factual
causation, and in applying this test, it asked what mCubed should have done in the
circumstances. The Court was satisfied that, but for the misrepresentation by mCubed, the
R10m would not have been invested with mCubed at that stage.
The trustees contended that this was the end of the ‘but-for’ enquiry and that factual
causation had thus been established. The Court disagreed, having taken the following two
aspects into consideration: (i) the unpredicted strengthening of the rand against the US
dollar, and (ii) that the evidence indicated that, had the trustees known about the proposed
investment structure’s illegality, the R10m would likely have been converted to US dollars
and invested through some other investment company.
The Court reasoned that, but for the unexpected strengthening of the rand against the
dollar, the loss would not have been sustained. If the rand–dollar exchange rate performed
as predicted, the trust would have made a profit despite the trustees’ dissatisfaction with
mCubed and despite the early withdrawal of investment funds. The Court furthermore took
into account the fact that even if mCubed had acted as it should have and the trustees
accordingly did not invest the R10m with it, there was enough evidence to indicate that the
trustees would have invested the R10m, converted to US dollars, through other investment
companies where they could have sustained a similar loss due to the stronger rand. Thus,
the misrepresentation was not the cause of the loss; it was merely the cause of the
investment through mCubed. Therefore, factual causation was not established because, but
for the misrepresentations made by employees of mCubed, the trust would probably still
have suffered the loss.
Note that causing an event that precipitates a loss (that is, the investment) does not
necessarily mean that the person also caused the harm (that is, the financial loss). The
corollary is also true: a person who has not caused the event (for example, an accident)
could be held to have caused the harm (for example, physical injury), if such harm is caused
by some other conduct (for example, the failure to wear a seat belt or a crash helmet).
Do you agree with the Court’s reasoning in this case? Should the Court not have
considered that, irrespective of where the money was to be invested, an early withdrawal
would have been unlikely, but for the misrepresentations, and the funds would probably have
been left untouched until the rand–dollar situation recovered? How much speculation can
one allow in applying the ‘but-for’ test?

When postulating a possible causal scenario, a court may look at the situation either subjectively or
objectively. A subjective approach would take into account what the specific person would actually
have done in the hypothetical situation. An objective approach would consider the hypothetical
sequence of events that would have arisen if the person were to have acted reasonably in the
circumstances. In the subjective scenario, the person might have made choices that a reasonable
person would not have made, thus leading to a different result. Our law favours the subjective
approach.17
Minister of Safety and Security v Carmichele18
Coetzee had been charged with attempted rape and attempted murder, but was released on
warning. Subsequent to his release, Coetzee brutally attacked Carmichele and Carmichele
sued the Minister of Safety and Security and the Minister of Justice for damages. She
contended that members of the police, as well as the prosecutor in Coetzee’s case, had in
the circumstances owed her a legal duty to prevent her from being harmed and that they had
failed to comply with that duty. The basis for her contention was that the prosecutor had not
opposed Coetzee’s application for his release when both the investigating officer and the
prosecutor had relevant information that they should have placed before the Court.
Carmichele submitted that their failure to disclose that information led to her injuries. The
magistrate, who had the ultimate responsibility of deciding whether to keep Coetzee in
custody, testified that because of the emphasis on personal freedom and the lenient
approach to bail that prevailed at the time, he would still have released Coetzee even if the
information available to the investigating officer had been placed before him for consideration.
To determine factual causation, and because this case involved an omission, the Court
had to postulate what the facts would have been, had the police officers and the prosecutor
performed their duties properly. Would the magistrate have released Coetzee if the police
officers and the prosecutor provided the magistrate with the relevant information?
Various judgments were delivered in this matter. The judgment of the Cape High Court
was taken on appeal to the Supreme Court of Appeal, after which it came before the
Constitutional Court,19 where it was referred back to the Cape High Court. The second High
Court decision was again taken on appeal. The second Supreme Court of Appeal judgment,
which finally disposed of the matter, forms the basis of this discussion.

The Court referred to a moot point that the Constitutional Court had raised:20

An intriguing aspect raised by [the Constitutional Court],21 but left for later
decision, is whether an objective or subjective test should be applied in
determining causation. In the ordinary case the question does not arise, but in
this case, because one has to postulate a hypothetical judgment by a judicial
officer exercising a discretion, it does. An objective test would mean that the
Court has to determine what a reasonable magistrate, on the probabilities, would
have done. The subjective test requires the Court to establish what the relevant
magistrate would have done, something that would depend on the relevant
magistrate’s evidence or evidence of what he or she had done in similar cases in
the past.

It responded as follows:22

Apart from the fact that the Constitutional Court did not, as I read its judgment,
favour any approach, I have difficulties in accepting the logic of the argument of
the (CPD). The first leg of causation, being a question of fact, cannot depend on
policy considerations such as whether or not a judicial officer should be called
to testify. Causation in this type of case will then no longer be a factual matter of
what the effect of certain conduct on the probabilities ‘would’ have been; it
would then become a value judgment of what it ‘should’ have been. Factual
issues cannot be decided differently depending on the type of case. It has to be
conceded, however, that it would be inappropriate for a particular judicial officer
to testify in relation to the hypothetical question of how he or she would have
decided a particular case. The problem becomes more complicated if, depending
on the organisation of a particular court or hypothetical postponements and the
 like, the identity of the relevant magistrate cannot be established with any
measure of confidence.

The solution to the conundrum appears to be this: The inquiry is subjective in

the sense that a court has to determine what the relevant magistrate on the
probabilities would have done had the application for bail been opposed. In this
regard, the ex post facto evidence of the magistrate would generally amount to
an inadmissible opinion as to what his or her state of mind would have been at
some time in the past. To the extent that the evidence is admissible it would
generally be unhelpful because it would be speculative.

Courts of appeal are often called upon to decide what a reasonable judicial
officer should have done and this they do by establishing what a reasonable
judicial officer would have done. … The proper inquiry is, thus, what the relevant
judicial officer, who is factually assumed to make decisions reasonably, would,
on the probabilities, have done … .23

The Supreme Court of Appeal thus confirmed that one must use the ‘but-for’ test for
determining factual causation, whereby a person eliminates the actual scenario and replaces
it with a hypothetical scenario that would have arisen if the defendants had acted properly.
Based on the facts and evidence, the Court held that if the police and the prosecutor had
acted properly, the magistrate would not have released Coetzee. On a proper application of
the sine qua non test, factual causation was established. In reaching its conclusion, the Court
used a combination of the subjective and objective approaches: it assumed reasonable
decision making on the part of judicial officers in general (an objective element), but
considered what the particular judicial officer probably would have done in such
circumstances.

Lee v Minister for Correctional Services24

The applicant was imprisoned at Pollsmoor prison from 1999 to 2004, with the exception of a
two-month period in 2000 when he had been released on bail. The applicant did not have
tuberculosis (TB) when he entered prison in 1999 but contracted TB sometime before his
release from prison in 2004. The applicant sued the Minister of Correctional Services for
damages arising from him contracting TB. The Constitutional Court accepted that Pollsmoor
prison was overcrowded and that the applicant had been imprisoned and transported to and
from court in very close confines with other prisoners. It was also found that TB was rife in the
prison and authorities did not follow an adequate system of confinement and prevention of
the disease.
The Supreme Court of Appeal25 found that the applicant had not proven factual causation
in terms of the but-for test, but on appeal the Constitutional Court (by a majority) overturned
the decision. The majority noted that the but-for test was not inflexible and where the use of
this test resulted in injustice our courts could utilise a more flexible approach to factual
causation. This flexible approach had a long history in our law and should be used when
common sense had to carry the day.

The Constitutional Court stated:26

[T]he rule regarding the application of the test in positive acts and omission
cases is not inflexible. There are cases in which the strict application of the rule
would result in an injustice, hence a requirement for flexibility. The other reason
is because it is not always easy to draw the line between a positive act and an
omission. Indeed there is no magic formula by which one can generally
establish a causal nexus. The existence of the nexus will be dependent on the
facts of a particular case.
 As is evident from the statement of agreed facts, the applicant was not infected
with TB when he was admitted to Pollsmoor. It is common cause that, on the
evidence on record, it is more probable than not that Mr Lee contracted TB in
prison, rather than outside it. The Supreme Court of Appeal judgment proceeded
on an acceptance of this probability, but it non-suited Mr Lee on the basis that
he failed to prove that reasonable systemic adequacy would have ‘altogether
eliminated’ the risk of contagion; that he does not know the source of his
infection; and that had he known the source it is possible that he might have
been able to establish a causal link between his infection and the specific
negligent conduct on the part of the responsible authorities.

In my respectful view the Supreme Court of Appeal erred in adopting that

approach. The reasons for this are twofold. First, it was not necessary for the
substitution of reasonable alternative measures to determine factual causation
because our law allows for a more flexible approach. Second, even if the use of
a reasonable alternative substitution were necessary in the circumstances, our
law does not require evidentiary proof of the alternative, but merely substitution
of a notional and hypothetical lawful, non-negligent alternative. The purpose of
the exercise is to evaluate the evidence presented by a plaintiff, not to require
more evidence. If the substitution exercise is done in this way, probable factual
causation is established.

The majority held that nothing prevented a court from simply asking whether on the facts of
the case the wrongdoer’s omission probably caused the harm. The Court thus looked at the
circumstances surrounding the applicant’s imprisonment and asked whether these conditions
were more probably the cause of his contracting TB than any other situation. The majority
found that to prove factual causation it would be sufficient to show that the applicant’s risk of
contracting TB would have been reduced had the prison authorities instituted more stringent
systemic measures to reduce infection. Factual causation was found to be present.

PAUSE FOR Could the flexible approach make a difference?

REFLECTION Would the factual scenarios in S v Van As27 and Minister of Police v
Skosana 28 produce the same results if the flexible approach advocated in
Lee v Minister for Correctional Services 29 were to be applied to them?

PAUSE FOR Nature and purpose of factual causation

REFLECTION Factual causation is one component of the element of causation that has
to be satisfied before delictual liability can be established. The purpose of
the factual component is to establish a causal link between the
defendant’s conduct and the plaintiff’s harm or loss. The factual
component is concerned with the sequential progression of how one fact
arises out of another. Courts do this using a process of reasoning that
involves a retrospective analysis of what probably would have occurred,
based on the evidence, and what one can expect to occur in the ordinary
course of human endeavour. Note, however, while the choice of which
test one has to apply does involve a policy, one cannot establish facts by
applying policy considerations. Facts are established by evidence.
However, in Lee’s case the Court held that the evidence does not have to
be equivalent to ‘a control sample in scientific investigation’.30 In addition,
the law in omission cases does not require the plaintiff to provide
 evidentiary proof of what likely alternatives might have eventuated. A
court’s role is merely to examine and assess the evidence presented to it.

6.3 Critique of the conditio sine qua non test

Although courts use the conditio sine qua non test to determine factual causation, the test is not
perfect. There are three main points of critique:
1. The process of reasoning – determining a hypothetical result by eliminating or substituting
conduct – is clumsy and circuitous.
2. The test provides no answer where there are multiple causes.
3. The conditio sine qua non is not a true test for determining factual causation because it is
merely a way of expressing a causal link that has already been determined.

6.3.1 The elimination/substitution process of reasoning is clumsy and

circuitous
In instances that involve positive conduct, the conditio sine qua non test requires that one
eliminates the conduct. The criticism here is that in applying the test judges are required to
determine factual causation in an indirect manner.31 This leads to a circuitous process, which
provides no answer. Judges must eliminate the conduct and imagine how the sequence of events
would have progressed in the absence of the conduct. In this process, judges only have to eliminate
the wrongful act, while keeping the same all the other events and conditions that coexisted with the
wrongful act. In other words, judges are required to determine whether these events and conditions
caused the harm. Yet, to determine this, they also have to eliminate these events and conditions in
terms of the conditio sine qua non test. The result is that, since the act in question and the other
events and conditions have been mentally eliminated, only the harmful event remains and the
process has to start all over again.
In instances that involve omissions, one uses the substitution process. In these cases judges
first substitute the omission with a positive act that they consider would have been probable and
reasonable in the circumstances. They then imagine a sequence of events that would have followed.
In Minister of Safety and Security v Van Duivenboden 32 the Supreme Court of Appeal uttered some
cautionary words to those who apply this process:
There are conceptual hurdles to be crossed when reasoning along those lines (as in
International Shipping Co (Pty) Ltd v Bentley) for once the conduct that actually occurred
is mentally eliminated and replaced by hypothetical conduct questions will immediately
arise as to the extent to which consequential events would have been influenced by the
changed circumstances. Inherent in that form of reasoning is thus considerable scope for
speculation which can only broaden as the distance between the wrongful conduct and its
alleged effect increases. No doubt a stage will be reached at which the distance between
cause and effect is so great that the connection will become altogether too tenuous … .

Nevertheless, the Court was of the opinion that these ‘conceptual hurdles’ should not be ‘unduly
exaggerated’ since a plaintiff does not have to establish factual causation with absolute certainty.
However, the plaintiff does have to prove that the conduct probably caused the harm and that this
entails a ‘sensible retrospective analysis of what would probably have occurred, based upon the
evidence and what can be expected to occur in the ordinary course of human affairs’. 33
6.3.2 The conditio sine qua non test provides no answer in cases of
multiple (or cumulative) causes
Multiple (or cumulative) causation entails two independent acts that simultaneously cause the same
harmful event. For example, if A and B, independently of each other, set fire to C’s house after
which everything C owns is destroyed in the fire, both acts caused C’s loss. However, the conditio
sine qua non test is not effective in determining the factual cause of C’s loss, because when
considered separately, neither A’s nor B’s conduct would qualify as a necessary condition for C’s
loss. If A’s conduct is eliminated, C would still have suffered the loss because of B’s conduct. So,
according to the test, A’s conduct did not cause C’s loss. Similarly, if one eliminates B’s conduct,
C would still have suffered the loss because A’s conduct is still part of the enquiry; and in applying
the conditio sine qua non test, B’s conduct would also not be the factual cause of C’s loss. Clearly
this outcome is untenable, because common sense, logic and the tenets of justice clearly indicate
that both A and B caused C’s loss. In an American case, Summers v Tice,34 the plaintiff was injured
when two hunters simultaneously shot at a quail. If one applies the conditio sine qua non test
following the reasoning just outlined, one would conclude that neither hunter caused the injury,
despite common sense dictating that either one of them, or both, was responsible.

6.3.3 The conditio sine qua non test is not a true test for determining
factual causation
The argument is that instead of the conditio sine qua non test determining factual causation, it is
actually an ex post facto way of expressing a predetermined causal link. The gist of this critique is
that one can only apply the conditio sine qua non test if one already knows the factual cause of the
harm. Otherwise, how would one know which event prior to the harm has to be eliminated?
Consider the following example: A is a terminally ill patient and in a lot of pain. Doctor B, A’s
attending physician, administers a dose of strong pain medication. Shortly thereafter A dies. The
conditio sine qua non test would be unable to indicate the factual cause of A’s death without an
investigation as to what caused A’s death: the illness or a too strong dose of pain medication.

This means that the conditio sine qua non cannot be a test for determining factual causation
because the factual cause of the harm has already been identified by human experience and
knowledge. The conditio sine qua non theory is thus merely a method of expressing:
a priori conclusion, based on knowledge and experience, regarding the existence of factual
causation in terms of the traditional ‘but-for’ formula.35

6.4 Alternatives to the conditio sine qua non approach

Given that the conditio sine qua non approach is unable to cater for all situations, courts in other
countries have considered other methods of determining factual causation. They continue to accept
and apply the conditio sine qua non theory as the primary test, but they do not consider it to be the
only test for factual causation. Similarly, in some instances, our courts have resorted to a common-
sense standard in light of the evidence, human experience and knowledge,36 and also to the material
contribution test.37 Other options, such as the increase in the risk of harm, have also been offered.38
The following sections describe some of the possible alternatives to the traditional conditio
sine qua non approach.

6.4.1 Material contribution

The leading case in South Africa, Minister of Police v Skosana,39 confirmed the conditio sine qua
non as the test for factual causation. However, the Court explained that in determining factual
causation, the question is whether the defendant’s conduct ‘caused or materially contributed to’ the
plaintiff’s harm.
The material contribution test has been used in other jurisdictions,40 but South African courts
have referred to it in a few cases only.41 A possible reason for this is that the material contribution
test denotes a lesser standard of proof and so applying it has the potential to broaden liability,
especially where apportionment of fault is not possible. Trindade and Cane42 point out:
The curiosity of this principle is that a defendant can be held liable for the whole of a loss
even though all that can be proved on the balance of probabilities is that (the defendant)
contributed to it. The principle allows the courts simply to ignore the lack of evidence on the
issue of which factor caused which part of the plaintiff’s loss.

However, the purpose of the element of factual causation is not to limit liability. Its purpose is to
determine whether the defendant’s conduct ‘caused or materially contributed to’ the harm.
Limitation of liability falls to the legal causation enquiry where policy and other considerations
come into play. The tempering effect of legal causation ensures that there is a fair balance between
the interests of the plaintiff, the defendant and society at large.

6.4.2 Common sense

Strict adherence to and rigid application of the conditio sine qua non theory has led to dissatisfying
outcomes and consequent criticism of the theory. In Minister of Safety and Security v Van
Duivenboden 43 the Supreme Court of Appeal recognised that plaintiffs do not have to establish a
factual link with certainty. Plaintiffs only have to establish that the wrongful conduct was probably
the cause of the harm. They do this by ‘sensible retrospective analysis’ of what would probably
have occurred. In Minister of Finance v Gore NO 44 the Supreme Court of Appeal noted that the
‘but-for’ test is a ‘matter of common sense, based on the practical way in which the ordinary
person’s mind works against the background of everyday life experiences’. In Portwood v Svamvur
45
Beadle CJ noted that the conditio sine qua non approach would not lead to an acceptable outcome
on the facts before him, and so he instead followed a ‘direct common-sense approach of the man in
the street’. 46
Portwood v Svamvur47
The defendant’s dog had become entangled in a gate. In trying to release the dog, the
plaintiff was bitten several times. The plaintiff’s claim was based on the actio de pauperie
and, in the alternative, on the lex Aquilia. The Court found that any normal dog would have
bitten a stranger in such circumstances and the pauperien action was dismissed. The
alternative claim was upheld on the ground that the defendant had been negligent in allowing
his dog, which he knew to have a savage nature, to be at large without a muzzle. On appeal,
it was found that the plaintiff’s injuries had been caused by two factors: the savage nature of
the dog, and the fact that it had got caught in the gate. In this case, the savage nature of the
dog was not a conditio sine qua non of the accident. Nevertheless, the Court found the proper
approach in the circumstances would be a common-sense approach and stated that by taking
a ‘juryman’s view’ of the situation, the savage nature of the dog did play a part in it biting the
plaintiff. Based on this approach, the Court found that there was a factual link between the
savageness of the dog and the plaintiff’s injury.
This judgment received criticism for stating that the nature of the dog caused the injury. In
terms of factual causation, the proper approach is to determine whether the defendant’s
conduct caused the harm.48
As a test for factual causation, one cannot accept the common-sense approach, which has little
specific to offer. First, there is no uniform notion of common sense because notions of common
sense vary from person to person. Second, resorting to common sense as a test may avoid properly
explaining the reasons for arriving at the conclusion,49 which in turn might suggest that ‘causal
requirements are a matter of incommunicable judicial instinct’,50 which they are not.
Therefore, although a common-sense approach to factual causation may indicate the
shortcomings of the conditio sine qua non approach, or any other test for factual causation, it
cannot in itself serve as a test to determine a factual link. Instead, common sense should provide a
standard against which one judges the application of a test for factual causation.

PAUSE FOR But-for ’ or ‘common sense’?

REFLECTION Consider the judgments in Minister of Finance v Gore NO 51 and Lee v
Minister for Correctional Services.52 Did the courts apply the ‘but-for’ test
or did the judge apply the common-sense approach as an alternative to
the ‘but-for’ test? Consider the usefulness of having common sense and
human experience as considerations in the articulation of the conditio
sine qua non test, as opposed to common sense being a test separate
from the conditio sine qua non test.

6.4.3 Human experience and knowledge

Neethling, Potgieter and Visser would have us debunk the conditio sine qua non test completely.
They suggest that there is no need for a general all-encompassing test for factual causation because
the question of factual causation does not present many problems. Courts usually determine the
factual link in light of the evidence and probabilities in the circumstances: ‘(T)here are probably as
many tests for factual causation as there are causal links.’ 53 The basis of this approach is that
factual causation depends on the facts of a particular case. A causal link exists where one fact arises
out of another. The point that Neethling, Potgieter and Visser argue54 stems mainly from the fact
that factual causation depends on the facts of each case and is not something that one can apply to
all factual scenarios. Therefore, in terms of this approach, one should determine factual causation
by human experience and knowledge, based on the actual, and not hypothetical facts of the case.
So, where a defendant’s conduct has in any way contributed to the harm suffered by the plaintiff,
factual causation would usually be established.

COUNTER A variation of the common-sense approach

POINT Midgley contends that an approach based on human experience and
knowledge is merely a variation of the common-sense approach, and
thus subject to the same criticism. He suggests that the criterion amounts
to ‘intuition based on facts’, which is essentially no different from a
common-sense view based on facts.55
Consider also whether the common-sense and human experience
approaches would differ from the material contribution approach. What is
the significance in this regard of the approach followed in Minister of
Finance v Gore NO?56

6.4.4 Increasing risk and creating opportunities for occurrence of harm

Creating or increasing the risk of harm is not necessarily the same as directly causing harm. In
some instances of increasing risk, applying the conditio sine qua non test can lead to the absence of
factual causation.57 The rationale for this seems to be that the wrongdoer has not taken the initiative
of (positively) setting in motion the factual chain of events that caused the harm to occur. However,
in some instances there may be sufficient grounds for deviating from the conditio sine qua non test
where a person has increased the risk of harm occurring. In the English case of McGhee v National
Coal Board 58 the plaintiff worked in a brick factory and cycled home every day after work with
brick dust still clinging to his skin and clothes. He contracted dermatitis, and brick dust was a
recognised cause of the disease. The Court held the employer liable for the plaintiff’s harm, in spite
of the possibility of other factors that also could have contributed to the disease. The grounds were
that the employer materially increased the risk of his employee contracting an industrial disease.
Based on the evidence, it was clear that if the employer had provided showers for employees at the
end of a working day, he would have materially reduced the risk of contracting the disease. The
Court thus established factual causation.
When establishing factual causation, the difference between creating the risk of harm
occurring and directly causing the harm can be a very fine line. For example, if someone starts a
fire that causes harm to others, he or she has directly caused harm. If one leaves the door to a house
unlocked and a thief enters the house through this door and steals property, one has set in motion a
chain of events that caused harm by creating an opportunity for harm to occur. By using these
examples, the English writers Hart and Honoré 59 indicate that there is in fact no real difference
between making a material contribution to causing harm and materially increasing the risk of
harm.60 In their argument, the increase of risk approach and the conditio sine qua non test do not
exclude each other because in both situations the conduct (starting the fire and leaving the door
unlocked) set in motion a chain of events that caused harm.

PAUSE FOR Is the increase of risk a necessary or appropriate deviation from the
REFLECTION conditio sine qua non test?
The purpose of factual causation is to establish a link between the
conduct of the defendant and the harm suffered by the plaintiff.
Therefore, if the defendant has (materially) increased the risk of harm to
the plaintiff, can one say that the defendant’s conduct also materially
contributed to, or caused the harm (as required in Minister of Police v
Skosana)61 and therefore satisfies the conditio sine qua non test?
If the answer is yes, the conduct of the thief in the previous example
could be relevant when enquiring whether legal causation is present by
considering the foreseeability test or whether the thief’s conduct was a
novus actus interveniens.62

6.5 Alternative tests for multiple, cumulative and

successive causes
The conditio sine qua non test is not suitable for instances that involve multiple causes. Therefore,
these instances provide ideal opportunities to explore the effect of appropriate alternative tests. For
example, one could easily resolve the causation issue in the example of A and B simultaneously
setting fire to C’s house by applying the material contribution test. There is no need to go further.
However, multiple causes do not necessarily have to occur at the exact same moment. They
can also occur successively. The so-called ‘asbestosis cases’ described next illustrate the
inadequacies of the conditio sine qua non test in successive causation cases. Courts have used other
methods, such as material contribution, risk and common sense, to avoid injustices that would
otherwise result.
Fairchild v Glenhaven Funeral Services Ltd63
The deceased breadwinners had been employed by two employers, both of whom had
exposed the deceased to asbestos dust and fibres. The House of Lords had to decide
whether the successive employers had factually caused the breadwinners’ deaths. The
difficulty in such cases is that the disease develops through a complex process and scientists
do not know the level of exposure to asbestos dust and fibres that can be tolerated without
significant risk of developing the disease. A single fibre or many fibres could cause the
disease. However, once the condition exists it can apparently not be aggravated by further
exposure. Accordingly, if a person is employed successively by employers A and B, that
person could have inhaled the asbestos dust and fibres that caused the condition at any time
during either employment period. The person could have developed the condition by inhaling
a single fibre during employment at A, which means that the exposure during employment at
B would have had no effect on his or her condition. Alternatively, the person could have
developed the condition during employment at B, in which case exposure at employer A
would have had no effect on his or her condition. Therefore, there is no way of identifying the
source of the fibre or fibres that caused the disease. So, applying the conditio sine qua non
test would lead us to conclude that neither of the two employers is liable as one could not
prove that either of them caused the condition. However, Lord Bingham held that it was:

… just and in accordance with common sense to treat the conduct of A and B in
exposing the [breadwinners] to a risk to which [they] should not have been
exposed as making a material contribution to the contracting … of a condition
against which it was the duty of A and B to protect [them].64

Lord Hoffmann and Lord Rodger agreed with Lord Bingham but limited the application of
this approach to specific types of cases.65 Lord Nicholls based his conclusion on the
exposure to risk as long as the risk is not insignificant.66 Lord Hutton drew an inference of
causation in instances where defendants materially increase the risk, and reversed the onus
of proof, with the proviso that one employer can claim a contribution to the damages from the
other employer.67
The importance of this case is that it illustrates that the conditio sine qua non test is not the
exclusive test for factual causation. Underlying enquiries into each and every aspect of
delictual elements is the overall objective of the law of delict ‘to define cases in which the law
may justly hold one party liable to compensate another’.68 So, the conditio sine qua non test
should not be slavishly applied. Where it would lead to an unjust result, it should make way
for a test that would lead to a just result.

PAUSE FOR Cook v Lewis69

REFLECTION In this Canadian case, a single pellet had struck the plaintiff after two
hunters had simultaneously discharged their shotguns in his direction. Is
this an instance of multiple causation? Would the sine qua non test lead
to an appropriate result? Is this an instance in which one should apply an
alternative test? Would alternative tests not lead to the same conclusion
as that reached when applying the sine qua non test? Would a ‘basket of
tests’ in which one looks at a particular case from a variety of angles –
necessary cause, risk, material contribution, common sense and policy –
not be a more satisfactory way of determining factual causation?
6.6 The appropriate test for factual causation
Despite criticism, the ‘but-for’ test continues to be the preferred approach to factual causation.
However, the question arises whether, in light of the flexible approach advocated in Lee v Minister
for Correctional Services,70 courts could follow a different approach in particular circumstances;
and if yes, under which circumstances?
The starting point for determining factual causation should be Minister of Police v Skosana: 71
the conduct in question should have caused (by being a conditio sine qua non) or materially
contributed to the harm. The Skosana approach should be applied in such a way that the result
conforms to common sense and produces a result that does not offend our sense of justice. In most
cases, this should not be difficult. In the cases where the result might initially appear to be over-
extensive in its range, the legal causation element tempers its effect.72
In some instances the test will not be able to produce an appropriate outcome. In these cases
courts should consider all the relevant circumstances and, in the light of relevant policy
considerations, determine whether to apply another test, and if so, which test. It is only when the
circumstances and policy factors justify a deviation from the ‘but-for’ test that one should relax the
initial standard. Therefore, the enquiry is whether the conduct materially contributed to the result. If
one follows this approach, common sense and policy are not tests for determining factual causation.
Instead, like constitutional norms and values, they provide the context in which one applies the
tests, and the yardsticks against which the tests and their results are measured.73

COUNTER Factual causation: ‘but-for’, material contribution or common

POINT sense?
In Minister of Police v Skosana 74 Corbett JA portrayed the test for factual
causation as one that entails both the ‘but-for’ test and an enquiry as to
whether the conduct of the defendant materially contributed to the
plaintiff’s harm;75 and furthermore, that there may be exceptions to the
conditio sine qua non as the (only) generally accepted test for factual
causation.76
In International Shipping Co (Pty) Ltd v Bentley 77 Corbett JA focused
purely on the ‘but-for’ test, and did not mention material contribution as
an alternative.
In Mukheiber v Raath 78 the Supreme Court of Appeal confirmed the
International Shipping Co (Pty) Ltd v Bentley 79 approach, stating that:

as far as factual causation is concerned, this Court follows

the conditio sine qua non – or ‘but-for’ – test.80

Has the material contribution test been discarded? Are there perhaps
instances in which the conditio sine qua non test would be inadequate
and where the material contribution test might nonetheless be more
appropriate?
Consider also Minister of Finance v Gore NO 81 where Cameron JA
stated that the application of the ‘but-for’ test is not based on
mathematics, pure science or philosophy, but is rather a matter of
common sense based on the practical way in which an ordinary person’s
mind works against the background of every-day life experiences. The
Constitutional Court appears to have endorsed this approach in Lee v
Minister for Correctional Services.82
Which view should hold sway? Is there room for all three approaches,
or would that lead to inconsistency?
6.7 Conclusion
It is important to remember that the purpose of establishing a factual link between the defendant’s
conduct and the plaintiff’s harm is to establish factual liability on the part of the defendant. As
indicated in the introduction to this chapter, when determining whether conduct is the factual cause
of harm one needs to conduct a factual investigation into how the harm came about. The normative
question of whether the (factually established) liability should also be recognised in law is a
question that is addressed in terms of the legal causation enquiry.
The prevailing test to determine factual causation is the conditio sine qua non test, which
should be the point of departure for any enquiry into factual causation.83 However, in some
instances this test is not satisfactory and one can use other methods of establishing the factual link.
In deciding upon an alternative method, bear in mind that the ambit of liability is not the main
concern; it will be addressed when applying the tests for legal causation. Nevertheless, any
alternative method should have some flexibility, because courts need to be able to make just and
equitable decisions in terms of the specific circumstances of each case.84 In other words, although
one is concerned with factual issues (as opposed to normative or policy issues) when determining
factual causation, one has to take into account some policy considerations when deciding upon an
alternative test for factual causation. Such policy considerations are thus aids (or way-marks) in
arriving at an appropriate test for factual causation, in the same way that constitutional values
would inform our choice of an appropriate test.85

Figure 6.1 Factual causation

 1 mCubed International (Pty) Ltd v Singer NNO 2009 (4) SA 471 (SCA); H v Fetal Assessment Centre (2015) 2 BCLR
127 (CC), 2015 (2) SA 193 (CC) paras 54 and 60.
2 1977 (1) SA 31 (A).
3 Prosser (1971) Handbook of the Law of Torts 4 ed at 237.
4 Minister of Police v Skosana 1977 (1) SA 31 (A) at 34–35; see also International Shipping Co (Pty) Ltd v Bentley
1990 (1) SA 680 (A) at 700.
5 See also International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) and mCubed International (Pty) Ltd v
Singer NNO 2009 (4) SA 471 (SCA).
6 1977 (1) SA 31 (A).
7 Minister of Police v Skosana 1977 (1) SA 31 (A); S v Van As 1967 (4) SA 594 (A); International Shipping Co (Pty)
Ltd v Bentley 1990 (1) SA 680 (A); Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA) 327 et seq.
However, as will be seen later in this chapter, the conditio sine qua non test is not suitable in all instances. Depending
on the circumstances of each particular case, courts are sometimes required to deviate from the traditional ‘but-for’
test to reach a just and reasonable outcome. These alternative methods for determining factual causation are discussed
in section 6.4.
8 Minister of Finance and others v Gore NO 2007 (1) SA 111 (SCA) para 32.
9 Minister of Finance and others v Gore NO 2007 (1) SA 111 (SCA) para 32.
10 1990 (1) SA 680 (A) at 700.
11 International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 700.
12 Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) para 25.
13 1967 (4) SA 594 (A).
14 1977 (1) SA 31 (A).
15 At 35D–E.
16 2009 (4) SA 471 (SCA).
17 Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA).
18 2004 (3) SA 305 (SCA).
19 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC).
20 2004 (3) SA 305 (SCA) paras 57–58.
21 Para 76.
22 Paras 59–61.
23 Our emphasis.
24 2013 (2) SA 144 (CC).
25 2012 (3) SA 617 (SCA).
26 2013 (2) SA 144 (CC) paras 41–43.
27 1967 (4) SA 594 (A).
28 1977 (1) SA 31 (A).
29 2013 (2) SA 144 (CC).
30 Lee v Minister for Correctional Services 2013 (2) SA 144 (CC) para 56.
31 Neethling and Potgieter (2015) at 187.
32 2002 (6) SA 431 (SCA) para 25.
33 2002 (6) SA 431 (SCA) para 25.
34 199 P 2d 1 (1948).
35 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 176.
36 For example, Portwood v Svamvur 1970 (4) SA 8 (RAD) at 14–15; Thoroughbred Breeders’ Association v Price
Waterhouse 2001 (4) SA 551 (SCA); Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) para
25 as quoted above; Ncoyo v Commissioner of Police, Ciskei 1998 (1) SA 128 (CkS) at 137G.
37 Minister of Police v Skosana 1977 (1) SA 31 (A).
38 For a detailed discussion on possible alternatives, see Midgley ‘Revisiting factual causation’ (2006) at 288–302.
39 1977 (1) SA 31 (A) at 34F.
40 For example Bonnington Castings Ltd v Wardlaw [1956] 1 All ER 615 (HL); Athey v Leonati 1996 140 DLR (4th)
355.
41 See, for example, Kakamas Bestuursraad v Louw 1960 (2) SA 202 (A) at 222; Humphrys NO v Barnes 2004 (2) SA
577 (C).
42 Trinidade and Cane The Law of Torts in Australia 3 ed (1999) at 478.
43 2002 (6) SA 431 (SCA) at 434F.
44 2007 (1) SA 111 (SCA) para 33.
45 1970 (4) SA 8 (RA).
46 At 15.
47 1970 (4) SA 8 (RA).
48 For example, see Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 413–414.
49 Midgley (2006) at 293–294.
50 Fairchild v Glenhaven Funeral Services Ltd [2002] 3 WLR 89; 2002 (3) All ER 305 (HL) para 53.
51 2007 (1) SA 111 (SCA).
52 2013 (2) SA 144 (CC).
53 Neethling and Potgieter (2015) at 185.
54 Neethling and Potgieter (2015) at 185–187.
55 Midgley (2006) at 294–295.
56 2007 (1) SA 111 (SCA).
57 Midgley (2006) at 297–298.
58 1973 (1) WLR 1 (HL).
59 Hart and Honoré Causation in the Law 2 ed (1985) at 59 and 133.
60 See also Fairchild v Glenhaven Funeral Services Ltd [2002] 3 WLR 89; [2002] 3 All ER 305 (HL) para 21 per Lord
Bingham.
61 1977 (1) SA 31 (A).
62 See Sadomba v Unity Insurance Co Ltd 1978 (3) SA 1094 (R).
63 [2002] 3 WLR 89; [2002] 3 All ER 305 (HL).
64 Fairchild v Glenhaven Funeral Services Ltd; [2002] 3 WLR 89; [2002] 3 All ER 305 (HL) para 34.
65 Paras 108, 111, 117 (Lord Hoffmann) and para 170 (Lord Rodger).
66 Para 41.
67 Paras 61 and 71.
68 Para 9.
69 (1952) 1 DLR 1 (SCC).
70 2013 (2) SA 144 (CC).
71 1977 (1) SA 31 (A).
72 Midgley (2006) at 301–302.
73 Midgley (2006) at 302.
74 1977 (1) SA 31 (A).
75 At 35E.
76 At 35C–D.
77 1990 (1) SA 680 (A) at 700.
78 1999 (3) SA 1065 (SCA).
79 1990 (1) SA 680 (A).
80 Para 34.
81 2007 (1) SA 111 (SCA) para 33.
82 2013 (2) SA 144 (CC).
83 Midgley (2006) at 302.
84 Midgley (2006) at 302.
85 Midgley (2006) at 302.
 Chapter 7

Legal causation

7.1 Introduction

7.2 General test – The ‘flexible approach’

7.3 Subsidiary tests

7.3.1 Direct consequences
7.3.2 Reasonable foreseeability
7.3.3 Adequate cause
7.3.4 Intent

7.4 The talem qualem rule |

7.5 Novus actus interveniens

7.6 Conclusion

7.1 Introduction
To hold a defendant delictually liable, there has to be a causal link between the defendant’s conduct
and the harm that the plaintiff suffered. The causation element consists of two components: factual
causation and legal causation. The primary test for factual causation is the ‘but-for’ or the conditio
sine qua non test. However, as was explained in International Shipping Co (Pty) Ltd v Bentley:1
… demonstration that the wrongful act was a causa sine qua non of the loss does not
necessarily result in legal liability. The second enquiry then arises, viz whether the wrongful
act is linked sufficiently closely or directly to the loss for legal liability to ensue or whether,
as it is said, the loss is too remote. This is basically a juridical problem in the solution of
which considerations of policy may play a part. This is sometimes called ‘legal causation’.
A factual link between the defendant’s conduct and the harm is therefore not enough to establish
liability. No legal system will hold people responsible for all the harmful consequences of their
conduct. This would not be fair and just. A person is liable only for the consequences that are
closely linked to his or her conduct, either directly or sufficiently closely and this is where legal
causation, the second component of the causation enquiry, plays a role. Legal causation is used to
limit any liability to those consequences that one can fairly attribute to the defendant. Where
consequences are not linked closely enough to the defendant’s conduct, or where the link for other
reasons is not strong enough, the link is insufficient to hold the defendant liable in law. In such
instances, courts sometimes say that there is no legal causation, while on other occasions they say
that the consequences are too remote.
Since the focus of the legal causation enquiry is on whether liability should arise, this part of
the causation enquiry is a normative one, often involving the weighing up of different factors and
policy considerations. In Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2 the
Court recognised the control function that the legal causation element plays and also that the policy
decision in this regard is not the same as that in respect of wrongfulness:
Even where negligent conduct resulting in pure economic loss is for reasons of policy found
to be wrongful, the loss may therefore, for other reasons of policy, be found to be too remote
and therefore not recoverable.3

The following example from case law illustrates how legal causation operates in practice:
International Shipping Co (Pty) Ltd v Bentley4
The respondent was the appointed auditor of the Deals Group of companies. International
Shipping carried out the business of financiers and shippers. Early in 1976, International
Shipping agreed to make certain financial facilities available to the Deals Group. In March
1979, Bentley issued group financial statements as well as reports in respect of each of the
companies in the Deals Group. These reports were not qualified. Also, Bentley stated that he
had examined the financial statements and that they fairly represented the financial position
of the Group as at 20 December 1978. International Shipping continued to provide financial
facilities until April 1981, when the Deals Group was liquidated. International Shipping at this
stage was owed R977 318, but managed to recover R593 826. This left the company with a
loss of R383 492. International Shipping then proceeded to claim damages from Bentley on
the basis that the financial statements he had prepared were materially false and misleading
in a number of respects. International Shipping alleged that had the 1978 financial statements
fairly presented the financial position of the Deals Group, its constituent companies and the
results of their operations, International Shipping would have terminated the facilities on
receiving those statements. It furthermore would have required the Deals Group to make
good its indebtedness to International Shipping. International Shipping contended that the
loss it had sustained was a consequence of Bentley’s conduct and that Bentley had to make
good its loss.
The Appellate Division found that there was conduct, and that the conduct had been both
wrongful and negligent. The Court also found that the conduct was a conditio sine qua non
for the loss suffered. The question that remained was whether the factual connection
between the conduct and the harm was sufficiently close for the Court to attribute liability to
Bentley.
The Court held that there were a number of factors that indicated that although the conduct
was a sine qua non of the harm, the harm could not be imputed to Bentley. Some of its
reasons included:
• Two years had elapsed between the respondent’s financial reports and the loss.
• International Shipping had decided to provide a support programme for the Deals Group at a
stage when it already knew that the Group’s financial situation was fairly bleak.
 • International Shipping had allowed the Deals Group’s indebtedness to escalate in an
uncontrolled way.
• The relationship between International Shipping and the Group had changed in the
meantime. International Shipping had become involved in the Group’s administration and
had greater insight into the Group’s financial situation.
• An executive of Deals Group had deceived International Shipping, but International Shipping
must have been aware of, or at least suspected, his dishonesty.
• When drafting the financial statements, Bentley could not have foreseen that International
Shipping would enter into a support programme.

The Court found that the connection between the conduct and the loss was not close enough
for liability to arise. It dismissed International Shipping’s claim. This case illustrates a clear
situation in which there is factual causation, but the harm cannot be imputed to the
perpetrator because the legal causation component has not been satisfied.

7.2 General test – The ‘flexible approach’

For many years courts used a number of tests to determine whether the harm that results factually
from the wrongful conduct should be imputed to the perpetrator. These tests include:
• The direct consequences test
• The reasonable foreseeability test
• The novus actus interveniens concept
• The adequate cause test.

At first the direct consequences test, and later the foreseeability test, were favoured. However, as
courts began to express ideas about the normative nature of legal causation decisions and the policy
considerations upon which the decisions are based, other tests emerged. The criminal case of S v
Daniëls 5 illustrates this point clearly. Two perpetrators were charged with murder. Perpetrator 1
had fired two shots into the victim’s back, after which perpetrator 2 fired one shot to the victim’s
head. An autopsy revealed that it was the head wound that killed the victim, but that the shots fired
in the victim’s back would have been fatal if they had not received immediate medical attention.
The Court accepted that even if the victim had not been shot in the head, he still would have died
from the shots in his back. The question that arose was which shot(s) had caused the deceased’s
death. Two of the judges investigated whether there had been a novus actus interveniens, a third
judge looked at the matter from the point of view of adequate causation, and a fourth adopted a
common purpose approach. Jansen JA, who applied the adequate cause theory, mentioned that
policy considerations require a limitation on liability, but he also stated that there was no agreement
as to what the criterion had to be.6
Eventually, in another criminal case, S v Mokgethi,7 the Appellate Division adopted an ‘elastic
test’ to encompass all the existing tests. In this case, five people were accused of several crimes,
including the murder of a bank teller employed at the bank where the five accused had committed
an armed robbery. The deceased had been paralysed after being shot in the spine by one of the five
accused. As a result, he lost all sensation in his lower body. After his discharge from hospital, he
had to avoid pressure sores from forming on his body and so could not sit still for too long.
However, because he had no sensation in his lower body, he could not feel any pain from the
pressure. Eventually he developed pressure sores, which then became septic. This resulted in severe
septicaemia, which eventually affected his organs and caused him to die. The question that arose in
this case was whether the gunshot had caused the man’s death. The Court held that the gunshot was
a sine qua non for his death, but not the legal cause.
 The elastic, flexible, or supple test that the Court developed was based on policy
considerations based upon reasonableness, fairness and justice. The Court also held that tests
previously used to establish legal causation would not be abolished, but could be used as subsidiary
tests depending on the circumstances of each case. Van Heerden JA described the flexible test as
follows: 8
I doubt then whether a legal system could get by without a dominant elastic criterion for the
establishing of legal causation. As is clear from the passages from Skosana and Daniëls …
policy considerations become relevant and the Court has to be careful that the liability of
the perpetrator does not exceed the boundaries of reasonableness, fairness and justice.
These considerations and concepts are not capable of clear definition.

Regarding the different criteria it seems as if they are not more precise than a criterion (the
supple criterion) in terms of which it is determined whether or not a sufficiently close
connection exists between conduct and its consequence. I do not say here that one or more of
these criteria is not capable of useful application as a subsidiary test in the case of a
particular set of facts, but merely that none of the criteria can be regarded as a more
concrete criterion for limitation in the case of all factual situations and for the purposes of
imputing any form of liability.

Soon after this the Appellate Division, in International Shipping Co (Pty) Ltd v Bentley,9 applied
the Mokgethi approach to delictual liability.
So, as a matter of policy, in our law defendants should not be held liable in respect of harm
that is not closely connected to their conduct. The test for legal causation is a flexible one whereby
courts determine whether the factual link is sufficiently strong and closely connected to the conduct
to say that it is fair, reasonable and just to hold the defendant liable for damages. Courts draw the
conclusions regarding fairness, reasonableness and justice after considering the circumstances of
the case and relevant social policy.

PAUSE FOR The flexible approach

REFLECTION It is not entirely clear what the content of the flexible test is. In Mokgethi’s
case Van Heerden JA wanted to develop a dominant elastic test, without
replacing the existing tests. However, when describing the flexible
criterion, he noted that concepts such as reasonableness, fairness and
justice are not easily defined. In relation to this, he referred to a
statement from Blaikie and Others v The British Transport Commission:10

The law has always had to come to some kind of

compromise with the doctrine of causation. The problem is
a practical rather than an intellectual one. It is easy and
usual to bedevil it with subtleties, but the attitude of the law
is that expediency and good sense dictate that for practical
purposes a line has to be drawn somewhere, and that, in
drawing it, the court is to be guided by the practical
experience of the reasonable man rather than by the
theoretical speculations of the philosopher.

This suggests that courts should be guided by practical considerations of

‘common sense’ and ‘expediency’. So, one should use the specific tests
only in appropriate circumstances, while the flexible criterion acts as a
conceptual container within which all the other tests are accommodated
and used when appropriate.
Mokgethi (and thereafter International Shipping) did not abolish the subsidiary tests. Any
uncertainty in this regard, or regarding the relationship between the flexible criterion and the
subsidiary tests, was clarified in Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd. 11
Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd12
SA National Roads Agency was authorised to collect toll fees at two toll plazas. Fourway
Haulage was transporting asbestos and one of their drivers caused an accident, which
resulted in this highly toxic substance spilling all over the road. The road was closed and
traffic diverted to a non-toll road while the spillage was being cleaned up. Because of this, the
SA National Roads Agency lost money and sued Fourway Haulage for damages. Regarding
legal causation the Court noted:13

What Van Heerden JA said in that case (Mokgethi) is not that the ‘flexible’ or
‘supple’ test supersedes all other tests such as foreseeability, proximity or
direct consequences, which were suggested and applied in the past, but merely
that none of these tests can be used exclusively and dogmatically as a measure
of limitation in all types of factual situations. Stated somewhat differently: the
existing criteria of foreseeability, directness, et cetera, should not be applied
dogmatically, but in a flexible manner so as to avoid a result which is so unfair
or unjust that it is regarded as untenable. If the foreseeability test, for example,
leads to a result which will be acceptable to most right-minded people, that is
the end of the matter ….

In this case it can, in my view, be accepted with confidence that any of the
various criteria will lead to the conclusion that the loss suffered by the Agency
is not too remote. If, for example, the direct consequences criterion is applied, it
is clear that the loss followed directly from the wrongful and negligent conduct
of Fourway’s driver; there was no so-called novus actus interveniens that broke
the chain of events. If, on the other hand, one applies the foreseeability test, it
was in my view reasonably foreseeable that a collision could cause spillage and
that, because of the dangerous nature of the cargo, spillage could result in the
closure of the toll road which could lead to a revenue loss by the Agency.

The Court here set out the relationship between the flexible approach and the subsidiary
tests. However, it also cautioned against using concepts such as reasonableness, fairness
and justice to determine legal causation:14

Considerations of fairness and equity must inevitably depend on the view of the
individual judge. In considering the appropriate approach to wrongfulness, I said
that any yardstick which renders the outcome of a dispute dependent on the
idiosyncratic view of individual judges is unacceptable. The same principle
must, in my view, apply with reference to remoteness. That is why I believe we
should resist the temptation of a response that remoteness depends on what the
judge regards as fair, reasonable and just in all the circumstances of that
particular case. Though it presents itself as a criterion of general validity, it is, in
reality, no criterion at all.

These comments signal that in the future there may be another variation in the development
of the elastic test for legal causation. If, as the Court suggested, the test is stripped of these
considerations as criteria for determining legal causation, then the decision becomes one of
policy in which ‘tests such as foreseeability, proximity or direct consequences’ play a part.
The flexible criterion would not replace these tests, but they would be applied ‘in a flexible
manner so as to avoid a result which is so unfair or unjust that it is regarded as untenable’.15
 Based on the facts, the Court found that the Road Agency’s loss was not too remote, for it
was reasonably foreseeable that an accident that involved a truck carrying hazardous cargo
could lead to a road closure, resulting in a loss of toll revenue.

7.3 Subsidiary tests

Prior to the Appellate Division’s adoption of the flexible test, courts used a number of tests to
determine legal causation. There were two main tests that they used: at first the direct consequences
test, and later the test of reasonable foreseeability. This section describes some of the tests that have
been used in the past.

7.3.1 Direct consequences

The direct consequences theory, also known as the ‘proximate cause’ test, originated in the English
case of In re Polemis and Furness, Withy & Co Ltd 16 and dominated for about four decades. In the
Polemis case the defendant had hired a ship. Prior to this contract, petrol had leaked into the ship’s
hold, producing petrol fumes. One of the defendants’ employees dropped a plank into the hold,
causing a spark, which in turn started a fire that completely destroyed the ship. The defendants
argued that they could not be held liable because the damage that resulted could not have been
foreseen. The Court rejected this argument and held:17
If the act would or might probably cause damage, the fact that the damage it in fact causes
is not the exact kind of damage one would expect is immaterial, so long as the damage is in
fact directly traceable to the negligent act, and not due to the operation of independent
causes having no connection with the negligent act, except that they could not avoid its
results. Once the act is negligent, the fact that its exact operation was not foreseen is
immaterial.

According to this approach, persons are liable for all the direct consequences of their conduct if
they should have reasonably foreseen that their conduct would have caused harm of some kind to
the plaintiff. However, liability is not limited to foreseeable or probable consequences: as long as
the consequences result directly from the conduct, they may be imputed to the defendant. Only
where a new intervening cause (a novus actus interveniens) breaks the causal link will there be no
legal causation and therefore no liability.
The ‘proximate cause’ test can result in exceptionally wide liability. As a result, courts tend to
limit liability to direct physical consequences. They also apply the policy of the foreseeable
plaintiff, in which liability is limited to the category of persons that could reasonably have been
foreseen as being likely to suffer harm as a result of the defendant’s conduct.
The direct consequences test has not found much favour in South African law. It was used in
an old case, Frenkel & Co v Cadle,18 while in other instances it was used together with the test for
reasonable foreseeability. For example, in Thandani v Minister of Law and Order,19 an unlawful
arrest and detention case, the plaintiff had been arrested by the South African Police and then
handed over to the Ciskei security police. Counsel for the defendant argued that while the
defendant could be held liable for the unlawful detention of the plaintiff by the South African
Police, he was not liable for the plaintiff’s detention by the Ciskei Police, even though the arrest
and detention by the South African Police was the factual cause of the detention by the Ciskei
Police. The Court found that the legal causation criterion had been satisfied:20
As I see the position the unlawful handing over of the plaintiff to the Ciskei Security Police
was the cause of his being incarcerated in Ciskei. His incarceration in Ciskei was not only a
 direct consequence of his being handed over to the Ciskei Security Police, but was also a
reasonably foreseeable and in fact an intended consequence thereof.

The Thandani approach also illustrates the complementary manner in which one can use the tests
when applying the flexible criterion.

7.3.2 Reasonable foreseeability

According to the foreseeability approach, the question is whether the defendant should reasonably
have foreseen the consequences that resulted from his or her conduct. If the answer is ‘yes’,
liability results. The test was first used in English Law when the Privy Council, in Overseas
Tankship (UK) Ltd v Morts Docks & Engineering Co Ltd (The Wagon Mound No 1),21 rejected the
Polemis test on the basis that the direct consequences approach could be unduly harsh on the
defendant in that case. In the Wagon Mound case the plaintiffs claimed damages for harm suffered
as a result of a fire. Workmen, who were employees of the defendant’s company, had negligently
allowed furnace oil to spill into the Sydney harbour. The oil spillage thickly coated the water and
drifted underneath the wharf. Some hot metal fell into the water and onto some cotton waste in the
water. This ignited the oil and started a fire that damaged the wharf and some equipment. The harm
was clearly a direct consequence of the oil spillage, but the Privy Council held that liability should
arise only in respect of harm that was reasonably foreseeable. Viscount Simonds remarked: 22
Enough has been said to show that the authority of Polemis has been severely shaken though
lip-service has from time to time been paid to it. In their Lordships’ opinion it should no
longer be regarded as good law. It is not probable that many cases will for that reason have
a different result, though it is hoped that the law will be thereby simplified, and that in some
cases, at least, palpable injustice will be avoided. For it does not seem consonant with
current ideas of justice or morality that for an act of negligence, however slight or venial,
which results in some trivial foreseeable damage the actor should be liable for all
consequences however unforeseeable and however grave, so long as they can be said to be
‘direct’. It is a principle of civil liability, subject only to qualifications which have no present
relevance, that a man must be considered to be responsible for the probable consequences of
his act. To demand more of him is too harsh a rule, to demand less is to ignore that civilised
order requires the observance of a minimum standard of behaviour.

The reasonable forseeability test does not require that one foresees all the harm or its full extent, or
that one sees the specific harm. What is required by this test is that one foresees the general type of
harm that occurred.23 It is also not necessary to have foreseen the precise way in which the harm
occurred.24 These points are well illustrated by the facts of Smit v Abrahams 25 and the discussion of
the foreseeability test in the decisions of both the Appellate Division 26 and the Court a quo.27 The
plaintiff’s vehicle had been damaged in a collision caused by the defendant. The plaintiff claimed
damages for the repair costs of the vehicle and also for loss of profits (the vehicle had been used as
a delivery truck). The defendant accepted liability in respect of the first claim, but contested the
second. The plaintiff argued that he had tried to mitigate his loss by renting a replacement vehicle
from an acquaintance because the car rental agencies were too expensive. This contract ended after
three months, and because the plaintiff could not afford to either buy or rent another vehicle, he
suffered a loss of profit. Both courts held that the plaintiff’s financial problems were reasonably
foreseeable and accordingly, the losses that he had suffered by being unable to replace his vehicle
had to be compensated.28

7.3.3 Adequate cause

The adequate causation test originated in The Netherlands and in Germany.29 In this approach, a
consequence is imputed to a defendant if it is adequately connected to his or her conduct. Whether
something is adequately connected depends on whether, according to human experience, that
consequence would follow the conduct in the normal course of events; in other words, whether that
conduct would have the tendency to bring about that consequence.
The adequate cause theory has not been used pertinently in the South African law of delict,
although Jansen JA referred to it in the criminal case of S v Daniëls: 30
Without the connection between the perpetrator’s conduct and the alleged consequence
there can generally be no liability. On the other hand it is also clear that a perpetrator
cannot be held liable for all the consequences of which his conduct is a sine qua non – his
liability would then be too wide and thus exceed the boundaries of reasonableness, equity
and justice. Policy considerations require that liability should somehow be limited.

In the present case the conduct of the first appellant was a causa sine qua non of the death of
the deceased; if the first appellant had not taken out the gun and shot the deceased in the
back so that he fell down, the second appellant would not have shot him in the head as it
happened in this case … . The fault requirement has been met and apparently also that of
unlawfulness. The shots fired by the first appellant were deadly and would in any event have
resulted in death. According to human experience the shots fired by the first appellant
would in the ordinary course of events have resulted in death as a result of a gun shot. These
shots could … be regarded as adequate regarding the death … .

PAUSE FOR The adequate cause test

REFLECTION Jansen JA included reasonableness, equity and justice among the policy
considerations to be considered when deciding whether, according to
human experience, a consequence is adequately connected to the
conduct. Is the flexible Mokgethi criterion that now prevails in our law not
merely the adequate cause test in another guise?

7.3.4 Intent
Boberg, in line with a commonly held belief, contends that ‘intended consequences, however
strangely they may come about, can never be too remote.’ 31
While we can agree with this view in most instances, intention cannot be the determining
limiting factor in all cases where intended consequences arise. The question in each case is still a
normative one, and one should ask whether the link between conduct and the intended
consequences is sufficiently strong for liability to arise. While one can fairly say that intended
consequences are not too remote in the sense that they were indeed foreseeable, liability might be
denied because of other policy factors. The determining test is the Mokgethi test and intention, as
with foreseeability, is a relevant but secondary means of reaching a conclusion. In some instances,
it might be fair, reasonable and just to impose liability, as was the case in Groenewald v
Groenewald.32 In this case the plaintiff’s husband had assaulted her and he repeatedly threatened to
kill her, holding a knife to her throat. He had also called various people to tell them that he intended
to kill her. When the defendant was out of the room, which was on the third floor of a building, the
plaintiff tried to escape by climbing out of the window onto a ledge, from where she fell to the
ground and sustained serious injuries. The defendant denied that he had caused the injuries,
contending that the plaintiff’s attempt to lower herself from the ledge, which resulted in her falling,
constituted a novus actus interveniens. The Court disagreed. It held that there were no policy
considerations that militated against liability and that it was reasonable, fair and just to hold the
defendant liable for the injuries that the plaintiff had sustained. In other cases, however, courts have
denied liability because the intended result came about coincidentally and in a materially different
way from what the defendant had contemplated.33
Also, it is clear that a person cannot be liable only for intended consequences and not for
unintended ones. For example, in cases of assault, a person could cause more harm than had been
intended, and it would be unreasonable to hold the person responsible only for the intended
consequences. In Brown v Hoffman 34 the Court noted that in cases of negligent conduct, people are
held liable for unintended consequences. So it would be untenable, in cases where some harm was
intended, not to hold a person liable because he or she had not intended the exact consequences that
occurred.

PAUSE FOR Legal causation and wrongfulness

REFLECTION Legal causation and wrongfulness both limit liability and are normative
enquiries. Also, in both elements, courts have to resort to policy
considerations and have to base their conclusions on concepts such as
fairness, reasonableness and justice. Is there a need for both elements?
This is what the Supreme Court of Appeal said in Fourway Haulage SA
(Pty) Ltd v SA National Roads Agency Ltd:35

In the final analysis, the issue of remoteness is again

determined by considerations of policy. Broadly speaking,
wrongfulness – in the case of omissions and pure
economic loss – on the one hand, and remoteness on the
other, perform the same function. They are both measures
of control. They both serve as a ‘longstop’ where most
right-minded people, including judges, will regard the
imposition of liability in a particular case as untenable,
despite the presence of all other elements of delictual
liability.

Since wrongfulness – in the context of omissions and pure

economic loss – and remoteness are both determined by
considerations of policy, a certain degree of overlapping is
inevitable. However, wrongfulness and remoteness are not
the same. They involve two different enquiries in respect of
two different elements of delict, each with its own
characteristics and content … . Even where negligent
conduct resulting in pure economic loss is for reasons of
policy found to be wrongful, the loss may therefore, for
other reasons of policy, be found to be too remote and
therefore not recoverable. An example of a case where this
happened is to be found in a decision of this court in
International Shipping Co (Pty) Ltd v Bentley (supra).

Would you agree?

7.4 The talem qualem rule

In some cases a victim may suffer more serious harm than a perpetrator intended, because of some
prior weakness of the victim, which the perpetrator may not have known about. This weakness may
be physical, psychological or financial. In terms of the talem qualem rule (also known as the ‘thin
skull’ or ‘egg skull’ rule), persons take their victims as they find them. So where a perpetrator
foresees some harm, but causes more harm to a victim than what would normally be expected,
because of a pre-existing condition, the perpetrator will be liable for the full extent of the harm. In
the English case, Smith v Leech Brain and Co Ltd,36 the plaintiff’s deceased husband had been
struck on the lip by a piece of metal, causing a burn. Three years later the burn became cancerous
and as a result, the husband died. Medical evidence established that he had suffered from a prior
physical condition that made him susceptible to cancer. The Court held that the ‘thin skull’ rule
applied, and that the deceased’s employer, who had failed to provide him with protective clothing,
was liable to his widow in damages. Smit v Abrahams,37 discussed previously, is another example of
how one can apply this rule. In this regard, Boberg remarks:38
The so-called ‘weak heart’ and ‘thin skull’ cases … are sometimes regarded as inimical to
the application of the foreseeability test to the consequences of conduct.

For the defendant is held liable for unforeseeable additional harm suffered by the plaintiff
as a result of a pre-existing physical condition that renders him more vulnerable to injury
… . It is submitted that the rule pertains to the extent of the harm suffered by the plaintiff,
which need not be foreseeable, and therefore coexists peacefully with the foreseeability test.

The final result of this rule is that where one foresees the general nature of the harm (for example,
that a failure to provide protective clothing could result in bodily injury to a worker), one would be
liable for all the harm within that general category of harm (bodily injuries). This is true
irrespective of any pre-existing condition that might result in harm that one would not normally
expect from such conduct.

7.5 Novus actus interveniens

A novus actus interveniens is:
an independent, unconnected and extraneous factor or event which is not foreseeable and
which actively contributes to the occurrence of harm after the defendant’s original conduct
has occurred.39

The presence of such an intervening cause breaks the causal link between the perpetrator’s conduct
and the ensuing harm. To determine this, courts now apply the flexible test for legal causation. The
intervening event could be the conduct of the victim, the conduct of another person, or it could be
due to other factors. In Mafesa v Parity Versekeringsmaatskappy Bpk (In Likwidasie) 40 the plaintiff
was involved in a motor vehicle accident and suffered a fractured leg. He underwent surgery to
insert a steel plate. When the bone had set sufficiently, the plaintiff was discharged and given
crutches to help him walk. His leg was still in a cast and he was told not to put unnecessary weight
on it. While walking on a smooth floor, the plaintiff slipped and fell because the crutches did not
have rubber tips. The plaintiff fractured his leg again in this accident, which required a second
operation on his leg. When the plaintiff sued the insurer of the motor vehicle, the Court found that
the second incident constituted a new intervening cause because it had not been reasonably
foreseeable that the plaintiff, who had been warned of the dangers, would recklessly attempt to
walk on a slippery floor. Liability was accordingly limited to those consequences that arose prior to
the second incident and the insurer was not liable for expenses arising from the second incident,
such as the costs of the second operation.
Mafesa’s case also illustrates another point: whether something is regarded as a novus actus
depends on whether that event was reasonably foreseeable. If an event is reasonably foreseeable, it
cannot be seen as an independent act. In RAF v Russell 41 the wife of a man involved in a motor
vehicle accident claimed loss of support on behalf of the children. The man had committed suicide
after suffering serious injuries. The defendant claimed that the suicide was a novus actus
interveniens. However, the Court found that the depression that caused him to take his own life was
due to brain damage that he had suffered as a consequence of the accident. The suicide was not
unrelated to the accident, and because the suicide was a direct consequence of the injury, it did not
constitute a novus actus interveniens.
In Premier of the Western Cape Province v Loots NO 42 a botched sterilisation led to Mrs Loots
falling pregnant. She was offered the opportunity to abort the baby, but declined for religious
reasons. Subsequent complications in the pregnancy led to her losing the baby and suffering
personal irreversible brain damage which left her almost blind, unable to walk and talk, and
suffering from dementia. The Western Cape Government argued that the harm suffered by Mrs
Loots was too remote for legal causation to be present, and that her decision not to abort the baby
functioned as a novus actus interveniens. The Court held that, for the decision to function as a
novus actus, it would have to be an unreasonable one, and that reasonable conduct cannot absolve
the wrongdoer of liability. The Court concluded that Mrs Loots’ decision not to abort the baby was
not unreasonable: her pregnancy appeared to be normal and healthy, and there had been no
indication that continuing with her pregnancy would endanger either herself or her child. As to the
foreseeability of the complication which led to the baby’s death and Mrs Loots’ harm, the Court
held that, while the specific consequence had not been foreseeable, it was foreseeable that any
pregnancy could deliver complications; and therefore harm of a general kind had been reasonably
foreseeable. Accordingly, the Court concluded that legal causation was present.

7.6 Conclusion
For liability to arise, there has to be a causal connection that is strong enough to be classified as a
causal connection in law. The test for legal causation limits liability in that harm factually linked to
the perpetrator’s conduct might not be imputed to the perpetrator, because the factual link is not
considered strong enough in law. So, there could be circumstances in which a court finds a person
to have acted both wrongfully and negligently, and that the conduct factually caused the harm, but
where in all fairness the connection between the conduct and the harm is too tenuous for liability to
arise. The causation element in delict, therefore, not only links the plaintiff’s harm to the
perpetrator’s conduct, but also restricts the extent of any liability that might result.
Figure 7.1Legal causation

In essence:
• To hold a perpetrator delictually liable, one has to establish a causal link between the
wrongful, culpable conduct and the loss that the victim suffered.
• The fact that the conduct was a sine qua non (a factual cause) of the loss is not enough to
satisfy the requirement of causation. The plaintiff must also establish that there was legal
causation; in other words, that the loss must not be too remote.
• Courts use a flexible test for legal causation. This test is based on the policy considerations of
reasonableness, fairness and justice.
• When the flexible test was introduced, courts did not abolish the various tests they had used in
the past; instead, the flexible test accommodates all these other tests.

1 1990 (1) SA 680 (A) at 700H–I.

2 2009 (2) SA 150 (SCA).
3 Para 32.
4 1990 (1) SA 680 (A).
5 1983 (3) SA 275 (A).
6 At 331C–D.
7 1990 (1) SA 32 (A).
8 S v Mokgethi (16/1989) [1989] ZASCA 105; [1990] 1 All SA 320 (A) (18 September 1989). At 40–41 (our
translation).
9 1990 (1) SA 680 (A).
10 1961 SC 44 at 49.
11 2009 (2) SA 150 (SCA).
12 2009 (2) SA 150 (SCA).
13 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) paras 34–35.
14 Para 33.
15 Para 34.
16 [1921] 3 KB 560.
17 At 577.
18 (1915) 36 NPD 173.
19 1991 (1) SA 702 (E).
20 At 705E–F.
21 (1961) AC 388.
22 At 422–423.
23 Overseas Tankship (UK) Ltd v Morts Docks & Engineering Co Ltd (The Wagon Mound No 1) [1961] AC 388;
Standard Chartered Bank of Canada v Nedperm Bank Ltd 1994 (4) SA 747 (A) at 768; Smit v Abrahams 1992 (3) SA
158 (C) at 164.
24 Da Silva v Coutinho 1971 (3) SA 123 (A) at 148; Smit v Abrahams 1992 (3) SA 158 (C) at 164.
25 Smit v Abrahams 1992 (3) SA 158 (C) at 163–165.
26 Smit v Abrahams 1994 (4) SA 1 (A) at 17–19.
27 Smit v Abrahams 1992 (3) SA 158 (C) at 163–165.
28 However, the Appellate Division differed from the Court a quo on the role of the foreseeability test:
In the judgment of the Court a quo, the question of reasonable foreseeability of the harm … was
regarded as the single determining criterion for determining liability. In the light of the decisions in
Mokgethi (supra) and International Shipping Co (supra) that approach was not correct. Reasonable
foreseeability may well be used as a subsidiary test in the application of the flexible test, but it cannot
supersede it. (Smit v Abrahams 1994 (4) SA 1 (A) at 17, our translation.)
29 In the Netherlands, it was initially applied to limit damages, but the Dutch Supreme Court eventually rejected it in
favour of the so-called ‘toerekeningsleer’, which considers a multiplicity of factors. The New Civil Code has also
adopted the ‘toerekeningsleer’ (Boonekamp, Art 98, aant. 2.5. in Bloembergen (Ed) Schadevergoeding (1992)).
30 1983 (3) SA 275 (A) at 331–332 (our translation).
31 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 440. See Thandani v Minister of Law and Order 1991
(1) SA 702 (E).
32 1998 (2) SA 1106 (SCA) at 1114.
33 Jowell v Bramwell-Jones 1998 (1) SA 836 (W) at 884. See also Neethling and Potgieter Neethling-Potgieter-Visser
Law of Delict 7 ed (2015) at 208–209 where they cite an example of a young man who persuades his aunt to undertake
a journey, expecting her to have an accident, which in fact does occur as a result of bad weather. They contend that
her death would not be imputed to the nephew in such circumstances.
34 1977 (2) SA 556 (NC).
35 2009 (2) SA 150 (SCA) at 31 and 32.
36 (1962) 2 QB 405.
37 1994 (4) SA 1 (A).
38 Boberg (1984) at 278–279.
39 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 184.
40 1968 (2) SA 603 (O).
41 2001 (2) SA 34 (SCA).
42 2011 JDR 0250 (SCA).
Chapter 8

Fault

8.1 Introduction

8.2 Accountability
8.2.1 Youth
8.2.2 Mental disease or illness and emotional distress
8.2.3 Intoxication
8.2.4 Provocation

8.3 Intention
8.3.1 Dolus directus or direct intention
|
8.3.2 Dolus indirectus or indirect intention
8.3.3 Dolus eventualis or intention by acceptance of foreseen result
8.3.4 First component of intention: Direction of will
8.3.5 Second component of intention: Consciousness of wrongfulness
8.3.6 Special cases: Intention comprising direction of will only
8.3.7 Difference between motive and intention
8.3.8 Proving intention

8.4 Defences that exclude intention

8.4.1 Mistake
8.4.2 Jest
8.4.3 Intoxication
8.4.4 Provocation
8.4.5 Emotional distress

8.5 Negligence
8.5.1 The concept of negligence
8.5.2 Characteristics of a reasonable person
8.5.3 The test for negligence
8.5.3.1 Foreseeability and preventability
8.5.3.2 Foreseeability of harm
8.5.3.3 Preventability of harm
8.5.4 Circumstances and factors that indicate the required standard of care
8.5.4.1 General practice
 8.5.4.2 Legitimate assumption of reasonable conduct of others
8.5.4.3 Sudden emergency and error of judgement
8.5.4.4 Breach of statutory duty
8.5.4.5 Dealing with inherently dangerous things, persons or circumstances
8.5.4.6 Danger to children or people with disabilities or incapacities
8.5.5 Attributes of defendants that influence the standard of care required for
reasonable conduct
8.5.5.1 Beginners
8.5.5.2 Experts
8.5.5.3 Children
8.5.6 Proving negligence

8.6 Conclusion

8.1 Introduction
To establish delictual liability, it is not enough to show that the harm was caused wrongfully. One
must also show that the defendant was at fault. The fault element has two components: (a) the
person must have been accountable at the time of causing the harm (that is, the person must have
had the capacity to be at fault), and (b) the person must have been culpable or blameworthy (that is,
the person must have acted either intentionally or negligently).1 So, in this chapter, we first consider
the issue of a person’s legal capacity to be at fault and we set out the requirements for establishing
a person’s accountability. Thereafter we discuss the nature of intention and negligence and the
requirements that need to be met before a person can be said to be culpable.
Accountability focuses on a person’s ability and maturity; culpability focuses on a person’s
mindset or conduct. So, in a sense these concepts have a subjective aspect to them. In the next
sections, we describe how accountability and intention are both entirely subjective, in that one must
investigate the ability and maturity (accountability) and/or the state of mind (intention) of the
particular individual in question, and how negligence involves a more objective assessment, in that
one measures the individual’s conduct against the standard set by society. However, negligence
also has subjective elements, because one assesses the situation by placing the reasonable person in
the position of the defendant and considering the specific circumstances at the time.

TERMINOLOGY Fault
Fault, as an element of delictual liability, requires that one must be
able to blame and hold a person responsible for the harm that was
wrongfully caused to another. One is thus concerned with the
question of whether or not a person is blameworthy. The Latin
term for fault is culpa in the wide sense. Therefore, a person’s
blameworthiness is referred to as a person’s culpability. Fault
generally takes two forms: intention (dolus) and negligence (culpa
in the narrow sense).

Dolus
The Latin term for intention is dolus. Under the actio iniuriarum,
intention is referred to as animus iniuriandi (translated as ‘the will
to injure’). The term ‘intention’ is a technical legal expression that
describes a wrongdoer’s will to achieve a specific wrongful
 consequence and refers to a person’s state of mind regarding
wrongful conduct and its consequences. In Minister of Justice v
Hofmeyr 2 the Court defined intention as ‘a reflection of the will’. In
contrast to negligence, the test for intention is entirely subjective,
because it involves evaluating a defendant’s mindset (or will) in
relation to the particular consequences and whether the defendant
actually knew that causing such consequences would be wrongful.

Culpa
In this chapter we use culpa in the context of its narrow application
of denoting negligence as a form of fault. Culpa, or negligence,
refers to a situation where a person has not met the standard of
conduct that society deems appropriate in the circumstances. This
means that one evaluates a person’s conduct according to a
general standard of care as required by law. The standard is
expressed with reference to a ‘reasonable person’. Therefore,
negligence involves an objective evaluation of a person’s conduct,
because one measures the conduct against a standard outside the
particular person’s mindset.

Accountability
Accountability in the law of delict means the capacity to be
blamed, or the capacity to be at fault. Our law will not hold
accountable someone who does not have the ability, or sufficient
maturity, to know the difference between right and wrong and to
act in accordance with such knowledge. So, before one can
enquire into whether a person is at fault, either in the form of
intention or negligence, one must establish whether that person is
capable of being blamed.

8.2 Accountability
Accountability is a prerequisite for finding a person blameworthy, or at fault. The concept refers to
a person’s capacity to distinguish between right and wrong, and then to act in accordance with that
distinction. If a person is not legally accountable, one cannot impute blame, and the element of
fault is not satisfied. Culpability refers to the law’s judgement of an accountable person’s state of
mind (intention), or of the inadequate quality of a person’s conduct as measured against society’s
standards (negligence). Therefore, to establish fault, the first step is to enquire into the defendant’s
accountability. Then, only when one has established accountability, does one move on to the
second step and evaluate the defendant’s culpability. Accountability is, therefore, the basis for
blameworthiness to the extent that if a person lacks accountability, at the time the delict was
committed, there can be no fault on that person’s part.
To assess whether a defendant is accountable, one must have an affirmative answer to two
questions.3 Did the defendant, at the time the delict was committed:
1. Have the mental capacity to distinguish between right and wrong, and appreciate the
difference?
2. Have sufficient maturity to act in accordance with the appreciation of a distinction between
right and wrong?
When answering these two questions, it is important to keep in mind the entirely subjective nature
of the concept of accountability: one is concerned with the defendant’s ability, at the time of the
delict, to distinguish between right and wrong, and to act accordingly.

PAUSE FOR Assessing accountability

REFLECTION When answering these two questions, there are four important aspects:
1. Both questions must be answered in the affirmative to establish that a
defendant is accountable. So, if the first question is answered in the negative,
there is no need to proceed to the second question. If the first question is
answered affirmatively and the second question is answered in the negative,
there is still no accountability, and no blame can be imputed.
2. Accountability is an entirely subjective concept where one assesses the
defendant’s ability to distinguish between right and wrong, and to act
accordingly. One does not assess the category of persons into which the
defendant might fall.
3. The ability to distinguish between right and wrong is based on both knowledge
and an appreciation of what is right and wrong.
4. The relevant moment for assessing accountability is the time when the delict
was committed.

There is a general presumption that persons (natural and juristic) are accountable, or culpae capax,
for harm that they have wrongfully caused. However, the law recognises that certain personal
attributes or circumstances will either exclude accountability, or negate the presumption of
accountability. There is no closed list, but the following are recognised categories where persons
could be found to lack accountability, or the capacity to be at fault:
• Youth
• Mental illness
• Intoxication or a similar condition induced by a drug
• Anger due to provocation.

The existence of these attributes or circumstances forms the grounds upon which a defendant can
show the absence of accountability. In other words, these attributes can be used as defences to
exclude accountability.

8.2.1 Youth
There are three categories where youth may exclude accountability:
1. Children who are younger than seven years (infantes): Children in this category are
irrebuttably presumed to be culpae incapax or ‘without legal capacity’. Therefore, irrespective
of the child’s actual mental capacity, children under seven years of age always lack legal
accountability.
2. Children between the ages of seven and fourteen: In these instances, there is a rebuttable
presumption that a child over the age of seven, but under the age of fourteen is culpae incapax.
Therefore, unless proven otherwise in any particular case, children in this category are
regarded as legally incapable of being blamed.
3. Children between fourteen and eighteen years of age: Children in this category are presumed
to be culpae capax. Unless proven otherwise, children in this category are legally accountable
and liable for their wrongful conduct.
Weber v Santam Versekeringsmaatskappy Bpk4
While playing in the sand in front of a block of flats, Marius Weber (seven years and two
months old) was run over by a motorist who was reversing from a parked position. The issue
was whether Marius had been contributorily negligent. For such a finding, the Court had to
determine Marius’ capacity to be at fault. Evidence indicated that Marius’ mother had on
previous occasions told him to keep a lookout for cars, and also that Marius had enough time
to assess the situation and ensure he was playing out of harm’s way.
The Appellate Division confirmed that the test for accountability is a subjective one where
the focus should be on Marius’ mental capacity at the time of the accident, with due regard to
the child’s abilities in the particular situation. It reiterated the cautionary rule in respect of
children aged just over seven, as well as the importance of presuming that children between
seven and fourteen are culpae incapax.5 The Court also warned against ‘placing an old head
on young shoulders’,6 and so, in assessing Marius’ capacity it took into account the fact that
he had only just turned seven, as well as his intellectual development, maturity and
impulsiveness. The Court found that, although Marius was aware of the danger of cars, the
‘inherent weakness associated with tender age and the propensity of children’7 caused him to
have a child-like preoccupation with making roads in the sand, which in turn deprived him of
a consciousness of the surrounding activities and his mother’s previous warnings. The Court
accordingly held that Marius was culpae incapax, not contributory negligent, and that he was
entitled to the full amount of his damages. In essence, therefore, while Marius might have
understood the difference between right and wrong, he lacked the maturity to act in
accordance with such understanding.

Eskom Holdings Ltd v Hendricks8

Hendricks’ son, Jacques (who was eleven years and eight months old), sustained serious
injuries when he went too close to a high-voltage power line that was suspended from one of
Eskom’s pylons. To reach the point where the accident occurred, Jacques had to climb to a
height of 14 m above the ground after first going through an anti-climbing device. Apparently
Jacques, his younger brother, and a younger friend were taking their dog for a walk when
they challenged each other to see who could climb the highest. As Jacques was climbing, he
stopped to take a rest and the glass insulators, which he described as greenish glass
saucers, caught his eye. Out of curiosity, he climbed further out onto the cross-arm of the
pylon to feel their texture. According to an expert’s evidence, Jacques’ head came too close
to the power line that was suspended from the cross-arm immediately above him. The strong
voltage caused the current to jump across the space between the power line and Jacques’
head. The shock, from 66 000 V, flung the boy from his perch. When Hendricks claimed
damages, Eskom denied liability and also claimed contributory negligence on Jacques’ part.
In assessing whether Jacques was indeed culpae capax at the time of the incident, the
Court referred to Weber v Santam Versekeringsmaatskappy Bpk 9 and reiterated that it
needed to recognise the inherent weakness in children to act irrationally and impulsively due
to their tender age. Therefore, although children can distinguish between right and wrong,
they will often not be able to act in accordance with that appreciation, because they can
become so engrossed in their play that they become oblivious of other considerations. The
Supreme Court of Appeal had no doubt that Jacques had appreciated that if he climbed past
the anti-climbing device he ran the risk of falling and hurting himself. However, while he was
climbing, the glass insulators, which had nothing to do with why Jacques climbed up the
pylon, attracted his attention. His fascination by these glass saucers prompted him to climb
towards them so that he could touch them. The Court said that Jacques’ emotional and
intellectual maturity had to be assessed according to this particular conduct. On the
evidence, the Court concluded that Jacques’ conduct was typical of the impulsive behaviour
in which children sometimes engage, and that this was an instance where the warning
against ‘placing an old head on young shoulders’ should be heeded. The Court held that the
conduct in question (that is, touching the insulators out of curiosity) was indicative of
Jacques’ inability to act in accordance with any appreciation he may have had of the danger
involved. Accordingly, the Court held that Eskom had failed to rebut the presumption that
 Jacques was culpae incapax at the time of the accident, and so he had not been contributory
negligent.

8.2.2 Mental disease or illness and emotional distress

A wrongdoer is culpae incapax if, at the time of the alleged delict, he or she suffers from any
mental illness or disease, or emotional distress that renders him or her incapable of distinguishing
between right and wrong, or of acting in accordance with an understanding of the distinction
between right and wrong. In S v Campher 10 the Appellate Division confirmed that not only must a
person have failed to control the impulse to commit the act in issue, but the lack of control must
have arisen from an ‘infirmity of will for which he was not answerable’.
S v Campher11
The accused was charged with murdering her husband. Their marriage had been unhappy
and the deceased had been emotionally and physically abusive towards the accused and
their children. According to the evidence, the constant physical and emotional abuse that the
accused suffered, together with the fact that their children were also emotionally abused, had
driven the accused to breaking point. On the morning the accused shot and killed her
husband, there had been yet another occasion of abuse, and she grabbed a revolver to
defend herself against the deceased who was threatening her with a screwdriver. She did not
shoot the deceased then, as he had grabbed her and dragged her outside where he
submitted her to further emotional abuse. The Court accepted the accused’s testimony that
she had felt as if she had descended into an emotional dark pit and all she could think of was
to get out of the situation in which she had found herself. The accused explained that she had
realised only after it had happened that she had shot the deceased. The Court accordingly
found that, at the critical moment, the accused had suffered from a severe, yet temporary,
mental disturbance. Even though the accused did not suffer from any mental disease, she
was found not guilty.

8.2.3 Intoxication
Intoxication, whether while under the influence of alcohol or drugs, may render a person culpae
incapax. However, this is not true in all cases. For example, if a person takes an intoxicating
substance before committing the delict, when still accountable, he or she may be liable for the prior
act. The basis for liability here is similar to the situation where an epileptic person, who knows that
fits can suddenly occur, still drives a vehicle and causes an accident while having a fit. Even though
at the moment of the accident the person is not accountable for the harmful result, liability still
arises. Therefore, although a defendant may not have been able to appreciate the harmful nature of
the conduct at the time the harm was inflicted, a court might still hold the defendant liable.12
S v Chretien13
After a party, and while under the influence of alcohol, Chretien drove a Volkswagen minibus
into a crowd of people standing on a pavement. One person died and five were injured. The
Court could not find beyond reasonable doubt that Chretien had deliberately driven into the
crowd. He claimed that he had thought that the people would disperse when they saw the
minibus approaching. The Court a quo found Chretien not guilty of attempted murder and of
common assault, since he lacked the necessary intention, and therefore the required
elements of the crimes were not satisfied. The State appealed against the finding and
submitted that Chretien should at least be found guilty of common assault.
The Appellate Division identified two opposite ‘poles’ for describing degrees of intoxication.
One extreme is when a person is slightly intoxicated and there is some altered behaviour. In
this case, the intoxication would not be such that the person could not control himself or
herself, and so would still be accountable. The other extreme is when a person is so
intoxicated that he or she passes out and the only movements made are due to involuntary
 muscle spasms, in which event such involuntary movements are not even categorised as
conduct. Such a person is also culpae incapax. Between these two poles, there are various
degrees of intoxication. Some people can be intoxicated to such an extent that others might
describe them as ‘very drunk’, and yet they seem to conduct themselves in a seemingly
rational manner. In such an instance, the intoxicated person would be regarded as
accountable. Others might be ‘dead drunk’, and unable to know what they are doing. In such
instances, there is no intention and the question of accountability depends upon the evidence
and the circumstances of the case. The Court found that the Court a quo was correct in
finding that intention was lacking and accordingly dismissed the appeal.
Parliament subsequently enacted legislation that provides for a statutory crime where
persons commit a crime while in an intoxicated state. However, this legislation does not
affect the law of delict, and common-law principles enunciated in the Chretien judgment are,
therefore, still applicable in this area of law.

8.2.4 Provocation
Provocation may, in some instances, exclude the element of wrongfulness.14 However, in
circumstances where it does not do so, it may, depending upon its effect on the defendant’s
behaviour, exclude either accountability or fault in the form of intention.15 A person can be
provoked to such an extent that the person loses control of his or her ability to act responsibly or, if
the provocation is not that extreme, to the extent that a person lacks consciousness of the
wrongfulness of his or her actions. In effect, the rage reaction creates a moment of temporary
insanity not unlike that which occurred in S v Campher,16 noted previously. The basis for this
defence is that loss of temper due to provocation may render the provoked person culpae incapax at
the time of inflicting harm.

PAUSE FOR The nature of the provoking conduct

REFLECTION In Bennet v Minister of Police17 the Court held that verbal prvocation
cannot justify a physicl assault in ‘retaliation’, and so does not exclude
wrongfulness. For the purposes of excluding accountability or intention,
would the nature of the provoking conduct similarly be relevant? What
considerations might point towards not recognising provocation as a
reason for excluding accountability?
Figure 8.1Intention

8.3 Intention
A person will be at fault when he or she intends to cause another person harm, knowing that it is
wrong to do so. So, when a court concludes that a defendant had intention, it demonstrates the
law’s disapproval of a defendant’s reprehensible state of mind. The enquiry into intention is
subjective, in that courts have to determine what the defendant actually had in mind at the time of
committing the delict.
There are principally three forms of intention:
• Dolus directus
• Dolus indirectus
• Dolus eventualis.

Irrespective of which form of intention is present, the concept has two aspects: direction of will,
and consciousness of wrongfulness.18 So, in law a person acts intentionally only when that person’s
will is directed at a specific outcome, knowing that what he or she is doing is wrongful. If either
one of the two components is absent, or cannot be proved, there is no intention.
Figure 8.2Intention

8.3.1 Dolus directus or direct intention

Direct intention exists where the wrongdoer’s primary aim is to achieve a particular consequence,
for example, if one person deliberately says something in order to hurt another person’s feelings. It
is irrelevant whether the wrongdoer simply foresaw the desired outcome as a possibility or as a
certainty. The fact that the wrongdoer desired a certain consequence and acted in a way to realise
that consequence is enough to establish direct intention. It also does not matter whether the result is
achieved in the exact manner foreseen or contemplated by the wrongdoer.

8.3.2 Dolus indirectus or indirect intention

Indirect intention exists where a person has direct intention in terms of a certain consequence
(consequence A). However, in effecting consequence A, the person foresees that another harmful
consequence (consequence B) will inevitably also take place if consequence A is to be realised. For
example, Sam wishes to steal a car radio from a locked car, and he realises and accepts that he will
also have to break a window or damage the car in some other way to get to the radio. The damage
to the car is not his main objective, but it is an inevitable consequence if he wants to execute his
main objective, that is, to steal the car radio. The owner of the car will thus be able to sue Sam for
damage to the car by relying on dolus indirectus as form of fault.

8.3.3 Dolus eventualis or intention by acceptance of foreseen result

Dolus eventualis exists where, in executing a plan to cause harm, a person foresees a wrongful
consequence that is not desired, but nevertheless reconciles himself or herself with the possibility
that it might arise and continues to execute the plan to cause harm. For example, Sam aims a large
stone at Bheki, who is standing with two friends. Sam foresees that he might hit one of the friends
instead, but he nevertheless proceeds to throw the stone at Bheki. He misses, and the stone hits and
injures one of the friends. Dolus eventualis is present, since Sam subjectively foresaw that he might
injure someone else and reconciled himself with the fact that one of Bheki’s friends might get hurt.
Dolus eventualis entails a two-part enquiry:
1. Did the wrongdoer (Sam) subjectively foresee or realise that the harmful consequence might
ensue?
2. Did Sam reconcile himself with that realisation by nevertheless continuing with his actions?
If Sam thought that he might hit one of the friends, but somehow genuinely believed that it would
not happen, he would not have acted with dolus eventualis because, subjectively, he had not
reconciled himself with that consequence. The form of fault that is present in this latter instance is
luxuria or conscious negligence.19 Also, if Sam did not at all foresee that he might hit someone else,
there can be no intention, and negligence would be the appropriate form of fault.
S v Humphreys20
Humphreys had been convicted in the High Court on ten counts of murder and four counts of
attempted murder, resulting from a collision caused when the minibus he was driving collided
with a train. He had entered a level crossing while the boom controls were down and the
warning signals flashing. The Supreme Court of Appeal agreed with the High Court that
Humphreys had subjectively foreseen the possibility of fatal injuries occurring as a result of
his reckless actions. However, the Court disagreed in respect of the second part of the
enquiry, finding that Humphreys had not reconciled himself to the possibility that his reckless
actions could lead to the death of his passengers. If Humphreys had reconciled himself to the
possibility of the death of his passengers, this would have meant he had reconciled himself to
his own possible death. The evidence did not support this inference, as there was no
evidence he did not value his own life. Instead, the evidence showed that Humphreys had
cleared the level crossing in similar situations in the past and had thought he would be able to
do the same on this occasion, and that the foreseen risk would not materialise. Accordingly,
since Humphreys had not reconciled himself to the possibility of death or fatal injury, dolus
eventualis was not established.

Director of Public Prosecutions, Gauteng v Pistorius21

Pistorius had shot and killed his girlfriend Reeva through a toilet cubicle door when he
allegedly mistook her for a thief. The Court a quo found Pistorius guilty of culpable homicide
and held that he could not be convicted of murder, as dolus had not been proved. On appeal,
the Supreme Court of Appeal considered especially whether dolus in the form of dolus
eventualis had been present, for if it had been, Pistorius should have been found guilty of
murder instead of culpable homicide. It noted:22

In contrast to dolus directus, in a case of murder where the object and purpose
of the perpetrator is specifically to cause death, a person’s intention in the form
of dolus eventualis arises if the perpetrator foresees the risk of death occurring,
but nevertheless continues to act appreciating that death might well occur,
therefore ‘gambling’ as it were with the life of the person against whom the act is
directed. It therefore consists of two parts: (1) foresight of the possibility of
death occurring, and (2) reconciliation with that foreseen possibility. This
second element has been expressed in various ways. For example, it has been
said that the person must act ‘reckless as to the consequences’ (a phrase that
has caused some confusion as some have interpreted it to mean with gross
negligence) or must have been ‘reconciled’ with the foreseeable outcome.
Terminology aside, it is necessary to stress that the wrongdoer does not have to
foresee death as a probable consequence of his or her actions. It is sufficient
that the possibility of death is foreseen which, coupled with a disregard of that
consequence, is sufficient to constitute the necessary criminal intent.

Pistorius sought to deny the presence of dolus eventualis by insisting he did not have the
necessary intention to kill Reeva, as he had believed a thief, not Reeva, had been concealed
in the toilet. In this regard the Court said:23

… although a perpetrator’s intention to kill must relate to the person killed, this
does not mean that a perpetrator must know or appreciate the identity of the
victim. A person who causes a bomb to explode in a crowded place will probably
 be ignorant of the identity of his or her victims, but will nevertheless have the
intention to kill those who might die in the resultant explosion.

The Court went on to say: 24

What was in issue, therefore, was not whether the accused had foreseen that
Reeva might be in the cubicle when he fired the fatal shots at the toilet door but
whether there was a person behind the door who might possibly be killed by his
actions.

PAUSE FOR Summary of intention

• With direct intention, the wrongdoer desires a certain wrongful consequence
REFLECTION
• With indirect intention, the wrongful consequence is an inevitable wrongful
consequence while attempting to realise a desired consequence
• With dolus eventualis, the wrongful consequence is brought about by reckless
disregard for the possibility of that consequence ensuing.

8.3.4 First component of intention: Direction of will

Direction of will, the first component of intention, requires that a person must have aimed to
achieve a certain result, or at least must have been willing to produce or accept the consequences
that might result. This involves determining the mental predisposition of the person obtaining a
specific result.25 There are three ways in which a person can direct his or her will:

• Directly (resulting in dolus directus)

• Indirectly (resulting in dolus indirectus)
• By accepting the possibility of other harmful consequences ensuing (resulting in dolus
eventualis).

8.3.5 Second component of intention: Consciousness of wrongfulness

Consciousness of wrongfulness was first recognised in defamation cases, but courts now accept it
generally as a component of intention.26 The second component requires that when a person directs
his or her will towards achieving a desired consequence, that person must know that the conduct
and the ensuing consequences are contrary to law and the legal convictions of society. Therefore, if
someone genuinely believes that he or she is acting in accordance with the law, he or she has not
acted intentionally for purposes of the law.27 For example, where a person believes that his or her
conduct is justified, even if it is not, that person has not acted intentionally because he or she
intended to behave lawfully, and not wrongfully. This example also illustrates how the subjective
nature of intention comes to the fore when we assess the knowledge of the wrongdoer.
Maisel v Van Naeren28
Maisel had fallen into arrears with his rent and he was also allegedly a nuisance to his fellow
tenants. The owner of the block of flats (Van Naeren) wrote to him on these issues and sent a
copy of the letter to the chairperson of the Rent Board in the bona fide, but mistaken, belief
that the block of flats and the issue at hand fell within the ambit of the Rental Housing Act 50
of 1999, which required that he send such copy to the chairperson.
Because he sent the defamatory letter to the chairperson of the Rent Board, it amounted to
publication of defamatory information. Maisel claimed damages under the actio iniuriarum for
defamation, but the Court held Van Naeren not liable for lack of intention, because he was
unaware that his conduct was wrongful. Van Naeren had genuinely believed that sending the
letter to the chairperson of the Rent Board was the lawful thing to do. Therefore, although he
 had directed his will, in that he had desired to send the letter to the chairperson of the Rent
Board and had done so, he had been unaware he was acting wrongfully. There was
accordingly no animus iniuriandi on Van Naeren’s part.

In Le Roux v Dey 29 Harms DP analysed the background and implications of the requirement of
consciousness of wrongfulness (‘coloured intent’), and decided that intention for the purposes of
the actio iniuriarum does not require consciousness of wrongfulness. However, on appeal the
Constitutional Court decided that, since it had been unnecessary for the Supreme Court of Appeal
to decide this issue, it was also unnecessary for the Constitutional Court to do so.30

8.3.6 Special cases: Intention comprising direction of will only

Under the actio iniuriarum there are specific forms of iniuria – wrongful deprivation of liberty,
wrongful attachment of goods, and possibly seduction – where policy dictates that the usual rules
for establishing intention should be modified, and that intention should comprise one aspect only:
the direction of will. A plaintiff may, therefore, establish intention even though the defendant had
no knowledge of the wrongfulness of his or her actions. So, in these cases it is not full animus
iniuriandi that is required, but an attenuated version. The resultant effect is that a defendant is
unable to exclude fault by pleading a defence aimed at negating consciousness of wrongfulness,
such as mistake. For example, police officers who incorrectly believe that valid warrants of arrest
exist and then unlawfully arrest persons will be found to have intended to effect the arrests, even
though, subjectively, they had thought that they were acting lawfully.31 Similarly, courts say a
man’s ignorance of a girl’s virginity will not suffice as a defence to exclude intention in seduction
cases.32
Minister of Justice v Hofmeyr33
Hofmeyr claimed damages under actio iniuriarum for unlawful detention. He had been kept in
solitary confinement during his imprisonment of five months, except for two brief periods of
about eight days each. The Minister relied on statutory justification as defence. The Prison
Emergency Regulations provided that detainees could be ‘segregated’ from other detainees
when the head of the prison deemed it ‘practicable’. However, the head of the prison had
misunderstood the nature of the discretion conferred on him, particularly in terms of the
meaning of ‘segregation’ in the context of the regulations.
In deciding whether the defence of statutory justification could succeed, the Court held that
the head of the prison had erroneously believed the segregation provided for in the
regulations had a punitive purpose. There had been no basis for this belief, since the
regulation had been enacted primarily in the interests of the detainees themselves.
Therefore, the detention had been wrongful.
It was nonetheless argued that there had been no animus iniuriandi, since consciousness
of wrongfulness had been absent. The Court acknowledged that consciousness of
wrongfulness had been absent, but noted that there are certain instances, particularly those
that involve wrongful deprivation of liberty, where consciousness of wrongfulness is not
required. Although not based on Roman and Roman-Dutch principles, legal policy required
that these instances be recognised according to what is sensible and just.

It is clear that without dolus the action for an iniuria would lie neither in Roman
law nor in Roman-Dutch law. … It is equally clear, however, that in a limited
class of iniuriae the current of precedent has in modern times flowed strongly in
a different direction. In this limited class of delicts dolus remains an ingredient
of the cause of action, but in a somewhat attenuated form, in the sense that it is
no longer necessary for the plaintiff to establish consciousness [of
wrongfulness] on the part of the wrongdoer of the wrongful character of his act.
 Included in this limited class are cases involving false imprisonment and the
wrongful attachment of goods.34

Direction of will alone, without consciousness of wrongfulness, was therefore sufficient

indication of intention and the Minister of Justice was accordingly held vicariously liable for
the improper and unlawful manner in which Hofmeyr was treated while detained.

It is uncertain whether the attenuated form of intention also applies to the Aquilian action,35 but the
Appellate Division has left open the possibility of this doctrinal modification, should policy
considerations require that a limited form of intention be recognised in a particular instance.36

8.3.7 Difference between motive and intention

The term ‘intention’ is a technical, legal expression that describes a person’s reprehensible state of
mind. ‘Motive’, on the other hand, is the reason that triggered the formation of intention, or the
‘actuating impulse preceding intention’. 37 Motive represents the mental aspect that leads to the
formation of intention, which is why courts often use it to prove intention. For example, Thandi,
who has substantial gambling debts, stands to inherit a great deal of money from her uncle. She
decides to kill her uncle so that she can pay off the debts with her inheritance. Thandi’s motive, or
reason for committing the crime, is to acquire the inheritance to pay off her debts, while her
intention is to kill her uncle.
Motive may also indicate whether consciousness of wrongfulness is present, in that a good
motive may sometimes indicate a lack of consciousness of wrongfulness, while a malicious motive
might indicate the intention to do harm. However, a good motive will not save a person who knows
that what he or she is doing is wrongful. For example, a doctor who was treating his dying father
decided to relieve his father’s pain and suffering by giving him a lethal injection. The doctor knew
that the deliberate hastening of death was wrongful and so, even though he had a good motive, he
still had the intention to commit murder.38

8.3.8 Proving intention

The plaintiff bears the onus of proving the defendant’s intention.39 It is obviously unlikely that
anyone other than the defendant would actually know what the defendant subjectively thought at
the time of the incident. So, courts often draw the conclusion by means of inferential reasoning –
by looking at the nature of the alleged delict and all the surrounding circumstances of the case, and
concluding that, based on the facts, the only reasonable conclusion that can be reached is that the
defendant must have had a reprehensible state of mind. A general rule when drawing inferences is
that a person intends all the necessary consequences of his or her conduct.
Under the actio iniuriarum, for example, in defamation cases, the plaintiff is assisted by a
rebuttable presumption of animus iniuriandi. This usually means that, once the plaintiff has shown
that a personality right has been infringed, a court can assume that intention is present, unless the
defendant can show otherwise. So, the onus of proof shifts to the defendant, who must then provide
sufficient evidence to rebut the presumption.

PAUSE FOR Inference

REFLECTION An inference is not a presumption. An inference is a final conclusion
drawn from the facts. A presumption is a device that allows courts to
draw a preliminary conclusion, and which then shifts the onus of proof
from one party to another. So, the notion that a person intends all the
necessary consequences of his or her conduct is not a presumption, but
 a general expression of the process according to which the inference of
intention is drawn.

However, the law treats different iniuriae differently regarding the presumption of intention. For
example, when plaintiffs prove infringement of their reputation or fama, two rebuttable
presumptions arise: that both wrongfulness and intention are presumed.40 However, with other
iniuriae, only a rebuttable presumption of intention arises when plaintiffs prove infringement of
their personality aspects.41

PAUSE FOR Should all iniuriaebe treated the same?

REFLECTION Van der Walt and Midgley42 argue that courts should reconcile the
different approaches and treat all iniuriae similarly when it comes to the
presumptions of wrongfulness and intention. In all instances, plaintiffs
should prove, on a balance of probabilities, that their personality interests
have been infringed. Once they have proved this, two rebuttable
presumptions, of wrongfulness and intention, should arise, which the
defendants would have to rebut. The reason the authors give is that the
different ways of handling iniuriae could create problems when claims are
made in the alternative, or where more than one personality aspect has
been infringed simultaneously (for example, infringement of both dignitas
and fama). Why would there be problems in these instances?

8.4 Defences that exclude intention

Once a plaintiff has proved intention, the defendant (who a court has found to be accountable) can
raise various defences to escape liability. Defences that exclude intention can be aimed at any one
or both of the elements for intention. In other words, the defences can be used to:
• Indicate that the defendant did not direct his or her will towards effecting the harm-causing
event
• Indicate that the defendant did not know that his or her conduct was wrongful
• Indicate that the defendant neither directed his or her will towards effecting the harm-causing
event nor knew that the conduct was wrongful.

8.4.1 Mistake
Intention involves subjectively evaluating the defendant’s state of mind when the delict was
committed. Therefore, a bona fide mistake regarding any aspect of the case, including the
lawfulness of a person’s actions, will exclude intention.43 The defence of mistake often turns on the
absence of the second element of intention, the consciousness of wrongfulness. However, mistake
may also negate the direction of a person’s will – for example, accidentally pressing a button that
sends a defamatory email message. In principle, it does not matter whether the mistake is one of
fact or one of law, or whether it is reasonable or unreasonable. The subjective nature of the test for
intention simply requires that in the mind of the wrongdoer there was genuine belief that the
conduct at issue was indeed lawful. However, when media defendants are sued for defamation or
infringements of privacy, only reasonable mistakes can rebut the presumption of intention.44
Maisel v Van Naeren45
 Van Naeren sent a copy of a letter to the chairperson of the Rent Board in the bona fide, but
mistaken, belief that the Rental Housing Act was applicable. Although he had directed his will
in sending the letter, he lacked consciousness of the wrongfulness of his conduct, in that he
had genuinely believed sending the letter to the chairperson of the Rent Board was the lawful
thing to do. There was accordingly no animus iniuriandi.

8.4.2 Jest
The basis of jest as a defence is usually that the defendant’s will was not directed at achieving the
particular consequence that ensued, but it may also exclude consciousness of wrongfulness.
However, the defendant must be able to show that the conduct was bona fide and genuinely meant
as a joke.
Masch v Leask46
Leask, an auctioneer, told participants in an auction that he would be holding other sales, the
details of which appeared in yellow flyers he had distributed among them. Another
auctioneer, Masch, who was standing in the crowd, shouted to Leask that he was lying and
that the yellow flyers did not mean anything. Leask sued Masch for defamation, in that Masch
had publicly made Leask out to be a liar and an untrustworthy person. Masch raised the
defence of jest, claiming that he had simply been joking.
The Court found that Masch had not proven that he had acted in jest, or that the
bystanders had perceived his words as a joke. On the authority of this case, the defence of
jest can only succeed if the words spoken were genuinely meant as a joke and if others also
understood them to be a joke.
How compatible is the requirement that others must have understood the comment to be a
joke with the subjective nature of intention? Does this introduce an objective aspect, similar
to that which is now required when a media defendant makes a mistake?

8.4.3 Intoxication
In exceptional instances, people may be so intoxicated that they cannot develop an intention.
Intoxication is generally used as a ground to exclude accountability. However, a person can also
use it to exclude either one of the two elements of intention in cases where a court finds a defendant
accountable. In Chretien’s case47 the Court found that when persons are so drunk that they cannot
properly direct their will, or they do not realise that their conduct is wrongful, there is no intention
on their part. Of course, it would still be possible to find that they have acted negligently and base
liability on negligence as a form of fault.

PAUSE FOR Should voluntary intoxication be a defence?

REFLECTION Could one argue that there are policy considerations that would cause
the intoxication defence to fail when people voluntarily create the state of
intoxication? What would these policy considerations be?

8.4.4 Provocation
A person can raise provocation as a defence to exclude intention. Although a provoked person may
still have directed his or her will to injure the person doing the provoking, courts can exclude
intention where the provoked person did not realise that his or her conduct was wrongful.

8.4.5 Emotional distress

In certain instances, people can suffer from such severe emotional distress that they cannot develop
an intention. While a person can generally use emotional distress as a ground to exclude
accountability, a person can also use it to exclude either one of the two elements of intention in
cases where a court finds a defendant accountable.

PAUSE FOR Excluding accountability and intention

REFLECTION Emotional distress, provocation and intoxication are defences that can
exclude both accountability and intention. How are the requirements for
each defence applied differently to exclude accountability on one hand
and intention on the other?
Provocation can be used as a defence to exclude wrongfulness, but
the requirements of such a defence are different. What are these
differences?

8.5 Negligence
Unlike intention as a form of fault, negligence is not concerned with the law’s disapproval of a
defendant’s state of mind. With negligence, liability is based on the law’s disapproval of a
defendant’s conduct. The following sections discuss the concept of negligence, the test for
negligence and the various circumstances and attributes that play a role in establishing the standard
of care that the law expects of persons.

8.5.1 The concept of negligence

An enquiry into negligence involves evaluating a defendant’s conduct according to a standard that
is acceptable to society. This standard is expressed with reference to a fictitious ‘reasonable person’
that represents society’s expectations of adequate and reasonable conduct. It represents an objective
standard that all legal subjects must adhere to by paying sufficient attention to ensure that their
conduct is in line with the standard of care that society expects. The test for negligence differs from
the test for wrongfulness, in that the reasonable person is placed in the position of the defendant.
Therefore, we do not evaluate all the circumstances (both known and unknown to the defendant) of
the harm-causing conduct, as is done in the wrongfulness enquiry. To determine whether a
defendant was negligent, we simply ask what the reasonable person, put in the same position as the
defendant, would have done.
If a defendant’s conduct does not conform to the standard of a reasonable person, the conduct
is blameworthy in law and the defendant will be considered to be at fault.

8.5.2 Characteristics of a reasonable person

The reasonable person is a fictitious or abstract concept that expresses the standard according to
which one measures the reasonableness of a defendant’s conduct.48 Therefore, the reasonable-
person standard requires an adequate and consistent level of care on the part of all legal subjects,
and it must also be sensitive to a society where people have various skills and levels of intellect,
and are of different ages. A reasonable person, therefore, does not represent a standard of
exceptional skill, giftedness or care, but it also does not represent a standard of underdeveloped
skills, recklessness or thoughtlessness. It is a standard of the ordinary individual who takes
reasonable chances and reasonable precautions to protect interests, while expecting the same
conduct from others.49
 The reasonable-person criterion is an expression of what society expects of its members in
their everyday life. The criterion, and the behaviour expected from a reasonable person, will change
according to the changes in society’s expectations. For example, where crime and threats to
personal safety are more prominent, a reasonable person might act with more nervousness and
fright than a reasonable person would where such threats are not prominent. Leaving a security gate
unlocked would probably be regarded as highly negligent in a society with high crime levels, but as
reasonable conduct in a society in which housebreaking was a very rare occurrence.
A point to remember is that the standard is not that the harm must have been avoided at all
costs and that no harm must have ensued. Rather, reasonable conduct means that a person must
have acted appropriately in the circumstances, and behaved in the same way that a reasonable
person would have behaved in the same circumstances. Should harm arise despite a person’s
reasonable behaviour, that fact does not affect the standard. The behaviour remains reasonable and
that person would not be at fault.

8.5.3 The test for negligence

In Sea Harvest Corporation (Pty) Ltd v Duncan Dock Cold Storage (Pty) Ltd 50 the Supreme Court
of Appeal reiterated that the benchmark for negligence is what a reasonable person would have
done in the same circumstances as the defendant. Courts have developed tests, and the most often
quoted is the one set out in Kruger v Coetzee.51 However, these tests are still no more than
guidelines or approaches for assessing how a reasonable person would have acted in the
circumstances. Any conclusion reached after applying one of these tests should articulate the
benchmark standard of reasonable behaviour.
Kruger v Coetzee52
Coetzee’s car was damaged in a collision with Kruger’s horse, which had strayed from
Kruger’s property through an open gate and onto a public road. The local divisional council
had put in the gate to give construction workers access to a construction site via Kruger’s
property. Drivers of the construction vehicles frequently left the gate open. Kruger had lodged
complaints about such conduct with both the divisional council and the building contractors,
but continued to keep his horses on that land.
Coetzee alleged, firstly, that Kruger had negligently allowed his horses to stray onto a
public road unattended, and, secondly, that in leaving the gate open Kruger had been unable
to exercise control of the horses. The Court articulated the test for negligence as follows:53

For the purposes of liability, culpa arises if:

(a) A diligens paterfamilias in the position of the defendant:
(i) Would foresee the reasonable possibility of his conduct injuring another in his person or
property and causing him patrimonial loss; and
(ii) Would take reasonable steps to guard against such occurrence; and
(b) The defendant failed to take such steps.

The Court noted that requirement (a)(ii) is often overlooked, and that there must be an
indication as to what reasonable steps a diligens paterfamilias in the position of the defendant
would have taken. Assessing such steps depends on the circumstances of each case, and
there is no generally applicable basis for constituting reasonable precautions for all situations.
In this instance, it was common cause that a reasonable person in Kruger’s position would
have foreseen the possibility of his horses straying through the open gate onto the main
road, and causing damage to motor cars that might collide with them. In fact, Kruger
conceded that he was aware of the possibility. Kruger further conceded that he had been
aware that the gate was often left open. Yet, he continued to keep his horses on the
property. The Court held that in such circumstances, a reasonable person would not have
‘shrugged his shoulders in unconcern’, and if there were reasonable steps that could have
 been taken, a reasonable person would have taken them. The defendant did in fact take
some steps to prevent the gate from being left open. He went to the divisional council twice
to complain about its employees leaving the gate open. He also complained to the
construction managers. So, the question that remained was whether there were any other
steps that Kruger should have taken. The Court found that insufficient evidence was
presented of any other reasonable measures that Kruger should have taken, and accordingly
found that Coetzee had not proved that Kruger had been negligent.

The test articulated in Kruger v Coetzee 54 points to four important issues that one must assess when
determining whether the defendant’s conduct was reasonable or unreasonable (and therefore
blameworthy for purposes of the law):
1. The first is to place a reasonable person in the same position as the defendant.
2. The second is to evaluate the situation and circumstances to see whether a reasonable person
in the defendant’s position would have foreseen the possibility of harm arising from the
conduct. If a reasonable person would have foreseen that the relevant conduct might cause
harm, then one can move on to the next issue.
3. The third issue raises the question of whether a reasonable person would have done anything
to prevent the harm from occurring if the conduct continued. To answer this question, one must
assess what steps were available to the defendant in the particular circumstances. This is done
by assessing the availability of alternative steps that would have prevented harm, and whether
they were reasonable and practical in the circumstances. If the defendant did in fact take some
measures to prevent the harm, the plaintiff must show that such measures were either
unreasonable or inadequate, with reference to what a reasonable person would have done in the
circumstances. In Kruger v Coetzee 55 the Court noted that one must first determine what steps
were available before one can assess whether a reasonable person would have taken any other
steps.
4. In the fourth instance, one compares the defendant’s conduct to the course of action that the
court thinks a reasonable person would have taken in the circumstances. If it appears that the
defendant did nothing, or did less than what a reasonable person would have, the defendant’s
conduct was ‘sub-standard’ and unreasonable, and therefore negligent.

PAUSE FOR Foreseeing harmful consequences

REFLECTION When assessing whether a reasonable person would have foreseen the
harmful consequences, one should avoid applying the objective
reasonableness criterion that is used for determining wrongfulness. The
Supreme Court of Appeal addressed this problem in Minister of Safety
and Security v Carmichele:56

In considering this question [what was reasonably

foreseenable], one must guard against what Williamson JA
called ‘the insidious subconscious influence of ex post
facto knowledge’… Negligence is not established by
showing merely that the occurrence happened (unless the
case is one where res ipsa loquitur)57 or by showing after it
happened how it could have been prevented. The diligens
paterfamilias does not have a ‘prophetic foresight’. … ‘After
the event, even a fool is wise. But it is not the hindsight of a
fool; it is the foresight of the reasonable man which alone
can determine responsibility’.
8.5.3.1 Foreseeability and preventability
The test for negligence stands on two pillars:
1.The foreseeability of harm
2.The preventability of harm.

It is not adequate simply to state that the defendant was negligent. There must be a concrete and
practical argument as to why and how the defendant was negligent in the circumstances. So, the
application of both the foreseeability test and the preventability test is the foundation for practically
evaluating the defendant’s conduct.

8.5.3.2 Foreseeability of harm

One cannot establish negligence unless one can prove that the harm arising from the defendant’s
conduct was reasonably foreseeable. This is assessed after considering the surrounding (or
objective) circumstances, as well as the qualities the law attributes to a reasonable person.
Therefore, reasonable foreseeability is a flexible concept because it depends on the circumstances
of each case, and a juridical evaluation of the risk in those particular circumstances. There are no
hard-and-fast rules to help us determine whether harm is reasonably foreseeable, but courts have
identified some broad guidelines that should be applied flexibly.

Lomagundi Sheetmetal and Engineering (Pvt) Ltd v Basson58

Basson contracted Lomagundi to build a roof on top of a silo. Basson’s bales of stover were
stacked up against the silo and, while Lomagundi’s employees were busy welding, some
sparks ignited the bales. When Basson sued Lomagundi, the Court had to determine the
issue of negligence. The Court identified the following broad and flexible guidelines for
evaluating whether harm was reasonably foreseeable:
• How real is the risk of the harm eventuating?
• If the harm does eventuate, what is the extent of the damage likely to be?
• What are the costs or difficulties involved in guarding against the risk eventuating?

From the evidence, it appeared that the risk of the stover igniting was not great, but it was
nonetheless a real possibility. The damage that a fire might cause was, in the circumstances,
quite extensive. Furthermore, the cost or difficulty involved in preventing the risk from
eventuating was very slight. All that was required was to move the bales a small distance
back from the silo and to sweep the space between the bales and the silo to remove
flammable material.
The Court found that any ‘prudent man’, before commencing the welding operations, would
have foreseen the risk of fire and would have moved the bales and cleared the space
between the bales and the silo to reduce the risk of the flammable material catching fire. The
real possibility of the bales igniting from welding sparks, weighed against the possible extent
of the damage and the relative ease with which the ordeal could have been prevented,
meant that the Court found that there was indeed negligence on the part of Lomagundi’s
employees.
The questions that one has to answer are:
• Was it, in the specific circumstances, reasonably foreseeable that harm to others would
occur as a consequence of the defendant’s chosen conduct?
• Having due regard to the specific circumstances of the case, what would the general nature
of the harm be and how would it generally occur?

There are, in principle, two diverging views to the application of the foreseeability test: the abstract
or absolute approach, and the concrete or relative approach. In the abstract approach, the question
of foreseeability depends only on whether the defendant foresaw that, in general, harm could occur.
The exact type of harm, and the extent of that harm, is not relevant in this approach. In the relative
approach, we can only regard a defendant’s conduct as negligent if the specific harmful
consequences were reasonably foreseeable.59 So, we can only say that the defendant acted
negligently if a reasonable person in the defendant’s position would have foreseen the specific
harm that the plaintiff is claiming for.
In terms of the abstract or absolute approach, legal causation is the element that is used to
determine whether defendants are liable for the specific harmful consequences that occurred as a
result of their conduct. Foreseeability of harm becomes a factor that could indicate whether or not
the harm is too remote. However, when the relative approach is followed, the application of the test
of foreseeability of the specific harmful consequences as part of the test for negligence renders the
legal causation enquiry superfluous.

PAUSE FOR Abstract or absolute approach

REFLECTION Is harm, that is, any harmful consequence(s), foreseeable?
See Herschel v Mrupe 1954 (3) SA 464 (A); Kruger v Coetzee;
minority judgment in Sea Harvest Corporation (Pty) Ltd v Duncan Dock
Cold Storage (Pty) Ltd 2000 (1) SA 827 (SCA).

Concrete or relative approach

Are the general nature and the general manner of occurrence of the
consequences that actually eventuated reasonably foreseeable?
See Ablort-Morgan v Whyte Bank Farms (Pty) Ltd 1988 (3) SA 531
(E); Mukheiber v Raath 1999 (3) SA 1065 (SCA); majority judgments in
Sea Harvest Corporation (Pty) Ltd v Duncan Dock Cold Storage (Pty) Ltd
2000 (1) SA 827 (SCA); Premier of the Western Cape Province v Loots
2011 JDR 0250 (SCA) at 13.

It appears that our law favours the relative approach, but in a hybrid form where legal causation
remains a tool for limiting liability, in addition to the negligence enquiry.60 Boberg 61 explains this
preference by pointing out that an enquiry into how a reasonable person would have acted in a
meaningful manner can be conducted only by referring to the consequences that were indeed
reasonably foreseeable. According to Boberg, if only ‘harm in general’ is foreseen, there is no
sensible way in which we can assess what a reasonable person’s course of action would have been.
In Premier of the Western Cape Province v Loots NO 62 the Court accepted the relative approach to
negligence:
According to this approach it cannot be said that someone acted negligently because harm to
others in general was reasonably foreseeable. A person’s conduct can only be described as
negligent with reference to specific consequences. Yet, the relative approach does not
require that the precise nature and extent of the actual harm which occurred was
reasonably foreseeable. Nor does it require reasonable foreseeability of the exact manner in
which the harm actually occurred. What it requires is that the general nature of the harm
that occurred and the general manner in which it occurred was reasonably foreseeable. At
some earlier stage there was a debate as to whether our courts should follow the relative
approach as opposed to the so-called abstract or absolute approach to negligence. But it now
appears to be widely accepted by academic writers, on good authority, that our courts have
adopted the relative approach to negligence as a broad guideline, without applying that
approach in all its ramifications.63
It is important to note, as pointed out in the above passage, that the relative approach does not
require that the reasonable person should have foreseen the exact or precise manner in which the
harm was caused. So, although the specific consequences must have been reasonably foreseeable,
the exact manner in which the consequences actually came about need not be.

PAUSE FOR The relative approach to negligence

REFLECTION Some advocates of the relative approach have argued that applying it
prevents the need for any enquiry into legal causation, because the
relative approach uses foreseeability to fix liability and to limit it. Strictly
speaking, this view is correct. However, courts tend to use both the
relative approach and legal causation, often in combination. See Smit v
Abrahams,64 where the Court noted that legal causation is part of our law,
and Sea Harvest Corporation (Pty) Ltd v Duncan Dock Cold Storage
(Pty) Ltd,65 where the Court applied the relative approach and yet legal
causation played an important role.
Does it really matter which theory is adhered to? Irrespective of the
approach, the result should be the same. The relative approach
combines the questions into a single enquiry, whereas the abstract
approach follows a two-stage process. The core issue in each is whether
there was reasonable foreseeability of the general type of harm that
occurred.

Irrespective of the approach that is followed, the fundamental factor is the nature or magnitude of
the risk. When assessing the magnitude of the risk of harm, two questions arise:
1. How strong is the possibility that the harm will occur?
2. How serious will the damage be if the risk materialises?

Magnitude of risk is the outcome of an interrelationship between foreseeability of harm and the
possibility of that harm eventuating. If the likelihood of the harm occurring is great, the harm is
normally foreseeable.66 However, where the risk of harm is very small, the conclusion might be that
harm is not foreseeable at all,67 unless the harm that might eventuate is potentially serious. Where
there is a likelihood of serious harm, as in Lomagundi Sheetmetal and Engineering (Pvt) Ltd v
Basson,68 harm will probably be foreseeable. The circumstances of each case will determine the
result.69
Stratton v Spoornet70
An eight-year-old boy and his friend walked to the railway line to watch for trains. To see
whether a train was approaching, the boy climbed to the top of a railway pylon, where he
received a serious electric shock. The boy’s father (Stratton) sued Spoornet for all medical
expenses related to the accident and at issue was whether Spoornet had been negligent.
Stratton had to prove that:
• A reasonable person in Spoornet’s position would have foreseen the reasonable possibility
of injury to others.
• A reasonable person would have taken reasonable preventative measures.
• Spoornet failed to take such measures.

The Court defined ‘reasonable possibility’ as a possibility that would not be ‘too remote or
fanciful’, while keeping in mind the particular facts of the case.71 The Court found that, while a
reasonable person would indeed have foreseen that children might be injured in a collision
with a train, it was not reasonably foreseeable that a child might be injured due to an
electrical shock. Therefore, even though the possible harm from an electrical shock was of a
serious nature, the circumstances indicated that a reasonable person would not have
foreseen the risk as it eventuated.72

Sea Harvest Corporation (Pty) Ltd v Duncan Dock Cold Storage (Pty) Ltd73
Sea Harvest Corporation owned goods that were stored in a newly built cold store in the
Table Bay harbour area that was owned by Duncan Dock Cold Storage. During New Year’s
Eve celebrations, someone fired a distress flare that landed on the store’s roof and set fire to
the roof guttering. The resultant fire destroyed the store. There was no anti-fire sprinkler
system in the roof because the experts who were involved in the store’s construction
considered the building to have a low fire risk. The building had been constructed with non-
combustible material and the guttering material could only ignite from a very high intensity
heat source. Sea Harvest sued Duncan Dock and Portnet (the employer of the port engineer)
for damages, alleging that Duncan Dock had been negligent in not installing an anti-fire
sprinkler system and also in that the managing engineer of the cold store had not remained
on site during the midnight celebration to make sure that there were no incidents. The
evidence showed that a sprinkler system would either have extinguished the fire, or at least
have contained it. The claim against Portnet was based on the port engineer’s negligent
failure to warn of the danger posed by distress flares fired over New Year.
Scott JA (for the majority) considered the test for negligence as developed in Kruger v
Coetzee,74 and as redeveloped in Mukheiber v Raath:75
(a) A reasonable person in the position of the defendant:
(i) Would foresee harm of the general type that actually occurred
(ii) Would foresee the general type of causal sequence by which that harm occurred
(iii) Would have taken steps to guard against it.
(b) The defendant failed to take those steps.

Scott JA reiterated that dividing the negligence enquiry into stages (as described) is only a
guideline in resolving the issue. The ultimate analysis for determining negligence is whether,
in the particular circumstances, the conduct complained of falls short of the standard of a
reasonable person. When applying the formula as set out in Kruger and Mukheiber, there
must always be some measure of flexibility to provide for grey area cases, since there is not
one universally applicable formula that can cater for every case. The present case involves a
grey area because the result (destruction by fire) was readily foreseeable, but the cause (fire
originating from an outside source with sufficient intensity to ignite roof guttering) was not. A
too-rigid approach in borderline cases, where culpability and remoteness of damage queries
are so closely linked, can lead to attributing culpability in a manner that has been termed
negligence ‘in the air’. Therefore, judicial judgement plays an important role while paying due
regard to the facts of each case.76
On the facts, the general possibility of fire was foreseeable and indeed foreseen, and there
was sufficient fire-fighting equipment inside the building. However, the manner in which the
fire started was very different from what was foreseeable. The question is whether a
reasonable person would have foreseen the danger of fire on the roof originating from an
external source with sufficient intensity to ignite the guttering.
Due to the low-risk material of which the building was made, and based on the specific
circumstances of this case, the majority held that a reasonable person would not have
foreseen the manner in which the harm had occurred. Thus, neither Duncan Dock nor
Portnet was held liable.
Since the harm was not reasonably foreseeable, there was no need to discuss the matter
of preventability.
Streicher JA (for the minority) agreed that the appeal should be dismissed, but he based his
finding on different grounds, choosing to follow the abstract approach. In his view, fault would
be established:

if a reasonable person in the position of the defendant would have realised that
harm to the plaintiff might be caused by (his) conduct even if he would not have
 realised that the consequences of that conduct would be to cause the plaintiff
the very harm she actually suffered or harm of that general nature.77

The failure to install a sprinkler system was negligent according to the abstract approach, but
the Judge found that there was no causation because the harm had been too remote.

8.5.3.3 Preventability of harm

Once it is established that harm was indeed reasonably foreseeable, one can then enquire whether
that harm was also reasonably preventable. However, in some instances78 the possibility of (serious)
harm resulting may be so slight that the reasonable person would not have taken preventative
measures. In such instances applying the negligence test does not go further than the question of
foreseeability.
Whether a reasonable person would have taken measures to prevent the harm from occurring
depends on the circumstances of each particular case. However, it is not enough merely to indicate
that a reasonable person would have taken some measures to prevent the harm. The plaintiff also
has to indicate which reasonable measures the defendant should have taken in the circumstances to
prevent the harm from occurring.79
City of Salisbury v King80
King sued the City for damages after she had slipped on vegetable matter at a City vegetable
market. However, she failed to prove that the City was negligent because there was no
evidence to establish that the City had a reasonable opportunity to remove the vegetable
matter as it fell on the floor. Although it was reasonably foreseeable that vegetable matter
would fall to the floor, the Court recognised that the standard of care imposed on the City
involves acknowledging the practical difficulties in keeping the floor free of vegetable matter
at all times. The size of the market area, and the practical circumstances indicated that no
reasonable precautionary measures could have ensured that there would be no vegetable
matter on the floor. The City was not liable for consequences that were reasonably
foreseeable, but which could not reasonably be prevented.81

When assessing whether a defendant would have taken steps to guard against the harm, there are
four basic considerations in each case that influence the reaction of the reasonable person regarding
the preventability of harm: 82
1. The degree or extent of the risk created by the actor’s conduct
2. The gravity of the possible consequences if the risk of harm materialises
3. The utility (social usefulness) of the actor’s conduct
4. The burden of eliminating the risk of harm.

To establish negligence, one must prove that the harm was reasonably foreseeable and reasonably
preventable. To determine reasonable foreseeability of harm, one first determines the magnitude of
harm by weighing the likelihood of the harm occurring against the seriousness of that harm. To
determine preventability, one weighs the magnitude of the harm against the utility of the conduct
and the burden that might be imposed in implementing the preventative measures. If the magnitude
outweighs the utility and burden, the conclusion is that a reasonable person would have taken steps
to prevent the harm, and so the defendant’s conduct must be measured against what a reasonable
person would have done in the circumstances. If the utility and burden outweigh the magnitude of
the harm, the reasonable person would not have taken preventative steps, and so the defendant also
did not have to take preventative measures.83
The basis for these assumptions is the standard of care that a reasonable person would adhere
to.
Ngubane v South African Transport Services84
Ngubane had been standing close to the doorway in a crowded train coach while holding onto
an overhead strap. People were still pushing to get in and out of the coach when the train
started with a jerk and moved forward. Disembarking passengers pushed against Ngubane,
causing him to lose his hold and fall backwards out of the open doorway, and down between
the platform and the train. He suffered a spinal fracture that resulted in permanent partial
paralysis.
The evidence showed that the ticket examiner’s role was to remain on the platform until he
was satisfied that it was safe for the train to depart. He had to ensure that all passengers had
embarked and disembarked. In turn, the guard had to ensure that the ticket examiner had
boarded the train and then the guard should have operated the switch to close the doors.
The guard should have signalled to the driver to proceed only after he had entered the
guard’s van and checked from the window that there were no more passengers entering or
leaving the train. These procedures were part of the defendant’s ‘Interdepartmental Working
Instructions’.
South African Transport Services argued that crowd control was a big problem at stations
and in trains, and that it was too expensive to employ more ticket examiners on each train
and to introduce a door mechanism in each coach that could withstand the malpractices of
often undisciplined commuters.
The Supreme Court of Appeal held that a reasonable person would have foreseen the
possibility of harm. In the circumstances of the case, the Court identified four basic
considerations for determining the reaction of a reasonable person in a particular situation:
1. The degree or extent of the risk created by the actor’s conduct
2. The gravity of the possible consequences if the risk of harm materialise
3. The utility of the actor’s conduct
4. The burden of eliminating the risk of harm.

The Court held that the considerations in points (1) and (2) would have prompted a
reasonable person to take steps to prevent the harm. The risk, according to the Court, of the
near certainty of serious or fatal injury due to the train starting when persons were still
boarding and leaving coaches was ‘as obvious as could be’,85 and the evidence demonstrated
clearly that the conduct complained of created a high risk of serious injury. To have carried
out the existing prescribed procedures would have involved no extra cost to the defendant.
Issues of delayed departure could, according to the Court, not be weighed against the other
considerations that required the necessary safety precautions to be taken. Therefore, factors
(3) and (4) did not outweigh factors (1) and (2), which meant that a reasonable person would
have taken steps to guard against the foreseeable harm. So, the failure of the South African
Transport Services’ employees to take steps to prevent the harm rendered their conduct
negligent.

Cape Metropolitan Council v Graham86

Graham suffered serious injuries when a landslide from above a rock cut along Chapman’s
Peak Drive in the Cape Peninsula struck his vehicle. He sued the Cape Metropolitan Council,
the local government body responsible for the management and maintenance of the road, for
delictual damages and alleged that the Council had been negligent in, firstly, failing to warn
the road users of the risk of harm from falling earth and rock, the contention being that two
‘falling rocks’ signs were inadequate, and, secondly, failing to close the road temporarily prior
to the accident.
According to expert evidence, the unstable nature of much of the rock, the height and
slope of the rock cuts, and the steepness and inaccessibility of the mountainside above,
rendered it virtually impossible to prevent rockfalls and so ensure the safety of the road
users. The relevant slope failure had occurred during or after a rainstorm. Experts agreed
that, although rockfalls on this road were unpredictable and could occur in both wet and dry
periods, the risk of rockfalls and landslides associated with the characteristics of the slope
was greatly increased in wet conditions. The decision whether to close the road was left to
the assistant maintenance superintendent in charge of roads in the area. There was no
policy regarding when he should close the road, and there were no guidelines to help him
make this decision. All indications were that he would only take steps to close the road once
a rockfall or landslide had already rendered it unusable.
The Supreme Court of Appeal held that the circumstances of each particular case indicate
the precautionary measures that reasonably should have been taken. Assessing these
circumstances entails a value judgement in determining whether the precautions would be
reasonable in the circumstances or not. The Court listed the following four considerations to
take into account when making such an assessment:
1. The degree or extent of the risk created by the actor’s conduct
2. The gravity of the possible consequences if the risk of harm materialises
3. The utility of the actor’s conduct
4. The burden of eliminating the risk of harm.

The Court decided that, since the Council had knowledge of the particular problems
associated with the road, it had, or ought to have had, knowledge of previous landslides and
major rockfalls in wet weather. Moreover, the rainfall figures were readily available from the
Hout Bay Weather Centre for the Council’s perusal. The Council also had knowledge of other
incidents in which people had suffered injuries as a result of rockfalls. The Court concluded
that there would be times when the appellant would know, or ought to know, that there is an
increased risk of landslides and rockfalls in wet weather. This increase in risk was such that
the road signs warning of falling rocks were no longer adequate to convey to motorists the
true extent of the risk, which would then place the Council in a position to decide whether the
road ought to be closed.
The Council contended that there were utility considerations to take into account in
deciding whether to close the road. If Chapman’s Peak Drive is closed, commuters must take
an alternative route, which increases their travelling distance by 14 km, and commuters often
complain about this. However, the Court was of the opinion that this difficulty amounted to a
mere inconvenience, which did not weigh up against the great likelihood of serious harm
eventuating. There was accordingly no justification for keeping the road open in
circumstances when it should be closed.
The Court reiterated that to establish whether the Council was liable, it had to ask whether,
in the circumstances, the Council’s failure to close the road prior to the accident had been
unreasonable. In answering this question, the Court took into account the following
considerations:
• The rainfall figures for the week preceding the accident and the forecast of further heavy rain
during the weekend
• The major rockfall during the same month in the previous year had occurred at virtually the
same point during or after a rainstorm, resulting in the road being closed for a lengthy period
of time
• On the day before the accident, as well as on the morning of the accident, the assistant
maintenance superintendent had been called out to remove rocks that had fallen onto the
road.

In light of these considerations, the Court found that there could be no doubt that the risk of
slope failures had greatly increased, and that the Council ought to have known this from its
officials who had knowledge of the particular problems associated with the road. Accordingly,
the Court found that the risk of major slope failure and of harm to road users had increased to
such an extent that it outweighed the alleged utility of keeping the road open. In the
circumstances, the Council had been negligent in failing to close the road.

Enslin v Nhlapo87
One evening at 7pm, Nhlapo was driving his motor vehicle on a road alongside a farm and
came across a small herd of cattle. Nhlapo was unable to avoid the cattle and collided with a
young Brahman bull. He sued Enslin, the owner of the farm, for damages. Evidence showed
 that the cattle did not belong to Enslin, but Enslin had the power to decide where on his farm
the cattle could graze, and therefore had control over the cattle. The alleged grounds of
negligence were that Enslin had failed to ensure that the cattle were properly fenced in, had
failed to prevent the cattle from straying onto a public road, and finally had failed to warn
approaching motorists of the presence of cattle on the public road although he could have
done so.
Enslin admitted that the cattle had strayed onto the public road on a prior occasion. The
Court found that Enslin must have been aware that if the cattle on his farm, and in his
grazing camps, were to stray onto the adjoining public road, they could endanger the lives of
road users. So the foreseeability aspect of the enquiry had been satisfied.
The Court found that a reasonable person in Enslin’s position would have taken steps to
prevent the cattle from straying onto the public road, particularly at night. It was common
cause that Enslin had taken some steps, in that a fence and gates separated the grazing
camp from an access road that ran from the public road to a neighbouring property. He had
also told his employees that the gates had to be kept closed. Yet, on the night in question,
the gates were left open by an unknown person. So, the next question was whether a
reasonable person in Enslin’s position would have taken further steps to prevent the cattle
from straying onto the public road. In this regard, the Court found that a reasonable person
would indeed have taken more precautions. According to the Court, a padlock to secure the
steel gate that separated the access road and the public road, or installing a cattle grid on
the access road shortly before it joined the public road, are easy, inexpensive and effective
measures to prevent the cattle from straying onto the public road. Therefore, Enslin’s failure
to take either of these precautions established that he had been negligent. The appeal was
accordingly dismissed.

Note the essential difference between Enslin v Nhlapo 88 and Kruger v Coetzee.89 In Enslin, the
plaintiff managed to show that there were additional alternative steps that a reasonable person
would have taken. In Kruger, the plaintiff did not do so. Hence the different outcomes in
substantially similar circumstances.
Shabalala v Metrorail90
Shabalala boarded one of Metrorail’s trains one evening on a line that ran from Dunswart to
Benoni. Almost immediately after the train pulled away, three men stood up from among the
passengers. One man approached Shabalala and demanded money at gunpoint. Shabalala
replied that he did not have any money, after which the man shot him three times in the legs
and arm. When Shabalala fell to the floor, the assailant proceeded to search his pockets and
took R130 and a train ticket from his wallet. When the train stopped at the Benoni station,
Shabalala managed to get off the train and called out to two security officers who stood on
the platform. He reported what had happened and pointed out the assailants, who were still
on the train. The officers did not take action against the robbers since they were more
concerned with Shabalala, who was injured and bleeding profusely. Shabalala was taken to
hospital and remained there for one week.
In his pleadings, Shabalala stated that he did not see any security officials on the train or
on the platform before the train left the station that evening. There were no security officials
that evening to check the authenticity of train tickets and control who boards the trains. He
furthermore testified that there were about eleven other people on the platform waiting to
board the train and none of them looked suspicious. Also, nothing about the robbers’
appearance was untoward. Shabalala was unable to indicate whether the robbers were in
the group of people on the station platform or whether they were on the train already. He was
also unable to indicate what steps Metrorail should have taken in order to prevent the attack
from occurring.
The Supreme Court of Appeal noted that the grounds of negligence that Shabalala relied
on were of a general nature and ‘relate to a systemic failure on the part of the respondent’. In
other words, the alleged failure did not relate to a failure on the part of an individual
 employee to act in a specific manner in relation to the incident in question, but to a general
failure on Metrorail’s part to put measures in place that would ensure the safety of
commuters.
It was evident that Metrorail had adopted measures to minimise or prevent crime on the
trains by employing security guards on the trains and on the station platforms. Therefore, the
question in this instance was whether Shabalala had discharged the burden of establishing
that those measures were unreasonable in the circumstances, and that, had such measures
been taken, the harm would not have occurred.
The Court emphasised that this did not mean, simply because the foreseeable harm
occurred, that the precautionary steps taken were necessarily unreasonable. Such an
approach would lead to the untenable situation where Metrorail would have had to provide an
absolute guarantee against harm from criminal activities on its trains. Based on the facts, it
appeared that avoiding the attack would have required at least one security guard on each
coach, especially in light of the assailants’ willingness to use their guns and frequent armed
attacks on security guards. Moreover, assuming that the presence of a security guard in the
coach would have prevented the harm from occurring, the question then arises of whether
Metrorail could reasonably be required to employ a guard on every coach on every train. The
Court found that such a requirement would be unreasonable because of the large number of
coaches involved, and in terms of the cost and effort. It might have been reasonable to
require security guards on every coach where lines have been identified as being particularly
vulnerable to criminal activities, but no evidence to this effect had been tendered. In the
absence of adequate evidence that a security guard should be employed on every coach and
that such action would have prevented or minimised the harm, an inference of negligence
could not be drawn. The Court therefore dismissed the appeal.

So, to establish negligence, there are two requirements that must be met:
• Reasonable foreseeability of harm
• Taking reasonable precautionary measures where harm was reasonably foreseeable.

In certain situations, such as in City of Salisbury v King,91 harm might be reasonably foreseeable,
but a reasonable person in the same situation as the defendant would not have taken preventative
measures. It is important to keep in mind that a plaintiff must indicate what (other) reasonable
measures were available to the defendant in the circumstances (as indicated in Kruger v Coetzee 92
and seen in Enslin v Nhlapo).93 However, reasonable precautionary measures are not necessarily
those that prevent or minimise the harm. Sometimes, as in Shabalala v Metrorail,94 the harm occurs
despite a person having taken reasonable measures. Then, it is not a question of prevention, but
rather a question of whether the defendant conformed to the standard of a reasonable person by
taking the steps that a reasonable person would have taken. If a reasonable person would have
taken precautions that the defendant did not take, even if these precautions might not actually have
prevented the harm from occurring, the defendant’s conduct would nevertheless be unreasonable,
but causation would be an issue.

8.5.4 Circumstances and factors that indicate the required standard of

care
In applying the test for negligence as set out in Kruger v Coetzee 95 and Sea Harvest Corporation
(Pty) Ltd v Duncan Dock Cold Storage (Pty) Ltd,96 the following circumstances play a role in
assessing the standard of care against which we should measure a defendant’s conduct.

8.5.4.1 General practice

A defendant who follows general practice in a particular set of circumstances will usually comply
with accepted precautionary measures, provided that a reasonable person would similarly have
followed that general practice. General practice usually sets out the standard of reasonable
behaviour, especially the precautionary measures to apply in particular situations. It also indicates
what society would regard as reasonable conduct or practices.97 Therefore, a defendant can escape
liability by showing that the conduct conforms with normal or general practice. If it did not, the
defendant is considered to have been negligent.
However, general practice is not always decisive, because a reasonable person would not
follow practices that are out of date or unreasonable for any other reason. Common practice on its
own cannot determine negligence, because this would mean that people, or groups of people, would
be able to develop their own standards of what is acceptable, even if they amount to careless
practices.98 The ultimate basis for determining negligence is still whether a reasonable person in the
same circumstances would have acted differently.

8.5.4.2 Legitimate assumption of reasonable conduct of others

People who act according to an assumption that others would reasonably comply with the standard
of a reasonable person, normally act reasonably. In other words, a reasonable person would expect
other people to be reasonable persons also. One can therefore assume that other motorists will obey
the road safety rules,99 or that lifts,100 pavements101 and shop floors102 are in a good condition and safe
for use. However, if one realises that the assumption is not justified, one should adjust one’s
conduct accordingly. Usually pedestrians will not be negligent when they assume that drivers will
stop at stop signs or pedestrian crossings. However, should they be aware that drivers generally fail
to do so, or where they see that an oncoming driver will not obey the traffic rules, the previous
assumption is no longer legitimate, and they must do what a reasonable person would do in the
circumstances and take the necessary precautions to avoid harm. So, if the conduct of a driver (even
if reckless or grossly negligent) is reasonably foreseeable, a person will be negligent if he or she
fails to take reasonable steps to avoid a collision.103

8.5.4.3 Sudden emergency and error of judgement

The so-called ‘doctrine of sudden emergency’ is based on the premise that the law cannot expect a
person who has to act quickly in the face of imminent danger to exercise the same standard of care
as a person who does not have to act under these circumstances. In a sudden emergency situation,
one does not have the opportunity to consider rationally all the consequences of one’s actions.
Meeting the following three requirements will indicate compliance with the reasonable-person
standard:
• The situation must have been one of imminent peril.104
• The situation of imminent peril must not have been caused by the defendant’s own
negligence.105
• The defendant must not have acted unreasonably in the circumstances.106

PAUSE FOR Common practice entailing dangerous measures

REFLECTION Where common practice entails dangerous measures, for example, in the
building industry, one must evaluate the likely danger in terms of the
ordinary experience of those in the building industry who constantly deal
with such likely dangers. This ordinary experience might indicate that the
 practices resorted to are not unreasonable and that it does not create an
unreasonable risk of danger.107

Courts will take into account that some activities require a degree of skill and experience to cope
with dangerous situations that might arise during such activities.108 For example, motorists must
have the skill and ability to cope with the possibility of road accidents, which usually occur
suddenly. However, the ultimate test is still that of reasonableness and how a reasonable person in
the same situation would have acted when faced with the sudden danger.
Although justified error of judgement is often present in instances of sudden emergency, errors
of judgement may also occur in normal or ordinary circumstances. The question is fundamentally
whether a reasonable person in the same situation would have made a similar error of judgement. In
other words, the error of judgement must have been bona fide and it must have been reasonable in
the circumstances.109 An error of judgement due to a lack of the required skill, knowledge and
prudence is negligent, because a reasonable person would either have possessed the necessary
knowledge and would have exercised the necessary skill to avoid making such an error,110 or would
have refrained from acting due to the fact that the requisite skill and knowledge is lacking.111

8.5.4.4 Breach of statutory duty

The legislature can prescribe the appropriate standard of behaviour for some situations. Courts
prefer the approach that breach of statutory duty is merely an indication of negligence.112 This is in
accordance with the fundamental principle that we must determine negligence in terms of the
reasonable-person standard.113 Breach of a statutory duty is merely a factor that we take into
account when determining negligence. So a breach of statutory duty does not necessarily constitute
negligence, and, similarly, compliance with a statutory duty is not necessarily conclusive in
establishing absence of negligence. The ultimate determinant will be whether a reasonable person
in the same situation would have complied with the statutory duty.

8.5.4.5 Dealing with inherently dangerous things, persons or

circumstances
Where a person deals with an inherently dangerous thing, or is in control of a dangerous situation,
that person is required to act with greater care. Examples of dangerous things or situations include
handling loaded firearms, transporting dangerous criminals,114 slippery floors in supermarkets115 and
swimming pools in the vicinity of young children.

8.5.4.6 Danger to children or people with disabilities or incapacities

Society expects people to exercise greater care and vigilance when it is likely that children, people
with disabilities, such as blind persons, or people with a limited capacity to direct their actions,
such as intoxicated persons, are in the vicinity.116 Children, especially, are known to act impulsively
and unpredictably, and a reasonable person would consider this and take precautions against
injuring a child.117 The question whether a reasonable person would have foreseen the presence of
children or persons with disabilities or incapacities depends on the circumstances of each case.118
The following factors are relevant when determining reasonable behaviour where children are seen
or anticipated:
• Visibility of the child
• The child’s apparent age
• The child’s proximity to the road, the path of the vehicle and the width of the road
• The child’s mobility or liveliness
• Indications that the child intends to cross the road
• Extent of supervision by a responsible person
• The child’s apparent awareness of the danger or of the approaching vehicle
• Motorists’ ability to stop the vehicle in the situation.
Santam Insurance Co Ltd v Nkosi119
Just before lunchtime on a Sunday afternoon, a toddler (almost three years old), Elizabeth
Ngwenya, was knocked down by a car and injured. An action for damages was instituted
against the vehicle insurer.
The road where the accident occurred was a narrow strip of tarmac with a T-junction at
one end. The driver of the vehicle, Maseko, was driving on the left-hand side of the road
towards the T-junction at a speed of approximately 30–40 km/h. A car was parked on the
edge of the tarmac on the right-hand side of the road in a manner that caused its rear to
protrude into the street. As Maseko approached the car, he saw an elderly man standing on
its far side. Just as he reached the car, Elizabeth emerged from behind the car and ran in
front of his vehicle. Maseko applied his brakes but the right front of his vehicle struck the
child. The vehicle stopped within a distance of 4,5 m. Under cross-examination, Maseko
conceded that he knew the area well and that there were a lot of children in the
neighbourhood. He also conceded that it was common for children to play in the street, on its
verges, and on the grassy areas next to the street.
The Court reiterated that there was a special duty of vigilance and care on motorists when
they drive near young children. The Court acknowledged that Elizabeth was out of Maseko’s
sight and would have been completely out of sight to any reasonable driver. The issue then
was whether the special duty also applies in instances where children are hidden from a
motorist’s view. If there is a duty regarding hidden children, there would be a greater demand
for care and vigilance.
The Court held that the ultimate test in any such situation is to consider whether a
reasonable person in the motorist’s position would foresee the possible presence of hidden
children. Based on the circumstances of the case (at lunchtime, a person would expect
children to be at home having lunch), and given that Maseko had, for a brief moment, an
unimpeded view of the area where the car was parked, the Court found that a reasonable
person in Maseko’s position would not have anticipated a young child to be behind the
parked car. Alternatively, even if a reasonable person would have anticipated that a young
child was behind the car, the surrounding circumstances meant that a reasonable person
would not have regarded it as a real possibility that required immediate preventative
precautions. Accordingly, the Court found that Maseko had not been negligent.

8.5.5 Attributes of defendants that influence the standard of care required

for reasonable conduct
Courts have recognised certain characteristics that affect the reasonable-person standard, and so
have adapted the standard to establish a realistic expectation of what is reasonable in the
circumstances. These instances include situations where defendants are beginners or experts.
Society does not expect the same of people in these categories as it would expect of the ‘average’
person. For example, if the defendant is a highly skilled medical specialist, applying the
reasonable-person test would be inadequate, because the special knowledge and skill of the
defendant represent a higher standard of care than one can attribute to an average person in society.
8.5.5.1 Beginners
Where beginners of particular activities expose other persons to a risk of harm, the question arises
whether one should take into account their ‘beginner status’ when assessing negligence. Although
novices are required to demonstrate some skill and care, one cannot expect them to demonstrate the
same skill and experience of a reasonably experienced person. Nevertheless, where the conduct of a
novice creates the risk of serious harm to others, there should be no lenience regarding the lack of
experience. The seriousness of the possible harm is therefore a fundamental factor that must be
taken into account. For example, an inexperienced person who is overseeing the burning of fire-
belts cannot escape liability because of that inexperience, because the severity of the foreseeable
harm to surrounding properties is such that one would expect the inexperienced to act with similar
proficiency as an experienced person.120 Similarly, courts will find negligence where a person
undertakes an activity for which expert knowledge is required, even though that person knows, or
reasonably ought to know, that he or she lacks the required skill or knowledge and should not
attempt the activity in question.
African Flying Services (Pty) Ltd v Gildenhuys121
A flying instructor allowed his pupil to fly his aeroplane, which the pupil crashed, and the
instructor claimed damages from the pupil. The accident did not cause harm to anyone else.
In assessing whether the pupil had been negligent, Tindall JA in a separate judgment122
indicated that because the instructor knowingly entrusted his interests to his inexperienced
pupil, the standard against which the pupil’s conduct had to be measured needed to be
adjusted downwards to take into account the risks inherent in the lack of skill and experience
of a pupil.
Van der Walt and Midgley point out123 that if there had been injury to third parties in the
African Flying Services case, the Court would have assessed the pupil’s conduct according
to the standard of care expected from an experienced pilot. Why?
Would the conduct of a newcomer to a sport such as golf124 be treated differently to a
person who is a learner driver? Why?

8.5.5.2 Experts
Where a defendant possesses a skill or competence gained by training and experience, we can
expect a higher standard of care. For example, courts assess the conduct of a medical doctor
according to the standards that a reasonable medical doctor would have adhered to in the same
circumstances.125
Although one can say that the diligens paterfamilias or reasonable person is ‘replaced’ by the
reasonable expert in the specific field in question, the reasonable expert is similar in all respects to
the reasonable person, except that one attributes a measure of the relevant expertise to the
reasonable person. The reasonable expert criterion does not require the highest skill and expertise,
but courts use it to assess the ‘general level of skill and diligence possessed and exercised at the
time by members of the branch of the profession to which the practitioner belongs.’ 126
Therefore, one can take into account the specific branch of an area of expertise when
determining negligence. For example, a medical practitioner who is a specialist will be required to
exercise a greater degree of skill than a medical practitioner who is a general practitioner, due to the
difference in their levels of specialisation and skill.
Although courts pay close attention to the opinions expressed by experts in a particular field of
expertise, the Court in Van Wyk v Lewis 127 reiterated that it is ultimately for courts to decide what is
reasonable under the circumstances. In other words, while courts have due regard to expert
opinions, courts do not have to follow these expert opinions. Therefore, the specific circumstances
of each case play a pertinent role in how courts approach the standard of care in cases of experts,
and also in how courts decide on what society would regard as reasonable for the expert’s conduct.
This approach conforms with the notion in our law that deviation from the general practice in a
particular field of expertise constitutes only prima facie negligence. Based on all the circumstances
and evidence, courts must be satisfied that an expert opinion of the general practice of that
particular field of expertise has a logical basis, and that the relevant risks and benefits were
considered.128
Van Wyk v Lewis129
Lewis was a surgeon who performed an urgent and difficult abdominal operation on Van Wyk.
When the operation was done, a swab that Lewis used was overlooked and it remained in
Van Wyk’s body for about one year. Evidence showed that it was general practice that the
attending nursing sister was responsible for checking and counting swabs. Also, Lewis
submitted that after the operation he had conducted as thorough a search as was allowed by
the patient’s critical condition and he, as well as the nursing sister, believed that all the swabs
were accounted for. The Court found that the mere fact that the accident happened was not
adequate proof of negligence on the part of Lewis. Although Lewis was bound to exercise all
reasonable care and skill associated with his profession, the general practice that the
attending nursing sister carried the responsibility to ensure all used swabs are accounted for
was indeed reasonable and proper, and Lewis was accordingly not negligent in complying
with this general practice.
The Court measured the reasonableness of Lewis’ conduct against the standard of
expertise required of a reasonable surgeon. The question was whether a reasonable
surgeon in the same situation would have acted differently and, if so, whether that
reasonable surgeon would have taken more (or other) precautionary measures than Lewis
did. The general practice of surgery indicates the required standard of care normally
expected of theatre personnel. Based on this general practice, the Court found that Lewis
acted as a reasonable surgeon would have done, and was therefore not negligent.

8.5.5.3 Children
Before 1965, the test for negligence of children in some cases was taken to be that of ‘the
reasonable child of the child defendant’s age and intellectual development’. 130 In 1965, the
Appellate Division held in Jones, NO v Santam Bpk 131 that the test for negligence is always
objective. In other words, once a court has established that a child is accountable in law, the fact
that we are dealing with a child becomes irrelevant for purposes of establishing negligence, and the
negligence of a child is determined according to the reasonable-person test. Courts, therefore, treat
a child defendant in the same way as an adult defendant.
The harshness of this approach is lessened somewhat by the rules that relate to accountability.
As indicated earlier, courts use a subjective test to determine whether a defendant was accountable
at the time that the delict occurred, and so courts take into account a child’s age and emotional and
mental maturity during the accountability enquiry. Only once a court has found a child to be
(subjectively) accountable will it apply the objective test. In assessing accountability, the Court in
Weber v Santam Versekeringsmaatskappy Bpk 132 said that one should give sufficient weight to the
impulsiveness of children due to their lack of maturity.

PAUSE FOR The reasonable-person standard and the negligent child

REFLECTION Although the reasonable-person standard leaves scope for sensitivity
regarding various levels of intellect and skill, there is no sensitivity
regarding the negligent child. Should our law recognise a reasonable-
child test?
 Haffejee v South African Railways and Harbours133
The plaintiff claimed damages on behalf of his son, Ahmed, who was injured by a train when
he was ten years old. The plaintiff’s house was about 150 m from a railway line, and the
street they lived in crossed the railway line. This crossing was also a pedestrian crossing.
Ahmed had been playing ball with friends when an older boy came along and ran off with the
ball. Ahmed chased after the older boy. They ran along a footpath towards the pedestrian
level crossing and ran blindly in front of a moving train. Ahmed was ‘pulled along’ by the train
and flung into the bushes.
The Court first assessed whether Ahmed was accountable at the time of the accident. The
Court found that, despite the fact that Ahmed sometimes acted like a child of seven, he had
the capacity to appreciate the dangers associated with railway lines and trains, and that he
had the ability to adjust his actions accordingly. Having found that Ahmed was indeed
accountable, the Court had to assess his negligence on the same basis as if he were an
adult. Since Ahmed was aware of the dangers and able to act according to that awareness,
and since he ‘ran blindly into the train’ with his mind set on retrieving his ball, his conduct
was, on an objective assessment, negligent and it was the cause of his injuries. Once they
had established negligence, the Court addressed the issue of contributory negligence134 on
the part of the defendant. The Court considered factors such as the dense vegetation around
the railway line, which may have meant that Ahmed did not see the train approaching, the
fact that there was no warning mechanism such as a light to warn people crossing the line,
as well as the fact that the railway line ran through a densely populated area and the
presence of children ought to have been foreseen, especially with the special duty of care
that arises with regard to children. The Court decided that damages should be shared
equally between the plaintiff and defendant.

COUNTER The accountability of children

POINT Compare the Haffejee case with the Hendricks case discussed under
accountability. In Haffejee, the child was ten years old at the time of the
injury. In Hendricks, the child was eleven years and eight months old. Yet
in Haffejee the child who ran in pursuit of his ‘stolen’ ball was found to be
accountable and 50% contributorily negligent. In Hendricks, the child was
found not to be accountable due to a child-like fascination with glass
insulators and therefore no liability could be attributed to the child. Both
cases deal with the child-like preoccupation with the situation that the
children found themselves in, yet the outcomes of the two cases are
dramatically different. The pressing question here is whether the law
should extend the recognition of youthful weakness to the question of
culpability by perhaps recognising a ‘reasonable child’ standard.

8.5.6 Proving negligence

The plaintiff bears the onus of proving, on a balance of probabilities, that the defendant acted
negligently.135 In cases where there is a statutory presumption of negligence, the defendant bears the
burden of rebutting that presumption.136
In some instances the facts of the case are such that one can draw an inference of negligence.
Here, the maxim res ipsa loquitur, translated as ‘the matter speaks for itself’, applies. For example,
this maxim will apply when someone is driving on the wrong side of the road. Such a situation
normally would not occur without negligence. The res ipsa loquitur maxim does not create a
presumption of negligence, since the plaintiff still bears the onus of proof.137 The plaintiff can rely
on this maxim to convince the court of the defendant’s negligence. Where more than one
reasonable possibility can be inferred from the facts, the decision will depend on the probabilities
of the case, and the court will normally hear evidence to indicate the most probable possibility.138
Sometimes a plaintiff cannot produce sufficient evidence of negligence, but the matter at hand
is peculiarly in the knowledge of the defendant. In such cases, less evidence is necessary to make a
prima facie case, because once the plaintiff has put forward all available evidence, the defendant
bears an evidentiary burden to indicate that the conduct in question complied with the necessary
standard of care.139
An irrebuttable presumption of negligence exists where, for example, a defendant keeps a wild
animal in captivity.140

Figure 8.3Fault

8.6 Conclusion
The general term ‘fault’ denotes both intention and negligence, and proof of fault is concerned with
blameworthiness on the part of defendants. However, before blameworthiness can be established,
there must be capacity for fault on the part of a defendant. Therefore, the first step towards
establishing fault is to determine whether the defendant is accountable.
Blameworthiness of plaintiffs, who contributed to their own loss due to intention or negligence
on their own part, involves an enquiry into contributory fault that could lead to a reduction of the
plaintiff’s damages.141

1 Intention and negligence have the same meaning in delict as in criminal law.
2 1993 (3) SA 131 (A) at 154D.
3 Weber v Santam Versekeringsmaatskappy Bpk 1983 (1) SA 381 (A) at 403.
4 1983 (1) SA 381 (A).
5 At 399H–400A.
6 At 400E–G.
7 At 400B–E (our translation).
8 2005 (5) SA 503 (SCA).
9 1983 (1) SA 381 (A).
10 1987 (1) SA 940 (A) at 958G–I.
11 1987 (1) SA 940 (A).
12 R v Victor 1943 TPD 77.
13 1981 (1) SA 1097 (A).
14 See Chapter 10.
15 15 Winterbach v Masters 1989 (1) SA 922 (E) at 925H.
16 1987 (1) SA 940 (A).
17 1980 (3) SA 24 (C).
18 Suid-Afrikaanse Uitsaaikorporasie v O’Malley 1977 (3) SA 394 (A); Minister of Justice v Hofmeyr 1993 (3) SA 131
(A).
19 Luxuria, or conscious negligence, is purely a form of negligence because it is assessed in terms of an objective test
where the wrongdoer’s conduct is measured against the standard of a reasonable person. The question in luxuria cases
would be whether the reasonable person, in the same situation as the defendant, would have proceeded to act when
realising that another harmful consequence might occur. See the discussion of the test for negligence.
20 S v Humphreys 2015 (1) SA 491 (SCA).
21 2016 (2) SA 317 (SCA).
22 Para 26.
23 Para 31.
24 Para 32.
25 Suid-Afrikaanse Uitsaaikorporasie v O’Malley 1977 (3) SA 394 (A) at 403C.
26 Dantex Investment Holdings (Pty) Ltd v Brenner 1989 (1) SA 390 (A) at 396.
27 Dantex Investment Holdings (Pty) Ltd v Brenner 1989 (1) SA 390 (A); Maisel v Van Naeren 1960 (4) SA 836 (C).
28 1960 (4) SA 836 (C).
29 2011 (6) BCLR 577 (CC); 2011 (3) SA 274 (CC) para 39.
30 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 137.
31 Minister of Justice v Hofmeyer 1993 (3) SA 131 (A).
32 Pauw Persoonlikheidskrenking en Skuld in die Suid-Afrikaanse Privaatreg – ‘n Regshistoriese en Regsvergelykende
Ondersoek (1976) at 194–196.
33 1993 (3) SA 131 (A).
34 At 154H–I.
35 Du Bois (Ed) Wille’s Principles of South African Law 9 ed (2009 revised impression) at 1129.
36 Dantex Investment Holdings (Pty) Ltd v Brenner 1989 (1) SA 390 (A) at 396.
37 Gluckman v Schneider 1936 AD 151 at 159.
38 S v Hartmann 1975 (3) SA 532 (C).
39 Dantex Investment Holdings (Pty) Ltd v Brenner 1989 (1) SA 390 (A) at 396.
40 See, for example, Suid-Afrikaanse Uitsaaikorporasie v O’Malley 1977 (3) SA 394 (A) at 401–402.
41 See, for example, Delange v Costa 1989 (2) SA 857 (A) at 861.
42 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 147, fn 6.
43 Minister van Veiligheid en Sekuriteit v Kyriacou 2000 (4) SA 337 (O) at 341J–342C; S v Motsepe 2015 (5) SA 126
(GP) para 21.
44 National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA).
45 1960 (4) SA 836 (C).
46 1916 TPD 114.
47 S v Chretien 1981 (1) SA 1097 (A).
48 Weber v Santam Versekeringsmaatskappy Bpk 1983 (1) SA 381 (A) at 410–411.
49 Herschel v Mrupe 1954 (3) SA 464 (A) at 490.
50 2000 (1) SA 827 (SCA).
51 1966 (2) SA 428 (A). This test was reformulated in Mukheiber v Raath 1999 (3) SA 1065 (SCA).
52 1966 (2) SA 428 (A).
53 At 430.
54 1966 (2) SA 428 (A).
55 1966 (2) SA 428 (A).
56 2004 (3) SA 305 (SCA) at 325E–G.
57 Res ipsa loquitur can be translated as ‘the matter speaks for itself’. It is relevant in proving negligence. See the
discussion at the end of this chapter on proving negligence.
58 1973 (4) SA 523 (RA).
59 Sea Harvest Corporation (Pty) Ltd v Duncan Dock Cold Storage (Pty) Ltd 2000 (1) SA 827 (SCA) para 23.
60 Standard Chartered Bank of Canada v Nedperm Bank Ltd 1994 (4) SA 747 (A).
61 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 276–277.
62 2011 JDR 0250 (SCA).
63 Para 13 (footnotes omitted).
64 1992 (3) SA 158 (C) at 163.
65 2000 (1) SA 827 (SCA).
66 Lomagundi Sheetmetal and Engineering Co (Pvt) Ltd v Basson 1973 (4) SA 523 (RA) at 525; Loureiro v Imvula
Quality Protection (Pty) Ltd 2014 (3) SA 394 (CC) paras 62–63; Jacobs v Transnet Ltd t/a Metrorail 2015 (1) SA 139
(SCA) paras 6–9; Minister of Justice and Constitutional Development v X [2014] 4 All SA 586 (SCA); 2015 (1) SA
25 (SCA) para 34.
67 Bolton v Stone [1951] AC 850.
68 1973 (4) SA 523 (RA).
69 Green v Naidoo 2007 (6) SA 372 (W) paras 38–44.
70 1994 (1) SA 803 (T).
71 At 809H.
72 At 810F–G.
73 2000 (1) SA 827 (SCA).
74 1966 (2) SA 428 (A).
75 1999 (3) SA 1065 (SCA). This formulation is a classic exposition of the relative approach to negligence.
76 Paras 22 and 25.
77 As quoted by Streicher JA at 845 para 3 from Groenewald v Groenewald 1998 (2) SA 1106 (SCA) at 1112I–J.
78 Ngubane v South African Transport Services 1991 (1) SA 756 (A); Shabalala v Metrorail 2008 (3) SA 142 (SCA).
79 Kruger v Coetzee 1966 (2) SA 428 (A) at 431G-432D.
80 1970 (2) SA 528 (RA).
81 See also Kruger v Coetzee 1966 (2) SA 428 (A) where the Court held that the plaintiff had the onus of proving
negligence, which includes the burden of indicating which reasonable precautions the defendant should have taken.
See also Avonmore Supermarket CC v Venter 2014 (5) 399 (SCA) para 20.
82 Ngubane v South African Transport Services 1991 (1) SA 756 (A); Cape Metropolitan Council v Graham 2001 (1)
SA 1197 (SCA).
83 Lomagundi Sheetmetal and Engineering (Pvt) Ltd v Basson 1973 (4) SA 523 (RA).
84 1991 (1) SA 756 (A).
85 At 758I–J.
86 2001 (1) SA 1197 (SCA).
87 2008 (5) SA 146 (SCA).
88 2008 (5) SA 146 (SCA).
89 1966 (2) SA 428 (A).
90 2008 (3) SA 142 (SCA).
91 1970 (2) SA 528 (RA).
92 1966 (2) SA 428 (A).
93 2008 (5) SA 146 (SCA).
94 2008 (3) SA 142 (SCA).
95 1966 (2) SA 428 (A).
 96 See section 8.5.3.
97 See, for example, Colman v Dunbar 1933 AD 141 at 157.
98 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 165.
99 Moore v Minister of Posts and Telegraphs 1949 (1) SA 815 (A); Van As v Road Accident Fund 2012 (1) SA 387
(SCA).
100 Faiga v Body Corporate of Dumbarton Oakes 1997 (2) SA 651 (W).
101 Stewart v City Council of Johannesburg 1947 (4) SA 179 (W).
102 Swart v Department of Economic Affairs, Environment and Tourism (Eastern Cape) [2001] 2 All SA 357 (E).
103 Van der Walt and Midgley (2016) para 165; Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed
(2015) at 143.
104 Msutu v Protea Assurance Co Ltd 1991 (1) SA 583 (C).
105 Brown v Hunt 1953 (2) SA 540 (A).
106 Bonthuys v Visagie 1931 CPD 75.
107 Colman v Dunbar 1933 AD 141.
108 Van der Walt and Midgley (2016) para 167.
109 Neethling v President Insurance Co Ltd 1978 (2) SA 744 (T) at 745; Pringle v Administrator, Transvaal 1990 (2) SA
379 (W) at 395–396.
110 Mouton v Die Mynwerkersunie 1977 (1) SA 119 (A).
111 See section 8.5.5.1.
112 Van der Walt and Midgley (2016) para 164; Neethling and Potgieter (2015) at 157–158.
113 Clairwood Motor Transport Co (Pty) Ltd v Akal & Sons 1959 (1) SA 183 (N); Bekker v Du Toit 1974 (3) SA 248 (O).
114 Van der Spuy v Minister of Correctional Services 2004 (2) SA 463 (SE).
115 Brauns v Shoprite Checkers (Pty) Ltd 2004 (6) SA 211 (E); Checkers Supermarket v Lindsay (123/08) [2009] ZASCA
26; 2009 (4) SA 459 (SCA); [2009] 3 All SA 487 (SCA) (27 March 2009).
116 Roelse v Commercial Union Assurance Co of SA Ltd 1981 (1) SA 1126 (A).
117 Santam Insurance Co Ltd v Nkosi 1978 (2) SA 784 (A); Knouwds v Administrateur, Kaap 1981 (1) SA 544 (C).
118 Road Accident Fund v Landman 2003 (1) SA 610 (C).
119 1978 (2) SA 784 (A).
120 Simon’s Town Municipality v Dews 1993 (1) SA 191 (A).
121 1941 AD 230.
122 At 245.
123 Van der Walt and Midgley (2016) para 172.
124 Clarke v Welsh 1975 (4) SA 469 (W).
125 Van Wyk v Lewis 1924 AD 438; Castell v De Greeff1993 (3) SA 501 (C); Mukheiber v Raath 1999 (3) SA 1065
(SCA).
126 Van Wyk v Lewis 1924 AD 438 at 444.
127 At 460.
128 Van der Walt and Midgley (2016) para 171, fn 38.
129 1924 AD 438.
130 Lentzner NO v Friedmann 1919 OPD 20.
131 1965 (2) SA 542 (A).
132 1983 (1) SA 381 (A).
133 1981 (3) SA 1062 (W).
134 We discuss contributory negligence in Chapter 35.
135 Ntsala v Mutual & Federal Insurance Co Ltd 1996 (2) SA 184 (T) at 190.
136 Gouda Boerdery BK v Transnet Ltd 2005 (5) SA 490 (SCA); see also Eskom Holdings Ltd v Hendricks 2005 (5) SA
503 (SCA) where Eskom did not succeed in rebutting the presumption of negligence.
137 Stacey v Kent 1995 (3) SA 344 (E) at 352.
138 Van der Walt and Midgley (2016) para 156.
139 Jamneck v Wagener 1993 (2) SA 54 (C) at 65.
140 Zietsman v Van Tonder 1989 (2) SA 484 (T) at 492.
141 See Chapter 35, where the issue of the consequences of a plaintiff’s contributory fault is addressed.
Chapter 9

Wrongfulness

9.1 Introduction

9.2 What is the role of wrongfulness in South African law?

9.3 When is the issue of wrongfulness likely to arise?

9.4 Wrongfulness – an attribute of conduct?

9.5 Wrongfulness presupposes both conduct and consequences, which

do not necessarily occur
| simultaneously
9.6 Wrongfulness – a matter of law

9.7 The general criteria for determining wrongfulness: Pathways to policy

9.8 Wrongfulness and the infringement of a right

9.9 Wrongfulness and breach of duty

9.10 Different concepts of ‘duty’

9.11 What is the content of the legal duty?

9.12 Policy considerations

9.13 Is wrongfulness determined with hindsight (ex post facto) or from the
perspective of the defendant at the time of the relevant conduct with
foresight (ex ante)?
9.13.1 Involvement of a strictly ex post facto perspective and exclusion of an ex ante
or actor-oriented perspective
9.13.2 Determining wrongfulness or fault first
9.13.3 The nature of fault is in some cases relevant to wrongfulness
9.13.4 Wrongfulness and negligence are sometimes based on similar factors
 9.14 Conclusion

9.1 Introduction
Wrongfulness is closely linked to the central idea of the law of delict, which is that liability is
imposed when a person unreasonably causes harm to another.

TERMINOLOGY Wrongfulness and unlawfulness

The terms ‘wrongfulness’ and ‘unlawfulness’ are interchangeable.
In the law of delict, both these terms have a special, technical
meaning. They indicate that causing harm to another person is
sufficiently unreasonable or unacceptable for the law of delict to
impose liability, as long as the other requirements for such liability
are also met. In this book, we use the terms ‘wrongful’ and
‘wrongfulness’, but they are interchangeable with the terms
‘unlawful’ and ‘unlawfulness’. If the term ‘unlawful’ is more
commonly used, as in the case of the delict of ‘unlawful
competition’, one uses this expression, although technically,
‘wrongful competition’ would also be correct.

Some of the aspects of wrongfulness that we discuss in this chapter have been hotly debated. These
include the relationship between wrongfulness and negligence, whether wrongfulness is determined
with hindsight, and the nature of the duty that is often the subject matter of the enquiry into
wrongfulness.1
In most legal systems the concept of wrongfulness exists, but there is no uniformity in how to
use the expression. In a broad sense, wrongfulness indicates the infringement of an interest worthy
of legal protection. However, this description is so wide that it could refer to the concept of
delictual liability generally.

9.2 What is the role of wrongfulness in South African law?

In South Africa, wrongfulness is a requirement for delictual liability, in addition to the following
other requirements for liability in delict:
• The conduct requirement is used to establish the presence of human conduct in the form of
positive conduct or an omission. An omission involves enquiring whether a duty not to cause
harm exists. This is a wrongfulness issue: The question is whether the harm fell within the
defendant’s scope of responsibility, so that society would regard failure to prevent the harm as
wrongful.
• The harm requirement deals with the effects of the infringement of a right or interest.
However, it does not explain what interests the law of delict will protect or to what extent it
will protect these interests. Physical integrity and tangible property are generally recognised as
fully protected interests, and the infringement of these interests by positive conduct can be said
to be prima facie wrongful. Boberg2 says in this regard:
 … it is settled law that all harm to person or property caused by a positive act is
prima facie wrongful.

• However, in the case of pure financial loss, privacy, reputation and mental distress, the extent
of protection is often a question of wrongfulness.
• The requirements of conduct, causation, harm and fault do not adequately deal with balancing
conflicting rights or interests, for example, reputation versus free speech, enjoyment of
property versus harm to a neighbour, freedom of competition versus harm to trading goodwill,
and freedom of action versus a duty to protect. Balancing these rights or interests is a
wrongfulness issue.
• The conclusion that the defendant’s conduct culpably (negligently or intentionally) caused the
plaintiff’s harm is not sufficient for liability. This is because one could in certain
circumstances justify causing harm that would otherwise be wrongful, for example, on the
basis of defence, necessity, consent or statutory authority. Justification is a question of
wrongfulness.
• The requirement of fault (negligence or intent) generally deals with the blameworthiness of the
defendant’s conduct, but not with the weight given to intent or reprehensible motive of the
defendant. Also, fault does not deal with the effects of mistakes made by the defendant, for
example, in cases of putative justification. Intent, awareness of the possibility of harm, a
reprehensible motive and conscious negligence (recklessness) can be indicators of
wrongfulness.
• The requirements of conduct, causation, harm and fault do not leave sufficient scope for policy
considerations relating to, for example, guarding against indeterminate liability, guarding
against hampering or disrupting public administration, the availability of an alternative
remedy, or the maintenance of free competition. Rights and duties in delict reflect the range of
protection that the law affords. The ‘right’ to this protection is linked to the ‘duty’ of a person
to not cause loss. However, the content of both the ‘right’ and the ‘duty’ reflect a value
judgement on the appropriate extent of protection, as judicially determined. This value
judgement forms part of the enquiry into wrongfulness.
• Under a fault-based system, the requirement of either negligence or intent acts as an important
filter in evaluating whether a court should impose liability. Where liability is strict, eliminating
the fault requirement does not mean that all risk of harm is indiscriminately transferred to the
defendant who caused the harm. Strict liability does not mean absolute liability. It still requires
wrongfulness based on reasonableness and policy.

In Le Roux v Dey 3 Brand AJ described wrongfulness as follows:

In the more recent past our courts have come to recognise, however, that in the context of
the law of delict: (a) the criterion of wrongfulness ultimately depends on a judicial
determination of whether – assuming all the other elements of delictual liability to be
present – it would be reasonable to impose liability on a defendant for the damages flowing
from specific conduct; and (b) that the judicial determination of that reasonableness would
in turn depend on considerations of public and legal policy in accordance with constitutional
norms. Incidentally, to avoid confusion it should be borne in mind that, what is meant by
reasonableness in the context of wrongfulness has nothing to do with the reasonableness of
the defendant’s conduct, but it concerns the reasonableness of imposing liability on the
defendant for the harm resulting from that conduct.

In Country Cloud Trading CC v MEC Department of Infrastructure Development Gauteng 4

Khampepe J explained the function of wrongfulness as follows:

 Wrongfulness is an element of delictual liability. It functions to determine whether the
infliction of culpably caused harm demands the imposition of liability or, conversely,
whether ‘the social, economic and other costs are just too high to justify the use of the law of
delict for the resolution of the particular issue’. Wrongfulness typically acts as a brake on
liability, particularly in areas of the law of delict where it is undesirable and overly
burdensome to impose liability.

In summary, wrongfulness involves the scope of protection that the law affords to various interests,
the scope of a person’s responsibility to act and the policy considerations that relate to whether the
law of delict should intervene. The fundamental questions in this enquiry are:
• Should a court, as a matter of policy, impose liability on the defendant in these circumstances?
• Is it reasonable to compensate the plaintiff for the loss, and for the defendant to bear the loss?
For a more detailed examination of the criteria for wrongfulness one must look more closely at
situations where the issue of wrongfulness arises.

9.3 When is the issue of wrongfulness likely to arise?

Wrongfulness does not usually cause problems in cases that involve positive conduct causing
bodily injury or damage to property. In law, all harm to person or property caused by a positive act
is prima facie wrongful. Wrongfulness usually causes problems in cases where the conduct is an
omission or a statement, where the harm is pure economic loss or psychiatric injury, or where there
is a conflict of rights. In these instances, courts generally use wrongfulness to determine whether
the infringement of an interest is worthy of legal protection and whether the harm caused falls
within the defendant’s scope of responsibility. In instances that involve a conflict of rights,
determining wrongfulness requires a value judgement on the question of which right should yield to
the other.
In contentious cases, wrongfulness involves applying wide and general criteria. The general
criteria for determining wrongfulness are:
• Reasonableness (sometimes referred to as ‘general reasonableness’)
• The legal convictions prevailing in the community
• Society’s boni mores.

Applying these criteria in the final instance involves public policy, as we explain more fully in
section 9.7.

9.4 Wrongfulness – an attribute of conduct?

Wrongfulness is not an attribute of conduct alone. It characterises the outcome of a sequence that
involves conduct and the harm caused by the conduct. Without harmful consequences, there is no
wrongfulness and no delict, however reprehensible the conduct in question may be. For example,
driving at high speed down a busy street is reckless and reprehensible conduct, but if a person
causes no harm by doing this, there is no wrongfulness and accordingly no delict.

PAUSE FOR Wrongfulness and conduct

REFLECTION Walker, commenting on the law of Scotland, writes:5

The act or omission by itself is not wrongful or delictual

unless it brings about as an immediate consequence some
harm to a legally protected interest of another person. The
 law of delict is concerned with harms, with wrongs to, or
infringements of, the interests of others, not with the bare
conduct whereby those harms come about.

In this respect the law of delict differs from criminal law. In criminal law, a particular act is
sometimes prohibited and therefore characterised as wrongful, for example, driving dangerously or
possessing a prohibited substance. Where a crime is mainly defined in terms of a particular causal
sequence, for example, culpable homicide, involving negligently causing the death of a human
being, wrongfulness characterises the outcome of this sequence, which involves conduct and the
harm it causes, as in the case of delict.
In Cape Empowerment Trust Limited v Fisher Hoffman Sithole 6 Brand JA confirmed that
wrongfulness in delict indicates the reasonableness of imposing liability:
… it should be borne in mind that, what is meant by reasonableness in the context of
wrongfulness has nothing to do with reasonableness of the defendant’s conduct [which is
part of the element of negligence], but it concerns the reasonableness of imposing liability on
the defendant for the harm resulting from that conduct.

Wrongfulness is therefore not only an attribute of conduct. The misconception that wrongfulness is
an attribute of conduct is reflected in expressions such as ‘wrongful conduct’ or ‘a wrongful act’.
Increasingly the use of these expressions reflects looseness or inaccuracy of terminology, rather
than attempts to convey the meaning that wrongfulness is only an attribute of conduct.

9.5 Wrongfulness presupposes both conduct and

consequences, which do not necessarily occur
simultaneously
Wrongfulness takes into account both the defendant’s conduct and the consequences of this
conduct. The conduct and its consequences may not happen at the same time or the same place. For
example, an assault may cause immediate injury, whereas the effects of building an unsafe wall
may not be seen until much later, when the wall collapses. In RAF v Mtati 7 negligent driving
caused an accident in which a pregnant woman was injured. She later gave birth to a child with
brain damage. The Supreme Court of Appeal decided that the harm occurred when the child was
born with injuries. The completed causal sequence at that time was characterised as wrongful.

9.6 Wrongfulness – a matter of law

Wrongfulness is a matter of law.8 Courts usually do not hear evidence on the issue of wrongfulness.
The onus is on the plaintiff to make and prove factual allegations from which wrongfulness, which
is a matter for judicial determination, can be deduced.9
A defendant who relies on justification for causing harm bears the onus to allege and
eventually present factual evidence that would enable one to draw an inference of such a
justification. Justification is part of the enquiry into wrongfulness and is a conclusion of law.

PAUSE FOR Wrongfulness must be pleaded

 REFLECTION The plaintiff in a delict action must allege the following in the particulars
of the claim:
• Wrongfulness
• Facts that indicate that the harm caused is wrongful
• Facts that substantiate the relevant policy considerations.10

If the plaintiff does not make these allegations, the defendant can raise
an exception on the basis that the pleadings do not disclose a cause of
action. A court must then decide whether the allegations of fact, if proved,
would establish that the defendant wrongfully caused harm. When
deciding wrongfulness in exception proceedings, courts assume that the
other elements of the delict, such as causation and negligence, are
present, as alleged.11
The following are examples of cases where the issue of wrongfulness
was decided by way of exception:
• Engineers negligently failed to carry out their professional duties in terms of a
contract for building a glass factory. This meant that expensive additional work
had to be done on the factory. The Court decided that the facts alleged did not
indicate wrongful causing of harm for the purposes of delict. The factory owner
(plaintiff) could only sue the engineers in contract. 12
• A civil engineering contractor, who negligently cut a cable supplying electricity
to a brick factory, caused harm in the form of loss of production to the factory
owner. The contractor cut the cable during excavation work and the factory
owner (plaintiff) alleged that the contractor knew where the cable was, and also
knew that the factory would lose production if he cut the cable. The Court
decided that the facts alleged indicated wrongful causing of harm.13
• Engineers negligently recommended an inadequate system of waterproofing
for aquarium tanks. The result was that later expensive additional work had to
be done on the aquarium. The alleged negligence of the engineers occurred
before the parties entered into a detailed contract. The contract could have
provided for liability arising from the pre-contractual work, but it did not. The
aquarium owners, therefore, sued in delict, but the Court decided that the
causing of harm in the pre-contractual phase was not wrongful for the purposes
of delict.14

9.7 The general criteria for determining wrongfulness:

Pathways to policy
Applying the general criteria or standards for determining wrongfulness (general reasonableness,
the legal convictions prevailing in the community and the boni mores) in the final instance involves
public policy and a value judgement. A number of judgments acknowledge that the decisions in
them were determined or influenced by policy considerations.
The legal policy makers of the community, such as the legislature and judges must take on
board the legal convictions of the community. There is constant interplay between the legal concept
of wrongfulness and the fundamental values of society. The general criteria, or standards, for
determining wrongfulness are of a legal rather than a social, moral, ethical or religious nature, but
the criteria do reflect societal values. In applying the legal convictions of the community, a court is
concerned with whether the community should regard the harm caused in a particular case as
wrongful for the purposes of delictual liability. It is not concerned with what the community
regards as socially, morally, ethically or religiously right or wrong. However, sometimes, this
involves a choice between two moral concepts, neither of which is wrong, for example, between the
two concepts that one should respect the privacy of others, and that one should speak out on matters
of public interest. In such instances a court has to decide which concept, based on the particular set
of facts, should be given preference when it comes to legal protection.
When enquiring into wrongfulness, one can either focus on the infringement of a right or on
the breach of a legal duty. In the final analysis, the decision involves an assessment of
reasonableness and public policy. A court must weigh up the interests of the people involved, and
also take into account the interests and convictions of the community. The personal views of the
judge, the parties, or a segment of the community are not the measure of what one should regard as
lawful or wrongful.
The meaning of general reasonableness, boni mores and the legal convictions of the
community should agree with the norms and values in the Constitution.

PAUSE FOR Constitutional norms

REFLECTION Consider how the constitutional norms of equality, non-discrimination,
personal security, and accountability of public officials have contributed to
the assessment of wrongfulness in the cases in Table 9.1.

Table 9.1 Constitutional norms contribute to the assessment of wrongfulness

Case Details Assessment

Du Plessis v Road Accident The right to support between partners in a same-sex Wrongful causing of
Fund relationship can form the basis of a claim for loss of harm by infringement
support against the person who negligently caused the of right
death of one partner

Carmichele v Minister of Safety Police and prosecutors failed to oppose the release of Wrongful causing of
and Security (Centre for Applied a person with a prior conviction for violence, while harm by breach of
Legal Studies Intervening) awaiting trial on a new charge duty

Minister of Safety and Security The police failed to deprive a person of his firearms Wrongful causing of
v Van Duivenboden and licence, although they knew that he was prone to harm by breach of
violence when drunk duty

Minister of Safety and Security The police failed to enquire into the psychological Wrongful causing of
v Hamilton fitness of an applicant for a firearm licence harm by breach of
duty

Van Eeden v Minister of Safety Police allowed a prisoner with a history of violent Wrongful causing of
and Security crime to escape, resulting in a further assault on the harm by breach of
plaintiff duty

The concepts of the legal convictions of the community and boni mores indicate objective and
normative standards for determining wrongfulness. Courts do not hear evidence on the content of
the legal convictions of the community or the boni mores. These are general guidelines for the
value judgement required of a court when assessing wrongfulness. These general criteria provide
courts with ‘a legal standard firm enough to afford guidance to the Court, yet flexible enough to
permit the influence of an inherent sense of fair play’, and this standard is based on ‘the general
sense of justice of the community, the boni mores, manifested in public opinion’.15
However, conclusions on wrongfulness based only on the general standards risk being vague
and difficult to analyse. Applying these general standards requires an open and structured process
of reasoning, with reference, inter alia, to:
• The specific rights and interests involved
• The relationship between the parties
• Relevant provisions of the Constitution and of other legislation
• Relevant policy considerations.

9.8 Wrongfulness and the infringement of a right

The enquiry into wrongfulness can focus either on the infringement of a right, or on the breach of a
duty. One can look at wrongfulness from either of these two angles and it does not really matter
which angle one prefers or starts off with. This is because rights and duties are related, and a breach
of a duty will, at the same time, constitute an infringement of a right. The choice of which to use
depends upon the facts and whether it is easier to recognise the right or the duty.
Rights to person and property are settled and easy to recognise. Courts work from the premise
that causing harm by infringing these rights is wrongful, without finding it necessary to refer to the
general criterion of reasonableness, the boni mores, the legal convictions of the community or
concomitant policy considerations. It is settled law that harm to person or property caused by a
positive act is prima facie wrongful. In the absence of some form of justification, such as self
defence or necessity, the infringement of such rights is wrongful.
The rights-based approach to wrongfulness mostly involves the following settled categories of
rights:
• Real rights in respect of movable or immovable property
• Personal rights in respect of an act or performance required from another person, such as
payment of a debt
• Personality rights in respect of aspects of human personality, such as bodily integrity, dignity
or reputation
• Immaterial property rights in respect of intangible products of the human mind, such as
patents, trademarks or copyright.

There is no closed list of protected rights, however, and one can also add sub-categories or new
categories of rights to these settled categories, such as:

• Personal immaterial property rights in the form of the right to earning capacity or personal
goodwill
• The right to information
• The rights to privacy, identity, goodwill and trade secrets.

Infringement of a right involves disturbing or limiting the holder of the right to enjoy, use or
dispose of the interest that is the object of the right. Examples of infringements of rights include:

• Damaging or appropriating property that belongs to another

• Causing a breach of privacy or damage to reputation
• Appropriating a patented process or a trademark.

Factual disturbance or limitation, however, is not enough. The infringement must be unreasonable
in terms of general criteria or standards based on boni mores, the legal convictions of the
community and public policy. One often needs to apply these general criteria to balance conflicting
rights or interests, such as reputation versus free speech, and enjoyment of property versus harm to
a neighbour.
Figure 9.1 Wrongfulness and Infringement of a right

9.9 Wrongfulness and breach of duty

The enquiry into wrongfulness often focuses on whether a duty exists, especially in cases where it
is not easy to identify a right. It is easier to look at the issue from a duty point of view. This occurs
where the defendant is blamed for an omission (failure to prevent harm to another person), for
causing financial loss by misstatement or unsound advice, or where breach of a statutory duty is
involved. Breach of a legal duty to another person also involves infringement of their right not to be
harmed, because to every obligation there is a right and a duty side, and a legal duty is the converse
of a right. Assessing wrongfulness from the duty side simply involves a difference in the initial
focus of the enquiry. The general criterion for determining wrongfulness remains the same: whether
the defendant unreasonably, or contrary to the boni mores or the legal convictions of the
community, failed to prevent harm to the plaintiff.
The breach-of-legal-duty cases involve situations in which one cannot say that the harm is
prima facie wrongful, as there is no clear right to be protected from such harm or loss.16 The focus
is on the existence of a duty. Prominent examples of such cases are liability for an omission and
liability for causing financial loss. In these cases, the enquiry often turns on responsibility and
relationship issues. Typical are claims:
• By a resident against a local authority for failing to repair a public facility or warn of danger17
• By a member of the public against the police for failing to provide protection18
• By a customer against a bank for furnishing incorrect information or making an incorrect
payment19
• By a client against a financial advisor for carelessly handling investments20
• By a patient against a doctor for inadequate professional services.21

In these cases the concept of duty ties in with the general function of the enquiry into wrongfulness.
The general function is to determine whether the affected interest of the plaintiff (judged either on
its own or in balance with a conflicting interest of the defendant) deserves protection from the
defendant’s action or lack of action, so that the burden of bearing the loss should be shifted from
plaintiff to defendant.

Figure 9.2 Wrongfulness and breach of a duty

9.10 Different concepts of ‘duty’
In this section we examine how the courts’ approach to the concept of ‘duty’ has developed over
time, and how the modern concept of ‘legal duty’ for determining wrongfulness differs from the
concept of a ‘duty of care’ as used in earlier South African cases, influenced by English law.

PAUSE FOR The concept of ‘duty of care’ in English law

REFLECTION This concept is described by Van der Walt and Midgley22 as:

Under the influence of the classic English doctrine of ‘duty

to take care’, courts have customarily described the duty as
a duty to take reasonable care, or to conform to a certain
standard of conduct. However, such an approach is not
tenable in terms of a theoretical structure of delict which
requires a distinction between the elements of
wrongfulness and fault. The duty to take care, or to act
reasonably, or not to act negligently, is a separate and
independent duty, concerned with establishing whether or
not the defendant was at fault, and which arises only after it
has been established that the defendant was in breach of a
legal duty not to harm the plaintiff. It is therefore incorrect
to express the legal duty in terms of a standard of care and
furthermore confusing to formulate of the standard of care
required in the particular case in terms of duty – for
example, [that a defendant has a legal duty to act without
negligence, or] that the motorist was under a duty to stop at
a stop-street.

In some earlier South African cases courts dealt with questions of wrongfulness by following the
‘duty of care’ approach of English law. Their view was that harm is not actionable unless the
defendant has a ‘duty of care’ not to cause harm, and that the duty of care is essentially based on
foreseeability. Boberg said that this contentious approach ‘brought trouble and strife’,23 and the
approach has attracted criticism on the basis that the concept of a ‘duty of care’ is an unnecessary
and misleading import from English law, which confuses negligence and wrongfulness.
A typical example of the earlier ‘duty of care’ approach of our courts is that of Schreiner JA in
Union Government v Ocean Accident & Guarantee Corporation Ltd.24 In this case, the Court
refused the claim of an employer for economic loss that resulted from the defendant negligently
injuring an employee. Schreiner JA said the following:
The expression ‘duty of care’ has sometimes been criticised as introducing an unnecessary
complication into the law of negligence, but, apart from the fact that it is endorsed by
considerable authority in this Court, it is so convenient a way of saying that it is the plaintiff
himself and no other, whose right must have been invaded by the careless defendant, that
the complication seems rather to be introduced by the effort to avoid its use. The duty of
care is in our case law rested upon foreseeability and this gives rise to a measure of
artificiality. But this is really unavoidable for, if there is to be control over the range of
persons who may sue, the test must be that of the reasonable man; what he would have
foreseen and what action he would have taken may not be calculable according to the actual
weighing of probabilities, but the device of reasoning on these lines helps to avoid the
impression of delivering an unreasoned moral judgment ex cathedra as to how the injurer
 should have behaved. The duty of care fits conveniently into the reasoning process and even
if it is no more than a manner of speaking it is a very useful one.

The ‘duty of care’ approach as set out in the Union Government case is based on the test of the
reasonable person. The test asks if harm was reasonably foreseeable and what action a reasonable
person would have taken to prevent harm. This enquiry into ‘duty’ is very similar to the test for
negligence. It uses the flexible concept of foreseeability, which, although it ‘may not be calculable
according to the actual weighing of probabilities’, allows courts to avoid ‘an unreasoned moral
judgment ex cathedra’. This approach combines wrongfulness and negligence, and uses the flexible
concept of foreseeability to cover value judgements and policy considerations that often remain
unexpressed.
In later cases courts moved away from applying the concept of a ‘duty of care’ in the context
of wrongfulness. Instead, they focused the enquiry on whether a ‘legal duty’ existed. However, the
current ‘legal duty’ terminology has not produced a uniform approach concerning the role of
foreseeability of harm in the enquiry into wrongfulness. Foreseeability of harm is a concept that is
central to negligence, and central to the ‘duty of care’ concept of English law. Using this concept to
determine wrongfulness tends to blur the distinction between these two elements of delict.
The judgment of Lewis JA in Premier, Western Cape v Faircape Property Developers (Pty)
Ltd is a more recent example of using foreseeability as a determinant of wrongfulness: 26
25

The foreseeability of harm to the plaintiff is also ‘a relevant consideration in the

determination of lawfulness’ … . Accordingly, even if it were to be found that the Minister’s
conduct had been negligent, this would not entail, necessarily, a finding that it was also
wrongful. One must ask whether it was wrongful … . In answering that question one must
consider also, therefore, whether the Minister should have foreseen that his conduct might
cause prejudice or loss … .

One of the enquiries, then, for determining whether the Minister was under a legal duty to
prevent harm … is whether the Minister should have foreseen that his conduct ‘might
endanger or prejudice others in regard to their legally protected interests’. A similar
question is inevitably repeated when one is determining the issue of negligence. In the
context of determining wrongfulness, the question relates only to whether there should be a
legal duty imposed on the Minister not to infringe a legal interest of an applicant. And it is
but one of several enquiries that must be pursued in order to determine whether, as a
matter of legal policy, an official or member of government should be visited with liability
for damages. Would a reasonable Minister have foreseen that an applicant for the removal
of restrictions would be prejudiced or would suffer loss if the application were granted?
Again, the answer must be no.

According to Lewis JA, the question of foreseeability is ‘inevitably repeated’ in the enquiries into
wrongfulness and negligence. For wrongfulness, ‘it is but one of several enquiries that must be
pursued’ to determine whether, as a matter of legal policy, liability for damages should be imposed.
She does not explain why foreseeability must be determined twice. Perhaps it is the same question
that is asked twice, or a ‘similar question’, or the same question is asked with a different purpose in
mind. She also does not explain how wrongfulness can properly indicate in which cases of
negligently caused harm liability should be imposed, if it shares a central attribute with negligence.
Another example of the foreseeability-based approach to wrongfulness is this passage in the
judgment of Scott JA in Gouda Boerdery BK v Transnet Ltd: 27
… Where the element of wrongfulness gains importance is in relation to liability for
omissions and pure economic loss. The inquiry as to wrongfulness will then involve a
determination of the existence or otherwise of a legal duty owed by the defendant to the
 plaintiff to act without negligence: in other words to avoid negligently causing the plaintiff
harm. This will be a matter for judicial judgment involving criteria of reasonableness, policy
and, where appropriate, constitutional norms. If a legal duty is found to have existed, the
next inquiry will be whether the defendant was negligent … . The courts have in the past
sometimes determined the issue of foreseeability as part of the inquiry into wrongfulness
and, after finding that there was a legal duty to act reasonably, proceeded to determine the
second leg of the negligence inquiry, the first (being foreseeability) having already been
decided. If this approach is adopted, it is important not to overlook the distinction between
negligence and wrongfulness.

According to Scott JA, the question of foreseeability is ‘sometimes’ determined as part of the
enquiry into wrongfulness. If the court finds there was a legal duty to act reasonably, the second
part of the negligence enquiry follows (the first part was deciding foreseeability). This seems to
indicate a split enquiry into negligence: the first part of the enquiry is combined with the enquiry
into wrongfulness, and the second part stands on its own. Scott JA does not explain why this makes
sense logically or what the practical benefits are.
In Telematrix (Pty) Ltd t/a Matrix Vehicle Tracking v Advertising Standards Authority SA28
Harms JA said the following in this regard:
But the fact that an act is negligent does not make it wrongful, although foreseeability of
damage may be a factor in establishing whether or not a particular act was wrongful. To
elevate negligence to the determining factor confuses wrongfulness with negligence and
leads to the absorption of the English law tort of negligence into our law, thereby distorting
it.

It appears from the extracts quoted that although the wrongfulness terminology has substantially
changed from ‘duty of care’ to ‘legal duty’, courts still sometimes regard reasonable foreseeability
as an indicator of a legal duty, and thus of wrongfulness. If reasonable foreseeability is a shared
attribute of wrongfulness and negligence, then these two elements of delict are at least partly
combined. However, this combination is denied in the judgments mentioned so far, which say that
although the elements of wrongfulness and negligence are intertwined, they are distinct. It seems
that the best explanation for this potentially confusing approach is that foreseeability is seen as a
factor that may be relevant in the wrongfulness enquiry, and as one of two core factors one must
consider in the negligence enquiry (the other being preventability). While foreseeability of harm is
a requirement for negligence, it might not be decisive in the wrongfulness enquiry and other factors
might override it. In some cases, foreseeability might add weight to the wrongfulness decision.

9.11 What is the content of the legal duty?

The concept of legal duty in case law is not uniform and its descriptions vary, for example, as a
duty not to cause harm, a duty to prevent harm and a duty to act reasonably. Some judgments and
academic comments say that where a breach of legal duty renders the negligent causing of harm
wrongful, the legal duty is a duty to act without negligence, or a duty not to be negligent. However,
to say that ‘where the breach of legal duty renders negligent conduct wrongful, the duty is a duty to
act without negligence’ is puzzling, because breach of a duty not to be negligent is nothing other
than negligence. The statement, in effect, says that negligence renders negligent conduct wrongful.
While it is true that liability based on negligence depends on an obligation not to be negligent,
duty is an indicator of wrongfulness in its fullest sense and is not confined to negligence. If liability
for negligence is in issue, the legal duty owed by the defendant to the plaintiff to act without
negligence means a legal duty to avoid negligently causing the plaintiff harm. Wrongfulness
assumes that all the other requirements for liability either have been met, or will be met. It seems
that the best way to describe the legal duty required for wrongfulness in its full sense is ‘the legal
duty not to cause harm negligently or intentionally’, or, in the case of strict liability, simply ‘not to
cause harm’.

9.12 Policy considerations

It is public policy that determines whether a legal duty for purposes of wrongfulness exists. In some
cases this involves considering the broad social and economic impact of imposing liability, and in
others, a more limited focus on legal and factual aspects of the relationship between the parties. In
each case, the question is whether it is reasonable for the law of delict to shift the burden of harm
from the plaintiff to the defendant. The policy considerations that the courts take into account
include the following (with illustrative cases):
• The social or economic consequences of imposing liability – in particular potential
indeterminate liability (‘opening the floodgates’):
◆ Shell and BP SA Petroleum Refineries (Pty) Ltd v Osborne Panama SA:29 The Court was
unwilling to recognise a legal duty towards the charterer of an oil tanker, who was one of
an indeterminate number of similarly placed potential claimants who suffered a loss when
there was a delay in offloading their cargo after the defendant had negligently damaged a
mooring buoy.
• The availability of alternative remedies – in particular a contractual remedy:
♦ Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd:30 The Appellate
Division held that a legal duty in delict did not fit comfortably into a detailed business
contract for professional services, inter alia, because recognising an action in delict could
mean contractual terms being avoided and could create ‘a trap for the unwary’ within the
contractual relationship. The denial of this action was limited to the case where the
alleged negligence consists of a breach of contract.
♦ Pinshaw v Nexus Securities (Pty) Ltd:31 The Court held that the contract between an asset
management company and its client did not rule out a legal duty in delict on the part of
the company’s employee towards the client.
♦ Holtzhausen v ABSA Bank Ltd:32 The Court held that an action is maintainable in delict for
a negligent misstatement causing pure pecuniary loss, even if a concurrent action is
available in contract.
♦ Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd:33 The Court held
that public policy does not require extension of the Aquilian action to rescue a plaintiff
who should have avoided risk of harm by contractual means, but did not.
♦ Freddy Hirsch Group (Pty) Ltd v Chickenland (Pty) Ltd:34 The Court held that the plaintiff
had an alternative claim in delict for economic loss caused by delivery to it of spices
containing a banned colourant. The Court accepted that the same facts may give rise to a
claim for damages ex contractu and alternatively ex delicto, but also that the breach of a
contractual duty is not per se wrongful for the purposes of Aquilian liability.35 The
negligent causing of pure economic loss is not prima facie wrongful, but in this case,
policy considerations, such as knowledge of the potential harm, the fact that the loss was
single and determinate, and the manufacturer’s general duty to take reasonable steps to
ensure that defective products do not reach the market, indicated wrongfulness for the
purposes of delictual liability.
• The need for accountability of state departments, public bodies and officials, which also
requires considering that potential liability could hamper persons or bodies in exercising
functions in the public interest, as in the case of tender boards or administrative tribunals:
♦ Steenkamp NO v The Provincial Tender Board of the Eastern Cape:36 The Court refused to
recognise a legal duty on the part of the tender board towards a successful tenderer, whose
tender award was later set aside because of negligence in the process of awarding the
tender. The Court took into account, inter alia, whether imposing liability for damages
would have a ‘chilling effect’ on the performance of administrative or statutory functions
by members of the board.
♦ Olitzki Property Holdings v State Tender Board:37 In this case, the claim of an
unsuccessful tenderer also failed. However, the result may be different if there is fraud in
the process of awarding a tender.
♦ Minister of Finance v Gore NO:38 The Court held that negligent conduct that caused pure
economic loss was wrongful only if, as a matter of legal and public policy, a legal duty
existed,39 and that there were no public or legal policy considerations that dictated that the
State should not be liable for the provincial officials’ fraudulent conduct in processing a
public tender.
♦ Telematrix (Pty) Ltd t/a Matrix Vehicle Tracking v Advertising Standards Authority SA:40
The Court held that the Advertising Standards Authority of SA (the ASA) did not owe a
legal duty towards an advertiser who had suffered a loss because of an incorrect decision
by one of the ASA organs. The relevant policy consideration is the protection of the
independence of persons or bodies with an adjudicative function to serve the public
interest, which imposes on them a duty to act impartially. These persons or bodies
(including the judiciary, arbitrators and other administrative tribunals) should be able to
adjudicate fearlessly. The threat of an action for damages could interfere with their
prompt dealings of litigation and disputes. Although both the person harmed and the loss
suffered due to an incorrect decision are foreseeable, negligently causing harm is not
considered wrongful.
♦ South African Hang and Paragliding Association v Bewick: 41 The respondent was
seriously injured while taking a tandem paragliding flight. She settled claims against the
pilot and his employer, leaving the Court to decide her claims against the two appellants,
the South African Hang and Paragliding Association and the South African Civil Aviation
Authority, who exercised direction and control over paragliding in South Africa. Tandem
paragliding for reward is illegal and the two appellants were aware that such illegal
activity was going on. The question was whether the appellants were under a legal duty to
take reasonable steps to terminate and prevent this illegal activity, but had negligently and
wrongfully failed to do so. The question of wrongfulness depended on whether it would
be reasonable to impose legal liability on the appellants. The Court decided it would not
be, referring to the following factors:
(a) The appellants were obliged by statute to ensure and promote the safety of civil aviation, but it
would not be reasonable to impose liability upon them for an omission which had no direct
impact on aviation safety;
(b) Identification of offending pilots would require widespread control and investigation by
inspectors appointed by the appellants, but there was insufficient evidence on available
resources, and inherent probabilities indicated that these extensive measures of control would
be unaffordable;
(c) Courts do not extend the scope of the Aquilian action to new situations unless there are
positive policy considerations which favour an extension;
(d) Courts avoid imposing liability if there is the apprehension of boundless liability; and
(e) The plaintiff had an alternative remedy: she could and did sue the pilot and his employer.
♦ Oppelt v Department of Health, Western Cape: 42 The plaintiff sustained a spinal cord
injury during a rugby match. Over the few hours following the injury he was treated at
 three hospitals under the control of the Department of Health. Eventually he was left
paralysed below his neck, and medically classified as quadriplegic. He claimed damages
for negligence arising from the medical treatment. His claim was based on the failure of
the three hospitals to provide him with prompt and appropriate medical treatment. On
wrongfulness the Court held that the legal convictions of the community demanded that
hospitals and health-care practitioners must provide proficient health-care services to
members of the public. Those convictions also demanded that those who failed to do so
must incur liability. Proficient health-care entailed providing urgent and appropriate
emergency treatment whenever a medical condition required it.
♦ Country Cloud Trading CC v MEC, Department of Infrastructure Development: 43 The
Department of Infrastructure Development entered into a building contract with iLima.
When the project ran into difficulties iLima borrowed R12 million from Country Cloud.
The loan agreement was subject to the condition that the Department would repay
Country Cloud's R12 million out of the amount payable by the Department to iLima in
terms of the building contract. The Department later cancelled the building contract,
thereby committing breach of contract, and this resulted in iLima’s liquidation and a loss
to Country Cloud, who then sued the Department in delict. The central issue was whether
the Department wrongfully caused harm to Country Cloud. Country Cloud's claim was for
pure economic harm, and Country Cloud could not show that the Department had
wrongfully infringed its rights or had a legal duty not to cause it economic harm. Country
Cloud relied on state accountability, but this consideration does not always give rise to a
private-law duty, particularly if, as in this case, the Department did not act dishonestly or
corruptly. Also, Country Cloud could have attempted to take steps to protect itself against
non-payment by iLima, by claiming repayment from iLima's liquidator under the loan
agreement, or taking cession of iLima's claim for breach of contract against the
Department, or by calling for payment from a surety.
• Constitutional rights may, in the context of the law of delict, imply a legal duty not to cause
harm or to prevent harm to another person. Rights that may imply such a duty include the
sanctity of life, the rights to freedom and security of the person, the right to privacy, and the
right to freedom of expression. Duties may be imposed on state officials to protect these
constitutional rights:
♦ Carmichele v Minister of Safety and Security: 44 The Court based liability on the duties of
the police and prosecutors who released a person, on his own recognisance, with a prior
conviction for violence. This person assaulted the plaintiff while he was awaiting trial on
a new charge.45
♦ Minister of Safety and Security v Van Duivenboden: 46 The Court held that the police failed
in their duty to withdraw the firearm licence of a person prone to violence when drunk.
♦ Minister of Safety and Security v Hamilton: 47 The Court held the police liable for failing
to enquire into the psychological fitness of the applicant for a firearm licence.
♦ Van Eeden v Minister of Safety and Security: 48 The Court held the police liable for
allowing a prisoner with a history of violent crime to escape, who subsequently injured
the plaintiff.
♦ Loureiro v iMvula Quality Protection (Pty) Ltd: 49 A private security firm, iMvula, was
held liable in delict for whatever damages might be proved, because one of their security
guards allowed robbers masquerading as policemen to enter premises, where they robbed
the plaintiffs. Wrongfulness in delict is determined according to the legal convictions of
the community, by weighing competing norms and interests. The convictions of the
community are underpinned and informed by the norms and values of society, embodied
in the Constitution. The Court had to determine the convictions of a community plagued
 by crime, on the issues of respect for the police and their role and interaction with the
ever-growing private security industry. The case posed questions about the interpretation
and development of the common law and thus raised constitutional issues concerning the
public role that security companies play in giving effect to fundamental rights. In this
regard Van der Westhuizen J concluded:50
There are ample public-policy reasons in favour of imposing liability. The
constitutional rights to personal safety and protection from theft of or damage
to one’s property are compelling normative considerations. There is a great
public interest in making sure that private security companies and their guards,
in assuming the role of crime prevention for remuneration, succeed in thwarting
avoidable harm. If they are too easily insulated from claims for these harms
because of mistakes on their side, they would have little incentive to conduct
themselves in a way that avoids causing harm. And policy objectives (such as
the deterrent effect of liability) underpin one of the purposes of imposing
delictual liability. The convictions of the community as to policy and law clearly
motivate for liability to be imposed.

• Certain factual circumstances may indicate a duty not to cause harm or to prevent harm,
including: proportionality of the risk of harm and the cost of prevention, control over a
dangerous object or situation, awareness of danger, prior conduct creating danger, a
relationship imposing responsibility, and professional knowledge.
♦ Administrateur, Transvaal v Van der Merwe: 51 This case illustrates the process or
reasoning involved where liability for an omission is in issue. The Court determined
whether there was a legal duty to prevent harm by enquiring into the proportionality of the
risk of harm and the cost of prevention. The question was whether provincial authorities
had a duty, in respect of a minor road, to make firebreaks or to take other precautionary
measures against fires breaking out and spreading to adjoining land. The Court held that
to determine whether a positive act or an omission is wrongful, it should weigh up, inter
alia, the different interests of the parties, their relationship with one another and the social
consequences of imposing liability in the type of case in question. Factors that play an
important role in this process are, inter alia, the probable or possible extent of prejudice to
others, the degree of risk of such prejudice occurring, the interests that the defendant or
the community, or both have in the act or omission in issue, whether there were
reasonable measures with which the defendant could avoid the prejudice, what the
chances were of the measures being successful, and whether the cost involved in taking
such measures was reasonably proportional to the damage that the plaintiff could suffer.
One should consider the affordability and proportionality between the potential damage
and the potential cost of prevention when deciding the question of wrongfulness. On
applying this proportionality test, the Court found that the Administrator’s control and
supervision over all public roads was only of a permissive nature. This meant that under
the applicable legislation, the Administrator was not obliged to make firebreaks or take
other precautionary measures against veld fires breaking out and spreading to adjoining
land. Given the nature of the road in question, the fact that it was seldom used, and the
cost of preventative measures, the mere fact that the Administrator exercised control and
supervision over all public roads did not in itself create a duty for purposes of delictual
liability. Without a positive danger-creating act, being in control of property and failing to
exercise this control, resulting in prejudice to another, is not per se wrongful. The crucial
issue is whether the precautionary measures that the controller should, according to the
aggrieved party, have taken to prevent the prejudice can in the circumstances be
reasonably and practicably required of him. The underlying philosophy is that a
 consequence is only wrongful if, after considering all circumstances, the defendant can be
reasonably expected to act.
♦ Za v Smith:52 This case involved a claim by dependents of a person who slipped on a
snow-covered mountain slope and fell over a sheer precipice to his death. The incident
occurred on a farm owned by the first respondent, where the second respondent conducted
the business of a private nature reserve for gain. In substance, the claim was based on
delictual liability arising from the wrongful and negligent failure by the first and second
respondents to take reasonable steps to avoid the incident that led to the death of the
deceased.
♦ On the issue of wrongfulness, Brand JA concluded:53
…In determining wrongfulness, the other elements of delictual liability are
usually assumed. Hence the enquiry is whether – on the assumption (a) that the
respondents in this case could have prevented the deceased from slipping and
falling to his death; and (b) that he had died because of their negligent failure to
do so – it would be reasonable to impose delictual liability upon them for the
loss that his dependants had suffered through their negligence. While denying,
of course, that these assumptions could validly be made, Counsel for the
respondent conceded that, if they were true, the answer to the question posed
must be ‘yes’. I believe that this concession was rightly and fairly made. Apart
from the fact that both respondents were in control of a property, which held a
risk of danger for visitors, the second respondent, with the knowledge and
consent of the first respondent, as owner of the property, allowed members of
the public, for a fee, to make use of a four-wheel drive route, designed to lead
directly to the area which proved to be extremely dangerous.

• The nature of the defendant’s conduct is taken into account. Causing harm by positive conduct
is more often regarded as wrongful compared to causing harm by omission (taking into
account the context in which the conduct occurred):
♦ BOE Bank Ltd v Ries: 54 The Supreme Court of Appeal refused to recognise an insurance
broker’s legal duty towards an intended beneficiary under a life insurance policy. The
broker did not make sure that the policy holder signed the necessary form to nominate the
beneficiary. The Court took into account that the broker’s conduct was not an assumption
of any professional responsibility regarding signing the form. The broker had become
involved in the process in passing, by doing a favour for a colleague. In effect, he acted as
a messenger, without undertaking any professional responsibility. Although it was
foreseeable that the intended beneficiary would not benefit from the policy if the holder
failed to sign the necessary nomination form before he died, foreseeability of loss was not
in itself enough to indicate a legal duty.
• The nature of the interest to be protected is taken into account. Courts more readily recognise a
duty in respect of physical injury and damage to property than a duty in respect of pure
economic loss:
♦ Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd: 55 The Appellate
Division held that a legal duty in delict did not fit comfortably into a detailed business
contract for professional services, inter alia, because the harm caused by the negligent
performance of the services was pure economic loss.

Table 9.2 Categories of policy factors relevant to wrongfulness

Category Factors

• Plaintiff’s interests
 Factors that are always • Defendant’s interests
present • Nature of the conduct
• Nature of the loss
◆ Any reasons of policy why a remedy should not be granted in the circumstances, for
example, availability of other remedies in contract or administrative law, and protection of
independent decision-makers

Factors that depend on • The seriousness of the injury and the size of the claim
the nature of the facts • Foreseeability of any harm
• Expert knowledge
• Social consequences
• Whether reasonably practical measures were available to the defendant to avoid the loss and
the chances of their success
◆ Administrative convenience
◆ Multiplicity of actions
◆ Direct and finite loss
◆ Relationship between parties
◆ Society’s ideas of morals and justice
• The extent of the prejudice to either of the parties and the risk thereof
• The costs involved in taking preventative steps and whether or not they were proportionate to
the loss incurred

•The nature of the defendant’s fault and state of mind (motive) is taken into account. Courts more
readily consider intentional harm-causing to be wrongful than negligent harm-causing. A
motive to cause harm will indicate wrongfulness.56 We deal with these matters in the next
section on wrongfulness and fault:
♦ Minister of Finance v Gore NO: 57 The Court held that fraudulent conduct in processing a
public tender that caused pure economic loss was wrongful, whereas negligent causing of
harm would not necessarily be wrongful.

9.13 Is wrongfulness determined with hindsight (ex post

facto) or from the perspective of the defendant at the
time of the relevant conduct with foresight (ex ante)?
It is somewhat contentious whether the enquiry into wrongfulness involves:

• A strictly ex post facto perspective and excludes an ex ante or actor-oriented perspective

• That courts should determine wrongfulness or fault first
• That the reasonableness of conduct is a factor relevant to wrongfulness.

9.13.1 Involvement of a strictly ex post facto perspective and exclusion of

an ex ante or actor-oriented perspective
We contend that both wrongfulness and fault are determined ex post facto. Since the events have
already occurred, the requirements for both these elements of the delict are determined from the
perspective of hindsight, but with a different focus. The wrongfulness enquiry necessitates that
issues also be considered that were not known to the defendant at the time, whereas the essence of
the negligence enquiry is to assess the conduct bearing in mind the defendant’s circumstances at the
time.
The focus of wrongfulness is wide and includes all the other elements of liability (conduct,
causation, affected interest, harm and fault), and finally it involves policy considerations and a
value judgement. Therefore, courts reach the final value judgement on whether the plaintiff’s
affected interest deserves protection from the defendant’s action or lack of action with an overall ex
post facto perspective.
 Fault is mainly concerned with the blameworthiness of the conduct of the actor. The question
asked is whether the defendant intended to cause the harm knowing that it was wrong to cause the
harm (intent), or, if there was no intention to harm the plaintiff, did the defendant behave
reasonably, and at that point, was the harm from the perspective of the actor, reasonably
foreseeable and preventable (negligence)? This enquiry requires an ex ante perspective.

9.13.2 Determining wrongfulness or fault first

There is no rule on whether to determine wrongfulness or fault first. Occasionally courts deal with
negligence first and a finding that the defendant was not negligent means that an enquiry into
wrongfulness is unnecessary. This process does not mean that either wrongfulness or fault logically
precedes the other. Both enquiries require the decision-maker to assume that the other element must
also be proven before liability can be imposed. One can determine fault after establishing or by
assuming that the defendant has wrongfully caused harm. If this assumption is unfounded, either
because the defendant did not cause the harm or because the harm caused was reasonable and
therefore not wrongful, the ‘fault’ is simply irrelevant or ‘legally neutral’. Likewise, courts can
determine wrongfulness on the assumption that there is fault or a particular form of fault, simply
because they find it convenient to deal with the wrongfulness issue first. If fault is a requirement for
liability, and it turns out there is no fault, assessing wrongfulness then becomes irrelevant. In Local
Transitional Council of Delmas v Boshoff 58 the Court said the following about the order and
manner of enquiry into the elements of wrongfulness and fault:
Depending on the circumstances it may be appropriate to enquire first into the question of
wrongfulness, in which event it may be convenient to assume negligence for the purpose of
the inquiry … . On the other hand, it may be convenient to assume wrongfulness and then
consider the question of negligence … .

9.13.3 The nature of fault is in some cases relevant to wrongfulness

The nature of the fault (in particular intent) and fault-related factors (in particular a motive to cause
harm) may be relevant to wrongfulness. Intentionally causing harm is usually not of value to
society, and is therefore likely to be wrongful.59 The wrongfulness of harm-causing
misrepresentation, for example, can depend on whether this occurred intentionally or negligently.
In Minister of Finance v Gore NO 60 the Court said:61
We do not think that it can be stated as a general rule that, in the context of delictual
liability, state of mind has nothing to do with wrongfulness. Clear instances of the contrary
are those cases where intent, as opposed to mere negligence, is itself an essential element of
wrongfulness. These include intentional interference with contractual rights … and
unlawful competition … .

Intentionally causing harm to others will not always be wrongful, for example, where justified
criticism harms the reputation of another, or fair competition causes financial loss to a trade
competitor. However, everything depends on the circumstances and intent does not necessarily
indicate wrongfulness.

PAUSE FOR Intentional causing of harm

REFLECTION Consider a passenger on a bus who knowingly offers bad investment
advice to other passengers, who are complete strangers to him. Will the
advisor be liable if his bad advice is followed with harmful effect? Does
he owe a duty towards his fellowpassengers in the area of investment
 advice? Does the intentional causing of harm indicate wrongfulness in
this case? On what grounds would you argue that the advisor owes no
duty to his fellow passengers?

An improper motive to cause harm to another in itself does not necessarily make causing the harm
wrongful. However, courts may take into account motive, together with other circumstances, when
assessing wrongfulness. For example, when a landowner exercises his or her rights of ownership in
a way meant to cause extensive harm to a neighbour, with little benefit to himself or herself, the
motive to harm may influence a court to decide that the harm was caused wrongfully. In Gien v
Gien 62 the Court considered causing excessive noise to deter animals and birds from damaging
crops, with a motive to annoy a neighbour, to be wrongful. In Kirsh v Pincus 63 the Court also
deemed it wrongful to plant deciduous trees along a boundary so that falling leaves harmed a
neighbour. In the area of unlawful competition, courts also take into account a motive to harm
rather than to compete, when determining wrongfulness.64 The exercise of rights which causes harm
to another can be a wrongful abuse of a rights if the sole or predominant intention was to harm
another and the act did not advance any appreciable or legitimate interest of the actor.65 These
examples all indicate some form of abuse of rights. In other words, behaviour that one would
normally consider reasonable becomes unreasonable because the person exercised the right for
what society considers an unacceptable purpose.
In another category of cases – malicious detention and malicious prosecution – an improper or
malicious motive is a prerequisite for a wrongfulness finding. In such instances, society considers
infringing another’s interest wrongful only if the defendant behaved maliciously. So, causing harm
by a careless detention or institution of a prosecution will not be wrongful, which is yet another
instance of where society protects the independence of a public functionary.
In Cape Empowerment Trust Limited v Fisher Hoffman Sithole 66 the purchaser of a business
relied on a certificate issued by the seller’s auditor, confirming that the business had made a profit
of R10 million. It turned out that this was entirely untrue and that the auditor had been grossly
negligent. On the issue whether the auditor owed a legal duty to the purchaser and had wrongfully
caused the purchaser’s economic loss, the Court decided that it was impermissible to take into
account the auditor’s gross negligence as a policy consideration indicating a legal duty and
wrongfulness. This would telescope the tests for wrongfulness and negligence into one. In the law
of delict in general and in the context of negligent misstatements in particular, the element of
wrongfulness introduces a measure of control. It serves to exclude liability in situations where most
right-minded people, including judges, would regard the imposition of liability as untenable,
despite the presence of all other elements of liability, including gross negligence.
Blameworthiness of conduct indicated by intent or a motive to harm is part of the mix of
factors that one takes into account when determining the question of wrongfulness.

9.13.4 Wrongfulness and negligence are sometimes based on similar

factors
Cases of liability for an omission, where there is negligent failure to prevent harm, illustrate that the
elements of wrongfulness and negligence often involve considering similar factors. If the question
is whether a local authority is liable for harm caused by, or related to public property, for example,
a hole in a pavement injuring a pedestrian,67 a broken merry-go-round injuring a child,68 a shop-
owner’s loss due to flooding from a burst pipe,69 or a land-owner’s losses due to an informal
settlement being established next to his or her land,70 the wrongfulness and the negligence issues
both require us to consider the extent of the risk and the possible harm, the cost of repair or
prevention, and the resources available to the local authority. If the pavement, merry-go-round,
pipe or settlement was under the control of the local authority, with concomitant responsibility to
maintain or administer it, the legal duty not to cause harm (wrongfulness) will generally not be
contentious. However, courts will not impose liability if the local authority was unaware of the
state of disrepair and could not reasonably have foreseen it. This could be due to the informational
and organisational constraints under which it operates, or if the costs of preventative measures
would be out of all proportion to the harm. In these circumstances, failing to repair or to prevent
harm is not negligent. If the issue is whether the public authority had the duty to exercise control,
rather than the extent of the control or preventative measures required, courts will likely treat it as
an issue of wrongfulness rather than an issue of negligence.
Therefore, it appears that there is some overlap between wrongfulness and fault. This does not
indicate a logical grey area or lack of clear definition of the elements of delict. Courts must
determine liability rationally and consistently, and a certain amount of overlap does not in itself
detract from either rationality or consistency. The aim is not to develop a theory of delict that is
made up of elements that fit into perfectly separate compartments. Wrongfulness and fault have
broadly different focus areas (in the case of fault: blameworthiness of conduct, and in the case of
wrongfulness: overall balance of interests and the scope of responsibility), but there are also
common areas. Both these elements of delict are based on reasonableness and involve value
judgements.

9.14 Conclusion
Wrongfulness is a matter of law. Courts do not hear evidence on the issue of wrongfulness.
However, the onus is on the plaintiff to make factual allegations and eventually present evidence
that indicates wrongfulness, which is a matter for judicial determination.
It is generally accepted that applying the general criterion or standard for determining
wrongfulness (alternatively referred to as general reasonableness, the legal convictions prevailing
in the community or the boni mores) in the final instance involves public policy and a value
judgement. When applying this general criterion, one can reduce judicial reasoning to the following
factors, which are often interrelated:
• Policy considerations that indicate whether the law of delict should intervene in respect of the
type of harm-causing (inter alia the social or economic consequences of imposing liability – in
particular potential indeterminate liability (‘opening the floodgates’), the availability of
alternative remedies, and the need for accountability of public bodies or officials)
• Consideration of relevant constitutional or other statutory rights and duties (inter alia the right
to freedom and security of the person, the right to privacy and the right to freedom of
expression)
• A grouping of factual circumstances that indicate a duty not to cause or to prevent harm in the
particular situation (inter alia the proportionality of the risk of harm and the cost of prevention,
control over a dangerous object or situation, awareness of danger, prior conduct creating
danger, a relationship imposing responsibility and professional knowledge)
• The nature of the defendant’s conduct (courts usually consider harm-causing by positive
conduct more wrongful than harm-causing by omission, and physical harm-causing more
wrongful than harm-causing by speech)
• The nature of the defendant’s fault and state of mind (courts usually consider intentional harm-
causing more wrongful than negligent harm-causing, and that a motive to harm is indicative of
wrongfulness)
• The nature of the interest to be protected or the consequences that resulted (causing physical
injury and damage to property is prima facie wrongful, while causing pure economic loss or
 emotional distress is not, and in the case of certain forms of harm, such as nuisance and
damage to reputation, courts judge wrongfulness by balancing conflicting interests).

The prevailing modern view is that wrongfulness is not only an attribute of conduct, but
characterises the outcome of a causal sequence that involves conduct and causing harm.
An enquiry into wrongfulness can focus either on the infringement of a right, or on the breach
of a duty. This is a matter of approach or methodology, and whatever approach one follows, the
conclusion often involves public policy and a value judgement. Breaching a legal duty to another
also involves infringing the right of the other person not to be harmed. To every obligation there is
a right and a duty side, and a legal duty is the converse of a right. The best description of the legal
duty that indicates wrongfulness in its full sense is the legal duty not to cause harm negligently,
intentionally, or without fault (in cases of strict liability).
Although the wrongfulness terminology has changed from the negligence-related ‘duty of
care’ to ‘legal duty’, courts still regard reasonable foreseeability of harm as one of the factors that
may indicate whether a legal duty for the purposes of wrongfulness exists. It seems that the best
explanation for this potentially confusing approach is that foreseeability is seen as a factor that one
may consider in the wrongfulness enquiry, together with other relevant factors, but it is one of two
core factors that one must consider in the negligence enquiry (the other being preventability). While
foreseeability of harm is a requirement for negligence, it might not be decisive in the wrongfulness
enquiry and other factors might override it. In some cases, foreseeability might add weight to the
wrongfulness decision.
There is no rule on whether one should determine wrongfulness or fault first. Occasionally,
courts deal with negligence first, and finding that the defendant was not negligent means that
enquiry into wrongfulness is unnecessary. This order does not mean that either wrongfulness or
fault logically precedes the other. One can determine fault after proving, or assuming, that the
defendant has wrongfully caused harm. Likewise, courts can determine wrongfulness on the
assumption that there is fault or a particular form of fault, simply because they find it convenient to
deal with the wrongfulness issue first.

Figure 9.3 Wrongfulness

Wrongfulness supplements the other elements of delict, adding a further value, or a policy-based
dimension to the enquiry into liability, and requires judicial discretion. With all the other elements
of liability (conduct, causation, harm and fault) proved or assumed to be present, wrongfulness
involves a value judgement on whether the affected interest of the plaintiff should prevail over a
conflicting interest of the defendant, or deserves protection from the defendant’s action or lack of
it, so that the burden of damage should be shifted from plaintiff to defendant. Wrongfulness is thus
essentially concerned with the scope of protection afforded to various rights and interests, the scope
of responsibility to act, and overall policy considerations that relate to the question of whether the
law of delict should intervene.

1 See Fagan ‘Rethinking wrongfulness in the law of delict’ (2005) 122(1) SALJ at 90; Neethling ‘The conflation of
wrongfulness and negligence: Is it always such a bad thing for the law of delict?’ (2006) 123(2) SALJ at 204; Nugent
‘Yes, it is always a bad thing for the law: A reply to Professor Neethling’ (2006) 123(4) SALJ 557 at 560; Neethling
and Potgieter ‘Wrongfulness and negligence in the law of delict: A Babylonian confusion?’ (2007) 70(1) THRHR at
120.
2 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 32.
3 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amicus Curiae) 2011 (6) BCLR
577 (CC); 2011 (3) SA 274 (CC) para 122 (footnotes omitted).
4 2015 (1) SA 1 (CC) paras 20–21.
5 Walker The Law of Delict in Scotland 2 ed (1981) at 33.
6 2013 (5) SA 183 (SCA) para 23. See also Mukheiber v Raath 1999 (3) SA 1065 (SCA) para 25.
7 2005 (6) SA 215 (SCA).
8 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC)
para 7; Cape Town Municipality v Bakkerud 2000 (3) SA 1049 (SCA) paras 14–17; Minister of Safety and Security v
Van Duivenboden 2002 (6) SA 431 (SCA) para 16.
9 Sanlam Capital Markets (Pty) Ltd v Mettle Manco (Pty) Ltd [2014] 3 All SA 454 (GJ).
10 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) paras 13–15.
11 Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd 2006 (3) SA 138 (SCA) paras 5 and 10.
12 Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A) at 498.
13 Coronation Brick (Pty) Ltd v Strachan Construction Co (Pty) Ltd 1982 (4) SA 371 (D).
14 Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd 2006 (3) SA 138 (SCA).
15 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 188.
16 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 22.
17 Cape Town Municipality v Bakkerud 2000 (3) SA 1049 (SCA).
18 Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA).
19 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
20 Pinshaw v Nexus Securities (Pty) Ltd 2002 (2) SA 510 (C).
21 Van Wyk v Lewis 1924 AD 438.
22 See, generally, Van der Walt and Midgley (2005) para 63; Van der Walt and Midgley Principles of delict 3rd ed.
2005. Lexis Nexis. Reprinted by kind permission of Lexis Nexis.
23 Boberg (1984) at 30–31.
24 1956 (1) SA 577 (A) at 585B–D.
25 2003 (6) SA 13 (SCA).
26 Paras 42 and 46 (footnotes omitted).
27 2005 (5) SA 490 (SCA) para 12 (footnotes omitted).
28 2006 (1) SA 461 (SCA) para 12.
29 1980 (3) SA 653 (D).
30 1985 (1) SA 475 (A).
31 2002 (2) SA 510 (C).
32 2008 (5) SA 630 (SCA).
33 2006 (3) SA 138 (SCA).
34 2011 (4) SA 276 (SCA).
35 Para 33.
36 2007 (3) SA 121 (CC).
37 2001 (3) SA 1247 (SCA).
38 2007 (1) SA 111 (SCA).
39 Para 82.
40 2006 (1) SA 461 (SCA) para 12.
41 2015 (3) SA 449 (SCA).
42 2016 (1) SA 325 (CC).
43 2015 (1) SA 1 (CC).
44 2004 (3) SA 305 (SCA) at 311.
45 See also Minister of Justice and Constitutional Development v X [2014] 4 All SA 586 (SCA); 2015 (1) SA 25 (SCA)..
46 2002 (6) SA 431 (SCA).
47 2004 (2) SA 216 (SCA).
48 2003 (1) SA 389 (SCA).
49 2014 (5) BCLR 511 (CC).
50 Para 56 (footnotes omitted).
51 1994 (4) SA 347 (A) at 361H–362A/B and 363C.
52 2015 (4) SA 574 (SCA).
53 Para 21.
54 2002 (2) SA 39 (SCA) paras 13–26.
55 1985 (1) SA 475 (A).
56 Minister of Finance v Gore NO 2007 (1) SA 111 (SCA) para 86.
57 2007 (1) SA 111 (SCA).
58 2005 (5) SA 514 (SCA) para 20.
59 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 40.
60 2007 (1) SA 111 (SCA).
61 Para 86 (footnotes omitted).
62 1979 (2) SA 1113 (T) at 1121.
63 1927 TPD 199.
64 See Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty) Ltd 1991 (2) SA 455 (W).
65 Koukoudis v Abrina 1772 (Pty) Ltd 2016 (5) SA 352 (SCA) para 31.
66 2013 (5) SA 183 (SCA) paras 24–25.
67 Cape Town Municipality v Bakkerud 2000 (3) SA 1049 (SCA).
68 Cape Town Municipality v April 1982 (1) SA 259 (C).
69 Mostert v Cape Town City Council 2001 (1) SA 105 (SCA).
70 Local Transitional Council of Delmas v Boshoff 2005 (5) SA 514 (SCA).
 Chapter 10

Grounds of justification: Defences

directed at the wrongfulness element

10.1 Introduction

10.2 Consent

10.3 Consent by assumption of risk

10.4 Prior agreement not to claim (pactum de non petendo in anticipando)

10.5 Necessity

10.6 Self-defence (private defence)

10.7 Provocation

10.8 Statutory authority |

10.9 Official capacity

10.10 Obedience to orders

10.11 Disciplinary powers

10.12 Impossibility

10.1 Introduction
Grounds of justification are special circumstances that make the factual violation of a right or
breach of a duty, reasonable and therefore lawful.1 The violation or breach would be wrongful
without these special circumstances. Grounds of justification are simply applications of the general
criterion of reasonableness to certain typical situations. For example, if one defends oneself against
a wrongful attack on one’s person or property, and in the process one injures the attacker, the injury
may be justified on the ground of private defence. In Clarke v Hurst NO, 2 grounds of justification
were described as follows:
The stereotyped grounds of justification are specific grounds of justification of otherwise
wrongful conduct which with the passage of time have become crystallised, with their own
rules limiting the scope of their application.

Grounds of justification deal with certain stereotyped situations where applying the general
criterion of reasonableness has become so standardised that certain sub-rules, or a framework for
applying the general criterion, have developed. There is no closed list of grounds of justification,
because these grounds merely represent applications of the general criterion of reasonableness,
which justify an infringement of a right or breach of duty that would otherwise be wrongful.
The onus of proving that the violation or breach is not wrongful is on the defendant, who has
caused harm by what appears to be an infringement of a right or breach of a duty. Therefore, the
defendant needs to justify causing such harm.3
The most common grounds of justification are: consent, private defence, necessity,
provocation, statutory authority, public authority, official command, disciplinary authority and
impossibility. We discuss each one of these in the following sections.

10.2 Consent
Where a person capable of expressing his or her will indicates to another person that he or she is
willing to suffer some harm or to run the risk of suffering some harm, for a lawful purpose, then the
causing of such harm by the other person is justified. Consent is a ground of justification based on
one of two arguments:
1. It involves a waiver of rights in respect of the harm concerned.
2. The causing of harm is reasonable in terms of the boni mores or the legal convictions of the
community where the person affected has indicated consent or willingness to suffer harm for a
lawful purpose.

One could also say that consent is a ground of justification because a legal duty not to cause harm
does not arise when the harm has been consented to.
In Roman and Roman-Dutch law the principle that consent can justify causing harm was
expressed in the maxim volenti non fit iniuria. South African law continues to apply this maxim
and its principle not only to intentionally causing specific harm for a lawful purpose, as in the case
of a medical operation, but also to accepting a risk of harm that may be negligently caused in the
course of a dangerous activity, such as taking part in a contact sport.4 In the latter case, consent is
also sometimes referred to as voluntary assumption or acceptance of risk.
Voluntarily taking part in a dangerous activity does not necessarily constitute consent to suffer
harm resulting from the negligence of others who also take part in the dangerous activity. However,
it may nevertheless constitute contributory negligence in respect of the resultant harm. Consent on
the part of a plaintiff is a ground of justification, and therefore a complete defence, which excludes
liability on the part of the defendant who caused the harm. However, contributory negligence on
the part of the plaintiff tends to reduce the damages to which the plaintiff is entitled as a result of
the harm caused by both the plaintiff’s and the defendant’s negligent conduct.5
Consent to the intentional causing of harm for a lawful purpose, as in the case of a medical
operation, involves a willingness to suffer specific harm. An example is the willingness that a
surgeon remove part of an organ because of cancer, together with acceptance of the pain and
inconvenience that accompany such an operation. Consent to the risk of harm is less specific. It
involves a willingness to risk suffering some harm during a dangerous activity, such as a sport that
involves the risk of injury.6 A particular situation may give rise to both forms of consent – for
example, a medical operation that involves not only some pain and inconvenience, but also the risk
of complications, or even death.7 A patient who is adequately informed of the risks and nevertheless
decides to undergo the operation, consents to the pain and inconvenience that will inevitably occur,
but also to the possibility of complications or death that may occur, but which hopefully will not.
The following is a well-known extract from the judgment of Innes CJ in Waring & Gillow Ltd
v Sherborne,8 which sums up the nature and requirements of the defence of volenti non fit injuria:
The maxim volenti non fit injuria embodies a principle which, when confined within right
limits, is both just and equitable. A man who consents to suffer an injury can as a general
rule have no right to complain. He who, knowing and realising a danger, voluntarily agrees
to undergo it, has only himself to thank for the consequences. But like so many other
maxims, the one under consideration needs to be employed cautiously and with
circumspection. The principle is clear; the difficulty lies in the application of it – in deciding,
in other words, under the circumstances of each particular case whether the injured man
was volens to undertake the risk. A consideration of the grounds upon which the doctrine
rests, and of the cases in which its scope has been discussed, leads to the conclusion that in
order to render the maxim applicable it must be clearly shown that the risk was known, that
it was realised, and that it was voluntarily undertaken. Knowledge, appreciation, consent –
these are the essential elements; but knowledge does not invariably imply appreciation, and
both together are not necessarily equivalent to consent …

The characteristics of, and requirements for, valid consent are:

• The plaintiff has to indicate that he or she is willing to suffer harm or run the risk of some
harm. An agreement or contract between the plaintiff and defendant is not required. The
defendant who caused harm to the plaintiff, who had indicated willingness to suffer the harm,
can rely on consent as a defence. The defendant does not have to prove that there was an
agreement of willingness between the parties. Where there was an agreement between the
parties that one person will not hold the other liable for harm that may ensue (pactum de non
petendo), the practical effect is the same as consent. Consent involves only the conduct of the
plaintiff who can revoke the consent at any stage prior to the defendant causing the harm. For
example, a person who has consented to an article being published about his or her private life
can revoke this consent before publication, and the publication would then be a wrongful
invasion of privacy.9
• The consenting party must have indicated consent in an obvious manner. If there was no
external sign of consent, for example, if the plaintiff remained quiet and passive when the
harm occurred, the plaintiff could of course simply deny that he or she had consented and the
defendant would not be able to prove the contrary.
• A person can give consent verbally, either expressly or by implication, or tacitly by conduct.
Encouragement or invitation to cause harm does indicate consent, but mere acquiescence or
knowledge that harm will ensue is not enough to constitute consent.

PAUSE FOR Consent

 REFLECTION In Jordaan v Delarey10 the plaintiff asked the defendant to repeat
defamatory words in the presence of two policemen, which the defendant
did. The Court held that this request amounted to consent. However, in
such circumstances, the request can also be interpreted as a challenge
to the defendant, for the plaintiff to obtain evidence in order to vindicate
his rights in court.

• Consent must be given before the harm occurs. As consent is a ground of justification that
involves a waiver of rights in respect of the harm concerned, causing the harm prior to giving
the consent will be wrongful. However, the person who suffers the harm may of course
afterwards also waive the right to claim damages, either unilaterally, or by way of an
agreement not to claim (pactum de non petendo).
• Consent must be given by a person capable of expressing his or her will, and who is the person
who will suffer the harm. This does not mean that majority or full legal capacity is required for
consent. However, the person who consents must have the mental ability to appreciate the
implications of his or her actions, to distinguish between right and wrong, and to act
accordingly. Courts will consider all the circumstances of the case to determine whether the
necessary capacity existed, including the nature and value of the interest affected, and the age,
intelligence, knowledge and experience of the person who is alleged to have consented. As
consent by a minor in respect of a patrimonial interest amounts to the alienation of such an
interest, courts will tend to require full legal capacity for such consent. However, they follow a
less strict approach in respect of a minor’s consent to conduct that affects personality interests
such as privacy or bodily integrity. If a person is not capable of expressing his or her will, such
as a young child, someone who can lawfully express a will on behalf of this person, such as a
parent or guardian, may give consent. Where parents refuse to consent to an operation a
physician considers necessary, or where the parents are deceased or incapacitated, the
responsible Minister can give the necessary consent.11 In an emergency, the medical
superintendent of a hospital may consent to the medical treatment of a child.12 Consent to
medical treatment of mentally ill persons is regulated by the Mental Health Act 18 of 1973.
The consent of a parent on behalf of a child must be reasonable and in the interests of the
child. For example, consent that the child may participate in a dangerous activity that is of no
value or benefit to the child will be unreasonable and invalid.
• Consent must be given freely and voluntarily. Courts will not regard as valid consent to suffer
or risk harm from a person who was under moral, social or economic pressure, for example, in
a situation where an employee undertakes dangerous work in the course of his or her
employment,13 or submits to physical punishment.14
• Full prior knowledge of the nature and extent of the harm or the risk of harm is required. To be
valid, the consenting party must have had information on the material aspects of the harm or
the risk of harm that is involved, so that the consent is informed. This is particularly important
in medical treatment. In Castell v De Greef 15 the Court reviewed the South African law in this
regard and set out the requirements for informed consent in the context of medical treatment.
For consent to operate as a defence:
◆ The consenting party must have had knowledge and been aware of the nature and extent of
the harm or risk.
◆ The consenting party must have appreciated and understood the nature and extent of the
harm or risk.
◆ The consenting party must have consented to the harm or assumed risk.
 ◆ The consent must be comprehensive, that is, extend to the entire transaction, inclusive of its
consequences.16
• In the context of medical treatment and its consequences, for a patient’s consent to constitute a
justification that excludes wrongfulness, the doctor is obliged to warn a patient of a material
risk in the proposed treatment. A risk is material if, in the circumstances of the particular case,
(a) a reasonable person in the patient’s position, if warned of the risk, would attach
significance to it, or (b) the medical practitioner is, or should reasonably be aware that the
particular patient, if warned of the risk, would attach significance to it.17 However, this
obligation is subject to the so-called ‘therapeutic privilege’. The ‘privilege’ allows medical
practitioners to withhold information, which in their opinion would be detrimental to the
patient in question, regardless of the negative impact that this has on the patient’s freedom or
independence.18 Although expert medical evidence is relevant in determining what the inherent
risks of treatment (surgical or otherwise) are, and might also have a bearing on their
materiality, this is not a question that courts should answer on the basis of expert medical
evidence alone. The ultimate question is whether the defendant’s conduct conforms to the
standard of reasonable care demanded by the law. That is a question for courts to decide, and
they cannot delegate the duty of deciding it to any profession or group in the community.19
• The consenting party must have been willing to suffer the harm, both where the harm is certain
to occur through intentional conduct, as in the case of a medical operation, and where there is a
risk of harm resulting from a dangerous activity. Waring and Gillow Ltd v Sherborne 20 states
this requirement as:
Knowledge, appreciation, consent – these are the essential elements, but
knowledge does not invariably imply appreciation, and both together are not
necessarily equivalent to consent.

• The consent must be lawful, that is, be permitted in terms of the general standard of
reasonableness, in accordance with the boni mores or the legal convictions of the community.
Even if all the requirements for consent as set out in the previous bulleted paragraphs have
been met, the consent will not be valid if consenting to the causing of harm offends the boni
mores, the legal convictions of the community, or contravenes a statute. Examples of this
include where a person consents to disfigurement or some other form of assault, where a minor
younger than the statutory age of consent agrees to sexual intercourse, where a woman agrees
to sexual intercourse in circumstances that amount to seduction, and ‘grooming’.21 Arguably,
consent to the risk of injury during a hazardous activity that has no redeeming social value,
such as a dangerous car race on a country road,22 or a ride with a drunken driver,23 could be
regarded as contra bonos mores and therefore invalid. One could also regard the injured
person’s voluntary exposure to such a risk as contributory negligence,24 rather than as consent.

PAUSE FOR Consent

REFLECTION The growth in the use of smartphones and social media apps has led to
an unexpected growth of ‘sexting’. ‘Sexting’ is the sending of sexually
explicit or suggestive photos or other material via electronic media. By
sending the explicit material, the sender can be taken to have consented
to share it with the intended recipient. However, problems may arise
where the recipient then shares the material with other users, which the
sender did not necessarily consent to. In National Media Ltd v Jooste25
the Court held that an individual has the right to decide on the scope of
disclosure of private facts, and to decide the conditions under which such
private facts may be made public. Accordingly, if the recipient of sexually
 explicit material shares this with third-party users without the consent of
the first sender, this could be wrongful invasion of the privacy of the first
sender.

10.3 Consent by assumption of risk

Where there is express consent it is a matter of construction to determine what exactly was
consented to. In the case of assumption of a risk, the requirement of subjective willingness to suffer
harm presents some difficulty. Mere knowledge of the possibility of harm is not enough to
constitute consent. The maxim is volenti non fit iniuria, and not scienti non fit iniuria.26 Where the
contention is that there was an assumption of a risk, the question is whether the person could
foresee the harm that eventually occurred and accepted it as falling within the ambit of the risk. The
enquiry is subjective and foresight of the possible harm is a key feature.
Lampert v Hefer NO27
The plaintiff was a passenger in the sidecar of a motor cycle and was injured in an accident
caused by the driver’s negligence. The driver was killed in the accident. The plaintiff claimed
damages from the driver’s executor and the Court upheld the defence of volenti non fit iniuria
on the ground that when the plaintiff entered the sidecar at the beginning of the journey, the
plaintiff knew that the driver was highly intoxicated. The plaintiff’s application for leave to
appeal in forma pauperis against the judgment of the Court was refused by the Appellate
Division. In delivering the main judgment, Fagan JA stated that ‘serious intoxication of the
driver of a motor-vehicle must always involve a risk of accident’,28 and concluded that the
plaintiff, who had previously also travelled with the driver while he was intoxicated, must have
or should have appreciated the risk of an accident and injury, even though she had been
fortunate before. Her assumption of the risk on this occasion could therefore be relied upon
as a complete defence by the defendant.
Fagan JA held that the defences of consent and contributory negligence may often
overlap. This case was decided before the Apportionment of Damages Act 34 of 1956 was in
force, and the all-or-nothing rule of the common law still applied in respect of contributory
negligence. Therefore, at the time, both consent and contributory negligence were all-or-
nothing defences. So, to make a decision in the Lampert case, it was not vital to categorise
the defence of assumption of risk as either consent or contributory negligence, as either one
was a complete defence. In terms of the Apportionment of Damages Act, contributory
negligence on the part of a plaintiff is no longer a complete defence, but leads to a reduction
of the damages that the plaintiff is entitled to.29
Since contributory negligence is no longer a complete defence and it has become possible to
produce a fair result by apportionment of damages, courts have been reluctant to recognise consent
in assumption of risk cases where there was negligence on the part of the defendant. In most cases
where the defence of consent in the form of assumption of risk has been raised, it has failed. Courts
recognise this defence in principle, but find that the plaintiff did not consent to the defendant
causing harm and that the voluntary exposure to risk amounts to contributory negligence. Only in
exceptional circumstances will courts find that a person consented to run the risk of another
person’s negligence.30
Santam Insurance Co Ltd v Vorster31
This is the leading case on consent in the assumption of risk where the plaintiff was a
passenger in one of two cars that were being driven in a race on a country road. The drivers
and passengers had placed bets on the outcome of the race. There was a collision when one
driver attempted to overtake the other on a bend in the road. The plaintiff was severely
injured and claimed damages on account of the negligent conduct of both drivers. The
 defences of consent and contributory negligence were raised, on the basis that the plaintiff
had known the risks of the race and had voluntarily exposed himself to those risks. It was
held that, in addition to knowledge and appreciation of danger, the plaintiff must have
foreseen the particular risk that culminated in the harm. The Court held that it is inherently
difficult to determine such foresight because of its subjective nature. Therefore, courts must
determine consent in the form of assumption of risk by analysing the proved facts. The
evidence of the plaintiff himself carries little weight. Based on the relevant facts, courts must
determine what the inherent risks of the particular hazardous activity were and whether the
risk that culminated in the harm fell within the ambit of these risks. Therefore, the essential
question is whether the plaintiff must have foreseen the particular risk that culminated in the
harm as part of the inherent risks of the activity, and should therefore be held to have
consented to these risks.32

Consent by assumption of risk is important in sport-related injuries. Taking part in sport often
involves certain inherent risks of injury as a result of the conduct of the other participants. A
participant normally consents to the risk of injuries that occur reasonably within the normal course
of a game or event. The fact that the injury-causing conduct of one player broke the rules of the
sport will not necessarily mean that the injury falls outside the ambit of another player’s consent. A
late tackle, or running into an opponent in a dangerous manner, may break the rules of a sport, but
it is nevertheless foreseeable in the normal course of a game. However, where the injury is caused
by gross contravention of the rules or deliberately dangerous conduct, the position is different.33 To
determine whether the injury occurred reasonably within the normal course of the game, and
therefore within the ambit of a participant’s consent, courts take all the circumstances into account.
Boshoff v Boshoff34
The plaintiff began action for damages for injuries sustained when he was hit by his
opponent’s racket during a game of squash. The defence of consent was raised. The Court
held that injuries of this nature are reasonably to be expected in a social game of squash
between amateurs. Although the plaintiff had not admitted that he had accepted the risk of
injury, the Court found that, had the plaintiff (an advocate) been asked before the match
whether he consented to this risk of injury, he would have answered in the affirmative. It was,
therefore, the ‘will’ of the plaintiff to run the risk of injury. Every intelligent person is, to a
certain extent at least, master of his own fate. It is not contra bonos mores for a person who
can express a will to consent to sustaining an injury or to run the risk of an injury in the course
of lawful sport or physical recreation. A bona fide sportsman, who causes injury to a fellow
player in a reasonable manner, as may be expected in the normal course of a game, can rely
on the defence of consent. The defence is based on the fact that the players know and
accept the risk of injury in the normal course of the game. In this regard, the concept of will
does not require a positive desire to be injured. It requires a legal ‘will’ or acceptance of injury
or the risk of injury. The Court said:35

If the conduct is deliberately intended to injure someone whose presence is

known, or is reckless and in disregard of all safety of others so that it is a
departure from the standards which might reasonably be expected in anyone
pursuing the competition or game, then the performer might well be held liable
for any injury his act caused. There would, I think, be a difference, for instance,
in assessing blame which is actionable between an injury caused by a tennis
ball hit or a racket accidentally thrown in the course of play into the spectators
at Wimbledon and a ball hit or a racket thrown into the stands in temper or
annoyance when play was not in progress.

Therefore, the defence of consent succeeded.

In sports injury cases defendants can also rely on defences other than consent by assumption of
risk. An example of such a defence is the absence of fault, that is, the injury was not caused
intentionally or negligently. In Clark v Welsh,36 the Court held that a person who hit a wayward golf
shot from a tee and struck a fellow player, who was standing slightly in front of the tee, was not
negligent in the circumstances of the case. The Court could also have argued that the injury was
caused lawfully because it was reasonable to cause such injury in the normal course of the game. In
Boshoff v Boshoff, 37 the Court held that, even if the plaintiff had not consented, causing an injury in
the normal course of a game of sport is not regarded as wrongful in terms of the general standard of
reasonableness. The Court also alluded to the possible absence of fault.
Roux v Hattingh38
Hattingh suffered serious injuries in the course of a rugby match when a scrum collapsed as
a result of an illegal and extremely dangerous manoeuvre that Roux had initiated. The
Supreme Court of Appeal noted that the mere fact that the manoeuvre contravened the rules
of the game did not mean that the act was wrongful and required the imposition of delictual
liability.39 However, a blatant flouting of the rules which was intended to cause serious injury
was not reasonably foreseeable and would attract delictual liability.40 On the issue of voluntary
assumption of risk, the Court said:41

The egregious nature of Alex’s conduct places it beyond the pale. Public and
legal policy, I have no doubt, require such conduct to be stigmatised as
wrongful. I also take the view, along with Fourie J in the court below, that
because this conduct amounted to such a serious violation of the rules, it is not
normally associated with the game of rugby and is extremely dangerous, it
would ‘not have constituted conduct which rugby players would accept as part
and parcel of the normal risks inherent to their participation in a game of rugby’.
In the result, the conduct is wrongful and the justification of consent cannot
avail Alex.

The existence of consent is a question of fact and the onus of proving these facts, lies with the
defendant.42 Courts do not readily accept that there was consent to harm, and so approach the
defence of consent with care.43 Where the defendant mistakenly thought that consent had been
given, the causing of the harm will still be wrongful, but the defendant may escape liability because
of absence of fault, that is, absence of the knowledge of wrongfulness as an element of intention.
Consent to the risk of death by a breadwinner, for example, when taking part in a hazardous
activity that may result in injury or even death, is not a defence against a claim by dependants for
loss of support, if the death of the breadwinner was caused by another person’s negligence.

PAUSE FOR Loss of support

REFLECTION A claim for loss of support by dependants is based on the infringement of
their right to support. There is no basis in law for accepting that the
breadwinner can, by consent, deprive dependants of their claim for loss
of support. It is in any event difficult to imagine this situation arising. The
breadwinner is not likely to consent to the risk that the negligent conduct
of another person may cause his death, and the person who acted
negligently will therefore almost never be able to prove consent. If there
was no negligence on the part of another person, there will in any event
be no liability. Even if the breadwinner did consent to the risk that the
negligent conduct of another person may cause his death, courts will
probably regard such consent as contra bonos mores and therefore
invalid. An agreement by the breadwinner that no claim will be made in
the event of injury or death (pactum de non petendo), even if it binds his
 estate and his dependants, will also not affect the right of his dependants
to claim compensation for loss of support in the event of his death. As the
Court decided in Jameson’s Minors v Central South African Railways,44
dependants are not bound by such an agreement, because they are not
parties to it, and the breadwinner cannot contract on behalf of his
dependants to their detriment. (This is discussed in more detail in the
section on contractual exemption clauses.)

10.4 Prior agreement not to claim (pactum de non petendo in

anticipando)
A prior agreement not to claim damages should the conduct of another person cause harm (pactum
de non petendo in anticipando), is contractual. Therefore, it differs from consent, which involves
the conduct of only the person consenting to suffer harm or run the risk of some harm. The effects
of these two legal phenomena are different, in that consent justifies causing harm, whereas a
pactum de non petendo in anticipando merely excludes the recovery of damages for wrongfully
caused harm. However, the practical effect of a pactum de non petendo in anticipando is the same
as that of consent, in that it provides a complete defence against a claim for damages as a result of
anticipated and later realised harm. In certain circumstances, the conclusion of such an agreement
could, at the same time, indicate a willingness to suffer harm or to run the risk of harm, and could
therefore also constitute consent, for example, where the parties informally agree, before taking
part in a dangerous activity, that their involvement is at their own risk and that liability for damages
is excluded. The requirements for a valid pactum de non petendo in anticipando are the same as for
any other contract. One person cannot conclude a pactum de non petendo in anticipando on behalf
of another without authorisation. As was decided in Jameson’s Minors v Central South African
Railways,45 an agreement by a breadwinner that no claim will be made in the event of his injury or
death, even if it binds his estate and his dependants, will not affect the right of his dependants to
claim compensation for loss of support in the event of his death.
Courts accept that a parent or guardian can validly conclude a pactum de non petendo in
anticipando on behalf of a minor child, for example, where a school or sports club requires this
undertaking from the parent or guardian to allow the child to participate in an educational activity
or sport. In Minister of Education and Culture (House of Delegates) v Azel 46 the Court decided that
the undertaking by a parent, which limited the school’s liability for damage to property or personal
injury of the child, did not have effect where there was negligence on the part of a responsible
teacher. In Durban’s Water Wonderland (Pty) Ltd v Botha 47 the Court decided that a parent had
agreed, on behalf of a child, to conditions that excluded the liability of an amusement park owner
for injury to the child while using the park’s facilities.

PAUSE FOR Prior agreement not to claim on behalf of a child

REFLECTION Arguably, a pactum de non petendo in anticipando agreed to by a parent,
on behalf of a minor child, is not to the benefit of the child, who should be
entitled to restitutio in integrum, to have the effects of the agreement set
aside.

The effect of a pactum de non petendo in anticipando depends on the precise ambit of the
agreement. Courts tend to interpret such agreements restrictively. An agreement to exclude liability
for harm caused intentionally will not be valid, but an agreement that excludes liability for harm
caused by negligence is regarded as valid.48 An agreement excluding liability for gross negligence
is void if the Consumer Protection Act 68 of 2008 is applicable.49

10.5 Necessity
Necessity can justify the infringement of the interests of an innocent person (that is, someone not
causing or threatening to cause harm wrongfully). This occurs in a situation where the infringement
is the only reasonable way of protecting one’s own interest or that of another person against danger
created by natural phenomena or human conduct.50 For example, a person may be justified in
breaking a window to enter another person’s house to telephone for help in the event of injury, or to
rescue someone trapped inside the house. In such a case, necessity justifies damaging the house.
An attack by an animal and danger created by forces of nature are natural phenomena that do
not cause harm wrongfully. Wrongfulness is an attribute of the outcome of a causal sequence set in
motion by human conduct. Therefore, defensive conduct when being attacked by an animal or
when faced with danger from forces of nature does not constitute private defence, but can be
justified on the ground of necessity.
The main difference between necessity and private defence is that necessity justifies causing
harm to an innocent person (that is someone who is not causing or threatening to cause harm
wrongfully), whereas private defence justifies causing harm to a person who is causing or
threatening to cause harm wrongfully. So, the requirements of private defence and necessity differ
accordingly. Necessity requires proportionality between the protected interest and the infringed
interest.
Necessity, as a ground of justification, simply involves applying the general criterion of
reasonableness or boni mores. To justify that the infringement of the interest was lawful, the person
relying on necessity must prove that protecting a legally recognised interest against danger created
by natural phenomena or human conduct was, in terms of the proportionality of the interests
involved and the nature and extent of the danger and the means of protection, acceptable in terms of
the boni mores or the legal convictions of the community. Reliance on necessity requires the
following:
• There must have been actual danger to a legally recognised interest. In accordance with the
general approach to assessing wrongfulness, the existence and extent of the danger must be
determined objectively. This must be on the basis of the actual circumstances, and not on the
perception or reaction of the person who relies on the state of necessity. However, as in the
case of private defence, courts have sometimes adopted a qualified objective approach. They
do this by determining whether the person who is said to have acted in necessity reacted to the
situation reasonably,51 instead of taking into account all the facts of the situation that become
known after the event. The following case study illustrates this approach.
S v Pretorius52
Necessity was raised as a defence against a charge of speeding. A parent had rushed to
hospital with his child after the child had taken a number of tablets. The parent believed that
his child’s life was in danger. It turned out that the child had taken pain tablets and would
have at most suffered from an upset stomach. Nevertheless, the Court held that the parent
had acted reasonably in a situation of necessity, and therefore set aside the conviction for
speeding.
There are two possible approaches to the situation encountered in the Pretorius case. The
court may reach either of the following conclusions:
• The person relying on necessity had overreacted and therefore wrongfully exceeded the
bounds of necessity. However, the person lacked fault, because a reasonable person would
 have reacted in the same way, and therefore liability cannot be imposed for causing the
harm.
• The person relying on necessity had acted as a reasonable person would have under the
circumstances, and therefore acted lawfully within the bounds of necessity, and so cannot be
held liable for causing harm.

In principle, the first approach is preferable. It is consistent with the accepted view that
wrongfulness is determined on the basis of general reasonableness and policy
considerations, rather than by applying the criterion of the reasonable person, which is central
to the enquiry into negligence.

• The endangered interest can be that of the defendant’s or of another person. For example,
where a child’s life is in danger, a parent may take protective measures that may infringe the
interests of another person. This necessity-induced conduct may both benefit and disadvantage
the same person, for example, where a physician needs to perform an emergency operation on
an unconscious patient, who then suffers the pain and discomfort of an operation, but gains a
health benefit.
Petersen v Minister of Safety and Security53
In this case, the police attempted to seize a number of bags of illegally harvested perlemoen
(abalone) in an area infamous for perlemoen poaching. A belligerent crowd gathered and
began stoning the police, forcing their retreat. That allowed the crowd to make off with most
of the perlemoen. The police attempted to stave off the attack by firing rubber bullets into the
crowd and, when they ran out of rubber bullets, shooting into the ground near the crowd with
sharp point ammunition. The plaintiff claimed damages for injuries sustained by her son in the
shooting. The police alleged that the shooting was justified by necessity. The plea of
necessity was upheld on appeal. The Court decided that the defence of necessity does not
require the harmful action to be directed at a person engaged in a wrongful attack. There was
no need for the police to establish that the injured boy was himself part of the attacking
crowd. They had to prove that danger existed, or was imminent, that there was no other
reasonable means of averting the danger, and that the means used and measures taken to
avert the danger of harm were not excessive, having regard to all the circumstances of the
case, including that, at the stage when the police started to fire live ammunition, their
attackers simply did not allow them to flee. The causing of the injuries by the police was
therefore not wrongful, but justified by necessity.

• Any legally recognised interest may be protected on the ground of necessity, for example, life
or physical integrity and property. Subject to the principle of proportionality, the necessity-
induced conduct may also infringe any type of interest, for example, life or physical integrity,
and honour or freedom.
• The danger can be created by any natural phenomenon, such as fire or flood, or by human
conduct, such as duress. There are conflicting views on whether defendants can rely on
necessity in situations where they themselves created a situation of danger and then, while
protecting their own interests or that of others from such danger, caused harm to others.

PAUSE FOR Using necessity

REFLECTION In criminal cases courts have on the one hand held that a defendant
cannot rely on necessity to escape the consequences of a situation that
he or she personally had created,54 while on the other hand, the Appellate
Division has accepted that defensive conduct may be justified where a
dangerous situation resulted from prior wrongful conduct by the
defendant.55 A person may start a sequence of events that creates
 danger and then, while protecting his or her own interests or that of
others from such danger, cause harm to others. In this situation, the
enquiry into wrongfulness should not be restricted to the part of the
sequence of events that involved a situation of necessity. To assess
wrongfulness courts must take into account the entire sequence of
events. For example, a parent has started a fire that endangers the life of
his or her child and then, while saving the child, causes damage to a third
party. In this situation the parent clearly had a duty to save the child, but
when taking into account the entire sequence of events, the conclusion
should be that he or she wrongfully caused damage to the third party.
The same applies in the case of a person who hides in a shop after
closing hours and then needs to break a window to get out.

• The danger must be present or imminent. As in the case of private defence, anticipatory
defensive conduct is not justified on the ground of necessity. Where a person expects a
dangerous situation to develop, it is not justified for that person to cause harm to others while
protecting his or her interests in anticipation of the danger. A declaratory order or interdict may
be the appropriate remedy where future danger is expected as a result of another person’s
conduct, or where danger is expected from a natural phenomenon such as a flood, and
reasonable measures to avert harm will affect the interests or require the cooperation of another
person. Where the danger is imminent, although not yet present, a person may reasonably
protect his or her interests without having to wait until the danger has actually arrived. Where
the danger has already ceased a person cannot rely on necessity, for example, if a neighbour’s
dog bites a child, the parent of the child cannot follow the dog and shoot it.
• A person wishing to rely on necessity must not be legally obliged to endure the consequences
of the dangerous situation. For example, the owner of a low-lying property must receive run-
off water from an adjacent higher property, even if it damages his or her property. He or she
may not take preventative steps that are harmful to the other landowner on the ground of
necessity.
• There must be proportionality between the protected interest and the interest infringed by the
protective conduct. Because necessity justifies causing harm to an innocent person, it is
unreasonable if the harm caused to the innocent person outweighs the harm prevented by the
protective conduct. However, even if the protected interest outweighs the infringed interest,
protective conduct will not necessarily be justified. Without a special duty or relationship it is
generally unreasonable to force a person against his or her will to sacrifice his or her interests
to protect the endangered interests of another person. For example, a person cannot be forced
to donate blood to save the life of a critically ill person, and a person wearing an expensive
leather jacket cannot take an umbrella from a person wearing an inexpensive garment if it
suddenly starts to rain. The general test of reasonableness applies, and courts must take into
account all the surrounding circumstances when assessing whether protective conduct is
justified.
• In terms of the general test of reasonableness the necessity-induced conduct must have been
necessary, or the only reasonable means of protecting the interests concerned. If the defendant
could have escaped from the dangerous situation by taking flight, he or she should have done
so.56
R v Dudley & Stevens57
The issue of proportionality of interests gives rise to the question of whether protecting a life
in a situation of necessity can justify taking another life. Taking one life to save another is not
regarded as justifiable in English law. This case dealt with the situation of people adrift in a
 small boat for 21 days after their ship sank. After they had spent eight days without food, two
of them killed and ate a cabin boy, who was the weakest of the persons on the boat. They
were later rescued, subsequently charged with murder, and raised the plea of necessity. The
Court held that necessity can never justify killing an innocent person. This is also the view of
courts in other European jurisdictions, on the basis that one cannot regard one life as more
valuable than another.
Before 1972, South African courts also refused to recognise necessity as a defence on a
charge of murder. However, in S v Goliath,58 the Appellate Division changed course and held
that necessity, in the form of compulsion or duress, can be a complete defence to a charge
of murder. It was emphasised though, that the success of this defence depends on the
particular circumstances of each case, and that such a defence will be considered with the
utmost care. In this instance, the accused A, was forced by accused B, to take part in killing
an innocent person C. B threatened to kill A if he did not assist in killing C. A, fearing for his
life, complied. In a key passage of the majority judgment, Rumpff JA held that an ordinary
person regards his own life as more important than the life of another, and that only a person
with qualities of heroism will sacrifice his own life for that of another. The law does not
require that a person acting under duress should conform to a higher standard than that of
the average person. Therefore, compulsion or duress can be a complete defence to a charge
of murder.
In terms of the general test of reasonableness, the effect of the majority judgment in the
Goliath case is that the boni mores or legal convictions of the community only require that the
person acting under duress acts as the average or reasonable person would under the
circumstances. Courts will be very cautious in recognising necessity as a defence where an
innocent person has lost his life. The standard will be that of an ordinary, average or
reasonable person. There is no basis for suggesting that courts will apply a different standard
in a civil case. Consequently, the dependants of the victim in the Goliath case would not
have succeeded with a claim for loss of support against the accused.

10.6 Self-defence (private defence)

Private defence justifies protecting a legally recognised interest against actual or imminent
wrongful attack. The policy considerations that underlie private defence were formulated as follows
in Ntsomi v Minister of Law and Order: 59
The principle that right does not have to yield to wrong has been a touchstone of civilised
legal systems through the ages. The victim of an unlawful attack has been entitled to defend
his person or property by virtue of a rule of law which has existed in all familiar legal
systems for many centuries.

The requirements for relying on private defence are the following: 60

• There must have been a wrongful attack. The attacker’s conduct must have wrongfully caused,
or threatened to cause, the infringement of an interest. Private defence is not justified against a
lawful infringement of an interest, for example, against a lawful arrest. The fact that a person
who has committed a crime for which he may be arrested without a warrant is running away
from the scene of his crime pursued by those who saw him do it, does not change him into a
threatened innocent with the right to use violence against those who are trying to arrest him.61
As we noted in the context of necessity,62 attacks by animals and danger created by forces of
nature are natural phenomena that do not cause harm wrongfully. Wrongfulness is an attribute
of a causal sequence set in motion by human conduct. Therefore, danger created by animals or
forces of nature does not constitute a wrongful attack. However, for the purposes of private
defence, defensive conduct in the face of such danger can be justified on the ground of
necessity. Although, where a person uses an animal as an instrument of attack, for instance
 inciting a dog to bite, there can be private defence against the human conduct causing or
threatening infringement of a right. In most cases private defence involves an attack in the
form of positive conduct. However, an omission can also give rise to private defence, for
example, where a salesman refuses to leave another person’s house, or where a fireman refuses
to extinguish a fire on request of a house-owner. Some measure of coercion may be reasonable
in such cases, but South African courts have not decided on this question.
• The attack must be directed against a legally recognised interest. People usually rely on private
defence in cases of attacks on life or bodily integrity. However, defence of other interests such
as honour 63 or property 64 can also be justified. There is no closed list of protectable interests,
and in principle, a wrongful attack on any legally recognised interest can give rise to a plea of
private defence.
• The attack must have commenced or must be threatening. Anticipatory defence is not justified.
Where a person expects another person to attack a legally recognised interest some time in the
future, it is not justified to use force against the attacker in anticipation.65 In such a case, an
interdict may be the appropriate remedy. Where an attack is threatening, the defender does not
need to wait until the attack has actually begun.66 Where a harmful attack has already stopped,
the person who suffered the harm cannot rely on private defence to justify retaliation.67
• The attacker does not have to be at fault. Private defence justifies warding off harm caused
wrongfully, regardless of whether the causing of the harm is intentional, negligent or without
fault. Private defence is therefore also justified against an attack by a person incapable of fault,
such as an infans or a mentally disturbed person.68
• The attack does not have to be directed at the defender. A person may defend another’s person
or property against wrongful attack. For instance, one policeman may be justified in using
force to ward off an attack on a fellow policeman,69 and a husband may prevent an
infringement of his wife’s honour.70 As private defence is based on the policy consideration
that right should not yield to wrong, a person can defend the interests of another person against
wrongful attack for purely altruistic reasons. There does not have to be any special relationship
between the defender and the person whose interests are wrongfully infringed or threatened.
• The defence must be directed at the attacker. Private defence justifies causing only the harm
that is reasonably necessary to ward off a wrongful attack. It does not justify causing harm to a
person other than the attacker.
• The means of defence must be necessary and reasonable to prevent the threatened harm. It is
lawful for a person to use a reasonable degree of force to protect himself or another person
against any wrongful attack. However, a harmful means of defence is not justified when the
threat could have been avoided in some non-harmful or less harmful way.71 Causing harm in
the course of private defence is not justified if the harm is either unnecessary, that is, if the
infringement of interest could have been prevented in some other way, or is disproportionate to
the protected interest.72 The issues of necessity and proportionality raise several further
questions.
• The defence against the attack must be necessary. Could the defender have taken flight? Courts
have accepted that the victim of an attack is not expected to flee from an assailant if flight
would be dangerous or if it would amount to a dereliction of duty.73

PAUSE FOR Proportionality between defence and attack

REFLECTION The question of proportionality between defence and attack is often the
most difficult part of the inquiry into private defence. The means used,
and the harm caused by the defender must not be out of proportion to the
threatened harm.74 In a number of cases, courts have taken into account
what the person who acted in private defence could reasonably have
known, or how he or she could reasonably be expected to react to the
situation, instead of taking into account all the facts of the situation that
were known after the event. Van Winsen AJ formulated this approach in
Ntanjana v Vorster and Minister of Justice:75

The very objectivity of the test, however, demands that

when the Court comes to decide whether there was a
necessity to act in self-defence it must place itself in the
position of the person claiming to have acted in self-
defence and consider all the surrounding factors operating
on his mind at the time he acted. The Court must be careful
to avoid the role of the armchair critic wise after the event,
weighing the matter in the secluded security of the
courtroom… . Furthermore, in judging the matter it must be
ever present to the mind of the Judge that, at any rate in the
particular circumstances of this case, the person claiming
to act in self-defence does so in an emergency, the creation
of which is the work of the person unlawfully attacking. The
self-defender is accordingly entitled to have extended to
him that degree of indulgence usually accorded by the law
when judging the conduct of a person acting in a situation
of imminent peril. ‘Men faced in moments of crisis with a
choice of alternatives are not to be judged as if they had
had both time and opportunity to weigh the pros and cons’,
per Innes JA in Union Government (Minister of Railways &
Harbours) v Buur.76

In this approach the defender may be judged to have acted reasonably

even if it appears with hindsight that he was not really in danger, or that
he caused more harm in defence than was required to ward off the
attack.77
The cases in effect indicate that the boni mores or legal convictions of the
community only require that the defender acts as a normal reasonable
person would under the circumstances. For example, an armed
policeman using force when faced with a threat of violence will be judged
according to the norm of a reasonable policeman.78
There is a strong dissent from the reasonableness-of-conduct
approach to mistaken (putative) defence in Kgaleng v Minister of Safety
and Security.79 The Court said:

The defendants may yet escape liability on the basis that

the second defendant’s bona fide (although erroneous)
belief that his conduct was justified, excluded
consciousness of wrongfulness – and thus fault in the form
of dolus – on his part, and provided a reasonable man
would not have reacted differently to the way in which the
second defendant reacted under the circumstances –
thereby excluding fault in the form of culpa.80

However, in Mugwena v Minister of Safety and Security 81 the Supreme

Court of Appeal also applied the test of a reasonable person to determine
whether the causing of injury was justified by defence.82 The two
approaches will usually produce the same result. One could argue on
one hand that the defender used disproportionate force and therefore
wrongfully exceeded the bounds of private defence, but that he lacked
 fault because he acted as any reasonable man would have under the
circumstances. The other point of view is that the defender caused harm
lawfully in the course of private defence, according to the standard of a
reasonable person in the particular circumstances.

• The interest that the defender protects need not be commensurate with or similar in character
to the attacker’s interest that is infringed by the defence.83 The interest threatened by the attack
and the interest affected by the defence are often different in nature and of unequal value. On
the basis that right should not yield to wrong, the attacker who wrongfully causes or threatens
harm should not have a cause of action against the defender for causing more harm in defence
than he or she would have suffered as a result of the attack.84 However, an extreme imbalance,
for instance where the defender kills the attacker to prevent a petty theft, indicates that the
protective steps were unreasonable.85
Ex Parte Minister van Justisie: In re S v Van Wyk86
This is a well-known instance of private defence that involves an imbalance of interests. A
shop-owner set up a gun in his shop to protect his property against thieves. His shop had
been the target of repeated burglaries, and other protective measures such as burglar-
proofing, night-watchmen and watchdogs had not worked. He set up the gun with a trip-wire,
so that an intruder setting off the wire would be hit in the legs. He also put up a notice on the
door of the shop, warning people that there was a gun inside the shop. An intruder set off the
gun, was wounded in the hip, and later died as a result of the wound. The shop-owner was
prosecuted for murder and eventually the Appellate Division had to decide whether:
1. A person can, in principle, rely on private defence where defence of property had resulted
in injury or death
2. The bounds of defence were exceeded in this particular case.

All the judges answered the first question affirmatively, holding in essence that killing in
defence of property, where the defender was in no personal danger, can be justified in
particular circumstances. However, the judges’ opinions on the second question differed. The
majority held that, in view of the repeated burglaries and the failure of other methods of
protection, setting up the gun was a reasonable method of defence. In his dissenting
judgment Rumpff JA held that the Court should determine the reasonableness of setting up a
gun that killed the intruder as if the shop-owner himself had fired the gun. This is because
what is wrongful in the form of direct conduct cannot be lawful in the form of indirect
conduct.87 However, this argument does not take into consideration that the shop-owner could
not reasonably have protected his shop by guarding it himself every night and therefore
needed to employ indirect protection, the reasonableness of which was in issue.

PAUSE FOR Private defence today

REFLECTION Given the constitutional protection of the right to life, it is likely that the
Van Wyk case would be decided differently today. The decision would be
based on the fact that there is an inherent disproportionality or imbalance
in weighing up the protection of property against taking a human life. In
cases where the proportionality of the defence is in issue, courts will
consider all the surrounding circumstances, including factors such as the
means of defence at the disposal of the defender, the nature and the
danger of the attack, and a possible imbalance of interests, assessed in
the light of the Constitution.
10.7 Provocation
There is authority in South African law that provocation, in the form of inciting words or conduct,
can be a complete defence to a claim for compensation based on alleged infringement of
personality rights. For example, where the provocative words or conduct of A causes B to insult A,
B can rely on provocation if A claims compensation for the infringement of his or her personality
rights.88
Provocation differs from private defence and necessity in that the person acting in response to
provocation does not anticipate or prevent harm. The person responds to it and retaliates. Courts are
cautious to recognise provocation as a complete defence, because no-one should be encouraged to
take the law into his or her own hands.89
There are different views on the nature of provocation as a complete defence:

• Provocation can exclude fault on the part of the person who responded to the provocation. This
is either because the provoked person did not have the capacity to form intent 90 or because
mental capacity to decide between right and wrong and to act accordingly may have been
affected by the provocation and the person may therefore lack the capacity for fault.91
• Provocation merely affects the amount of compensation that the person who acts provocatively
is entitled to for the consequent infringement of his or her personality rights. He or she may be
entitled to less compensation or may forfeit the right to compensation entirely.92 Courts may
even deny the right to compensation altogether in terms of this approach.
• The third view is that provocation can be a ground of justification. In other words, courts may
regard the person who caused the infringement of personality rights in response to provocation
as having acted reasonably and therefore lawfully.93

COUNTER Self-control
POINT Boberg 94 argues, with reference to Blou v Rose Innes,95 that self-control
should be displayed in the face of provocation and that an aggressive
response is ‘natural, but it is against the law’.

Bester v Calitz96
The Court in this case adopted the view that provocation can be a ground of justification. This
is the case that has so far dealt most comprehensively with the theoretical basis for
recognising provocation as a defence. The case dealt with a claim for damages that arose
from insulting remarks made in anger and elicited by prior provocative conduct, and an insult
of a similar nature. The Court held that in such circumstances, the elicited response is not
wrongful. The traditional justification grounds are merely clear expressions of the application
of the general criterion of wrongfulness, and should not be regarded as a numerus clausus of
defences. Where one cannot apply the traditional justification grounds, but it still appears that
the person’s conduct who responded to provocation was reasonable, one should not impose
liability. One determines wrongfulness according to the general criterion of reasonableness,
which is an objective standard based on the boni mores or legal convictions of the
community. When applying this standard, the approach should not be idealistic. One needs to
take account of human weaknesses and the average person’s inclination to respond angrily
to insult or other forms of provocation. According to the legal convictions of the community,
one must regard reacting angrily and swearing at a person whose conduct was provocative,
as reasonable and not wrongful.
The Court enquired whether the person who was incited by provocation reacted as a
normal reasonable person would. This is instead of adopting the idealistic approach that an
aggressive response to provocation is unnecessary and pointless once one has all the facts
 of the situation after the event, and that self-control is required in the face of provocation. The
Bester case provides a good theoretical basis for recognising provocation as a complete
defence by applying the general criterion of reasonableness according to the boni mores or
legal convictions of the community. Reasonableness in this case is equated to the reaction of
a normal reasonable person.

Relying on provocation as a ground of justification requires the following:

• The provocation must be of such a nature that one can regard the conduct in reaction to it,
whether by way of insult, defamation or even physical assault, as reasonable.97 The general
standard for determining wrongfulness in this case involves the question of whether a
reasonable person in the position of the defendant would have been similarly provoked.
• The reaction to the provocation must be ‘an immediate and reasonable retaliation’. 98 One may
regard an immediate, spontaneous reaction in the heat of the moment as reasonable, whereas a
deliberate or pre-meditated later retaliation may not be reasonable.
• The nature of the conduct in reaction to the provocation must be reasonably in proportion to
the provocation.99 Courts will assess the proportionality objectively, considering all the
circumstances of the case, including the value and nature of the interests affected by the
provocative and retaliatory conduct respectively. As a rule, verbal provocation will not justify
physical assault, even if the verbal provocation was gravely insulting or defamatory.100
Physical provocation may justify a retaliatory assault, provided that the retaliation is
proportionate to the provocation.101 Both physical provocation and verbal provocation may
justify insulting or defamatory remarks in retaliation.102 Where the retaliatory conduct was
disproportionate to the provocative conduct, the defence of provocation will not succeed as a
complete defence. However, the damages of the claimant who provoked the retaliation may be
reduced.103

10.8 Statutory authority

A statutory provision can justify causing harm that would otherwise be wrongful.104 The effect of a
statutory provision is to authorise the infringement of certain rights and correspondingly to restrict
the rights of the persons affected. The person who claims statutory authority must not exceed the
limits of that authority. That person also bears the onus of proving that the causing of harm was
within the bounds of the statutory authority.105
The question of whether the causing of harm was authorised by a statutory provision is one of
interpretation, and the normal rules of statutory interpretation apply. However, courts have also
adopted certain guidelines and presumptions to determine whether the causing of harm in a
particular instance is authorised by statute. Courts use the guidelines and presumptions described
below to help them in the interpretation process, but they are not conclusive.106
• The general rule is that the intention of the legislature will determine whether causing harm in
a particular instance is authorised. Courts determine the intention of the legislature according
to the normal rules of statutory interpretation.107
• If the statute is directory in respect of an action that involves the infringement of private
interests, and does not provide for compensation, this indicates that the infringement of rights
and consequent harm is justified.108
• If the statute is merely permissive and general in respect of an action that involves the
infringement of private interests, and does not provide for compensation, courts presume that
the infringement of rights and consequent harm is not justified. However, there is no such
presumption where a public body is permitted to act in the public interest or where the
permission to act is localised or specific.109
• In the case of certain empowering statutes, the intrinsic physical nature of the works or acts
authorised is such that their execution involves disturbing private rights.110 In Johannesburg
Municipality v African Realty Trust Ltd 111 the Court held that it was impossible to exercise the
power conferred (constructing streets and drains) without increasing the flow of water onto the
plaintiff’s land. In Breede River (Robertson) Irrigation Board v Brink 112 the Court held that it
was impossible to build miles of irrigation canal across the countryside without interfering
with the natural flow of surface drainage water. In Bloemfontein Town Council v Richter 113 the
evidence proved that removing silt from the dam would inevitably cause some damage to the
plaintiff’s adjacent property. In Germiston City Council v Chubb & Sons Lock and Safe Co
(SA) (Pty) Ltd 114 the Court held that constructing roads on sloping ground would inevitably
interfere with the rights of adjoining landowners by changing the natural drainage in that
location.
• Courts will apply the general criterion of reasonableness to determine whether the bounds of
statutory authority have been exceeded. For example, where alleged statutory authority for the
infringement of property rights is concerned, the question is whether there has been
unreasonable interference with the rights of the plaintiffs to enjoy their property.115 In earlier
cases, courts referred in this regard to the question of whether the statutory powers have been
negligently exceeded.116 However, subsequently courts have recognised that the question is one
of wrongfulness, not fault.117 Courts will take into account the extent to which it was
reasonably possible to avoid the infringement of private interests or to minimise harm resulting
from the exercise of statutory powers. The statutory power must be exercised diligently to
avoid harm to another. In this regard, the cost and effectiveness of measures to prevent harm
are relevant.118 In cases of alleged wrongful arrest or excessive use of force when making an
arrest, courts take into account the possibility of an alternative or less harmful conduct.119

10.9 Official capacity

Certain public officials, such as judges, magistrates, prosecutors and members of statutory licensing
boards, are authorised to perform functions that may negatively affect the interests of other persons
– for example, where a judgment by a judge or magistrate contains negative comments about the
credibility of a witness. The infringement of personality or other interests due to the reasonable
performance of official duties is justified in the public interest.120 In many cases, official authority is
conferred by statute – for example, the authority of the police to make arrests and conduct searches,
as well as the rules of statutory authority set out in section 10.8 in this book, will apply. Where
official duties involve discretion and are not defined in detail by statutory provisions, the question
is whether the performance of such duties was reasonable and within the bounds of official
authority.121 If a prosecutor at a bail application negligently fails to place all relevant information
before the court, and the accused is released and then commits a violent sexual crime, there is no
prosecutorial immunity and the State may be liable for harm suffered by victims. 122 An official who
performs official duties with an ulterior motive, malice or dishonesty acts unreasonably and outside
the bounds of authority (or, in the context of defamation, outside the ambit of a privileged
occasion), and the resulting infringement of private interests is accordingly wrongful.123

10.10 Obedience to orders

Obedience to orders as a ground of justification has so far only been raised in criminal cases.
However, it could also arise as a defence in a civil action, for example, where a policeman is given
a wrongful order by a superior officer to shoot at a person running away. The question is to what
extent causing injury or death by the junior policeman is justified by obedience to orders. In S v
Banda124 the Court held that obedience to a wrongful order can be a ground of justification in the
following circumstances:
• The order must come from a person in a position of lawful authority over the accused.
This requirement involves a question of fact.
• There must be a duty on the accused to obey the order given. This involves the question of
when a duty to obey a wrongful order exists, or when obedience to a wrongful order can be a
ground of justification for causing harm. Courts have accepted that there is no absolute or
blind duty to obey orders.125 According to some decisions, there can never be a duty to obey a
wrongful order,126 whereas others hold that only the execution of a deliberately wrongful order
is itself wrongful.127 Courts must then determine whether an order is deliberately wrongful
according to the judgement of a reasonable person.128 This approach is in line with the
approach of the Appellate Division in S v Goliath,129 where the Court held that the law does not
require that a person acting under duress conforms to a higher standard than that of the average
person. So, compulsion or duress can be a complete defence, even to a charge of murder. A
person who is under pressure to obey an order from a superior and who needs to decide at the
same time whether the order is lawful or not, is acting under compulsion or duress. Rumpff JA
held the terms of the general criterion of reasonableness, which agrees with the boni mores or
legal convictions of the community, three years after Goliath in the Ewels case.130 The person
acting under compulsion or duress only has to act as the average or reasonable person would
under the circumstances.
• The defendant must have caused no more harm than was necessary to carry out the order. This
requirement again involves the general criterion of reasonableness. Courts must take into
account all the circumstances of the case when determining whether it was possible to avoid
harm or to cause less harm.

10.11 Disciplinary powers

Persons who exercise lawful authority over children, such as parents131 and persons in loco parentis,
for example, guardians and teachers,132 may in terms of such authority lawfully punish the children
for education and correction.
The authority to discipline of persons in loco parentis is an original authority and does not
derive from delegation by parents,133 although a parent or guardian may delegate disciplinary
powers to another person. Delegated authority to discipline cannot be wider or more comprehensive
than the original authority.134 It involves discretion,135 which must be exercised reasonably and not
in an unpredictable way.136
Section 10 of the South African Schools Act 84 of 1996 prohibits corporal punishment in
public as well as independent (private) schools, by school principals, teachers and persons in charge
of school hostels. The implication of this section is also that parents may not delegate the power to
administer corporal punishment to a person in a public or private school.
The authority to discipline must be exercised moderately and reasonably,137 and in accordance
with the general criterion of reasonableness. The purpose of the punishment must be either to
correct behaviour or be educational.138 Repeating a punishment that has not worked139 and malice or
improper motive140 indicates unreasonableness. Punishment that may be regarded as cruel,
inhumane or degrading is unreasonable and wrongful.141
Courts take the following factors into account when determining the reasonableness of the
punishment:142
• The nature and seriousness of the transgression
• The means and degree of punishment or force used
• The physical and mental condition of the child
• The gender and age of the child
• The physical disposition of the child
• The purpose and motive of the person inflicting the punishment.

Punishment by a person in a position of lawful authority is presumed to be reasonable and without

malice.143 The person who alleges the opposite bears the onus of proving that the punishment was
unreasonable.144

10.12 Impossibility
The law will not enforce impossibility. Where the circumstances indicate that there was a duty to
act positively to prevent harm to another person, but it was impossible for the defendant to act, the
failure to prevent harm will not be wrongful. For example, a strong swimmer who would normally
be able to save a person from drowning may not be able to because of a cramp in his or her leg.
One could say that impossibility is a ground of justification in such a case, because in the
circumstances the defendant’s failure to prevent harm is reasonable and in accordance with the boni
mores or the legal convictions of the community. One could also say that a duty to act positively
does not arise at all in such circumstances, because it is not unreasonable for the defendant,
according to the boni mores or the legal convictions of the community, to fail to rescue the
swimmer. In theory, impossibility can be a separate ground of justification. However, courts also
take into account physical or economic impossibility when applying the general criterion of
reasonableness.145

Figure 10.1 Grounds for justification: Defences directed at the wrongfulness element

1 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 121.
2 1992 (4) SA 630 (D) at 650J.
3 Mabaso v Felix 1981 (3) SA 865 (A).
4 See, for example, Roux v Hattingh 2012 (6) SA 428 (SCA).
5 See Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 778.
 6 See Clark v Welsh 1976 (3) SA 484 (A) (golf); Boshoff v Boshoff 1987 (2) SA 694 (O) (squash); Roux v Hattingh
2012 (6) SA 428 (SCA) (rugby).
7 See Esterhuizen v Administrator, Transvaal 1957 (3) SA 710 (T); Castell v De Greeff 1994 (4) SA 408 (C); Broude v
McIntosh 1998 (3) SA 60 (SCA).
8 1904 TS 340 at 344.
9 National Media Ltd v Jooste 1996 (3) SA 262 (A). In Ketler Investments CC t/a Ketler Presentations v Internet
Service Providers’ Association 2014 (2) SA 569 (GJ) the Court held that the defendant’s post-publication withdrawal
of consent was irrelevant.
10 1958 (1) SA 638 (T).
11 Section 129(7) of the Children’s Act 38 of 2005.
12 Section 129(6) of the Children’s Act 38 of 2005.
13 Waring & Gillow Ltd v Sherborne 1904 TS 340.
14 R v McCoy 1953 (2) SA 4 (SR); S v Collett 1978 (3) SA 206 (RA).
15 1994 (4) SA 408 (C).
16 At 425I.
17 At 426G.
18 At 426H.
19 At 426I–J.
20 Waring & Gillow Ltd v Sherborne 1904 TS 340 at 344.
21 S v SM 2013 (2) SACR 111 (SCA). Also see the judgment of the Court a quo in S v M 2007 (2) SACR 60 (W) paras
35–37.
22 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A); Plumridge v Road Accident Fund 2012 JDR 1309 (ECP).
23 Lampert v Hefer NO 1955 (2) SA 507 (A).
24 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 778.
25 1996 (3) SA 262 (A) at 271–272.
26 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 779.
27 1955 (2) SA 507 (A).
28 At 514H.
29 Section 1(1)(a).
30 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 778.
31 1973 (4) SA 764 (A).
32 At 779–781.
33 Roux v Hattingh 2012 (6) SA 428 (SCA) para 43.
34 1987 (2) SA 694 (O).
35 At 702.
36 1976 (3) SA 484 (A).
37 1987 (2) SA 694 (O).
38 Roux v Hattingh 2012 (6) SA 428 (SCA).
39 Para 42.
40 Para 43.
41 Para 28.
42 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 779.
43 Santam Insurance Co Ltd v Vorster 1973 (4) SA 764 (A) at 779.
44 1908 TS 575.
45 1908 TS 575.
46 1995 (1) SA 30 (A).
47 1999 (1) SA 982 (SCA).
48 Government of the Republic of South Africa v Fibre Spinners & Weavers (Pty) Ltd 1978 (2) SA 794 (A) at 807.
49 Section 51(1)(c)(i) and 51(3). See also the discussion of exemption clauses in section 12.3 below.
50 See Maimela v Makhado Municipality 2011 (6) SA 533 (SCA) para 16.
51 Maimela v Makhado Municipality 2011 (6) SA 533 (SCA) para 19.
52 1975 (2) SA 85 (SWA).
53 [2010] 1 All SA 19 (SCA).
54 S v Kibi 1978 (4) SA 173 (E) at 179; S v Bradbury 1967 (1) SA 387 (A) at 393 and 404 (in this case, the view was
expressed that a person who voluntarily joins a violent gang cannot rely on a situation of necessity if he is later forced
by the gang to commit murder).
55 R v Mahomed 1938 AD 30, where the accused had kidnapped a girl and was later attacked by a group of people who
wanted to free the girl.
56 See S v Bradbury 1967 (1) SA 387 (A) at 390–393 and 404.
57 (1884) 14 QBD 273 DC.
58 1972 (3) SA 1 (A).
 59 1990 (1) SA 512 (C) at 526D–E.
60 See Ntsomi v Minister of Law and Order 1990 (1) SA 512 (C) at 526–527.
61 R v Ndara 1955 (4) SA 182 (A).
62 See section 10.5.
63 R v Van Vuuren 1961 (3) SA 305 (E).
64 Ntai v Vereeniging Town Council 1953 (4) SA 579 (A); Ex Parte Die Minister van Justisie: In re S v Van Wyk 1967
(1) SA 488 (A); S v Mogohlwane 1982 (2) SA 587 (T).
65 S v Mogohlwane 1982 (2) SA 587 (T); S v Kibi 1978 (4) SA 173 (E).
66 R v Patel 1959 (3) SA 121 (A).
67 S v Mogohlwane 1982 (2) SA 587 (T).
68 Cf R v K 1956 (3) SA 353 (A).
69 Ntanjana v Vorster and Minister of Justice 1950 (4) SA 398 (C).
70 R v Van Vuuren 1961 (3) SA 305 (E).
71 Chetty v Minister of Police 1976 (2) SA 450 (N) at 455–456; Thabethe v Minister of Police 1981 (3) SA 569 (D) at
573.
72 Ntsomi v Minister of Law and Order 1990 (1) SA 512 (C) at 527–528.
73 See R v Zikalala 1953 (2) SA 568 (A); R v Molife 1940 AD 202; R v Patel 1959 (3) SA 121 (A); Ntsomi v Minister of
Law and Order 1990 (1) SA 512 (C) at 530.
74 Ex Parte Die Minister van Justisie: In re S v Van Wyk 1967 (1) SA 488 (A) at 501; Ntsomi v Minister of Law and
Order 1990 (1) SA 512 (C) at 526.
75 1950 (4) SA 398 (C) at 406A–D. See also S v Ntuli 1975 (1) SA 429 (A) at 437E, and Ntsomi v Minister of Law and
Order 1990 (1) SA 512 (C) at 527–528.
76 1914 AD 273 at 286.
77 Cf Ntanjana v Vorster and Minister of Justice 1950 (4) SA 398 (C) at 406A–D; S v Ntuli 1975 (1) SA 429 (A) at
437E; Ntsomi v Minister of Law and Order 1990 (1) SA 512 (C) at 526–527.
78 Ntanjana v Vorster and Minister of Justice 1950 (4) SA 398 (C); Ntsomi v Minister of Law and Order 1990 (1) SA
512 (C) at 527–528.
79 2001 (4) SA 854 (W).
80 Para 16.
81 2006 (4) SA 150 (SCA).
82 2006 (4) SA 150 (SCA) at 158–159.
83 Ex Parte Die Minister van Justisie: In re S v Van Wyk 1967 (1) SA 488 (A) at 496–497.
84 Ex Parte Die Minister van Justisie: In re S v Van Wyk 1967 (1) SA 488 (A) at 497.
85 Ex Parte Die Minister van Justisie: In re S v Van Wyk 1967 (1) SA 488 (A) at 498.
86 1967 (1) SA 488 (A) at 498.
87 At 504–505.
88 According to D 48.5.22.3 a father, A, will not be liable for assaulting a man whom he catches having intercourse with
his (A’s) daughter. Voet, as translated by Gane, Commentary on the Pandects XLVII.10.20 (vii), states as follows:
It is not unlike this (self-defence) when on being provoked a person has retaliated with a wrong, since
wrongs appear in that way to have been set off in so far as a civil action could have been brought. The
retaliation of a wrong does not involve a wrong, since to suffer what you have done is no wrong, and
he who has meant on being challenged to avenge himself is to be pardoned.
In the following cases, provocation was accepted as a defence: Powell v Jonker 1959 (4) SA 443 (T) at 445–446;
Dzvairo v Mudoti 1973 (3) SA 287 (RA); Bennett v Minister of Police 1980 (3) SA 24 (C); Bester v Calitz 1982 (3)
SA 864 (O) at 875 and 880–881; Wapnick v Durban City Garage 1984 (2) SA 414 (D) at 419–420.
89 Bester v Calitz 1982 (3) SA 864 (O) at 876–877.
90 Jeftha v Williams 1981 (3) SA 678 (C).
91 Powell v Jonker 1959 (4) SA 443 (T); Wessels v Pretorius NO 1974 (3) SA 299 (NC) at 301; Winterbach v Masters
1989 (1) SA 922 (E) at 925.
92 Powell v Jonker 1959 (4) SA 443 (T) at 444–445; Winterbach v Masters 1989 (1) SA 922 (E) at 925.
93 Mordt v Smith 1968 (4) SA 750 (RA); Dzvairo v Mudoti 1973 (3) SA 287 (RA); Bester v Calitz 1982 (3) SA 864 (O)
at 875 and 878–881; Wapnick v Durban City Garage 1984 (2) SA 414 (D) at 419–420.
94 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 829.
95 1914 TPD 102 at 104.
96 1982 (3) SA 864 (O) at 880.
97 Bester v Calitz 1982 (3) SA 864 (O) at 875 and 878–881.
98 Powell v Jonker 1959 (4) SA 443 (T) at 445; Dzvairo v Mudoti 1973 (3) SA 287 (RA).
99 Powell v Jonker 1959 (4) SA 443 (T) at 446; Bennett v Minister of Police 1980 (3) SA 24 (C) at 31–32.
100 Blou v Rose Innes 1914 TPD 102; Bantjes v Rosenberg 1957 (2) SA 118 (T); Bennett v Minister of Police 1980 (3) SA
24 (C) at 31–32, but see D 48.522.3, stating that a father, A, will not be liable for assaulting a man whom he catches
having intercourse with his (A’s) daughter.
101 Powell v Jonker 1959 (4) SA 443 (T) at 446; Bennett v Minister of Police 1980 (3) SA 24 (C) at 31–32.
102 Kernick v Fitzpatrick 1907 TS 389 at 394; Kirkpatrick v Bezuidenhout 1934 TPD 155 at 158–159. Cf also Jeftha v
Williams 1981 (3) SA 678 (C) at 683, where Grosskopf J stated:
My own sense of what is reasonable leads me to consider that it should not be lawful for a person to
publish defamatory matter about another merely because he loses his temper, whatever the
provocation may be.
In this case, it was held that provocation can exclude fault in the form of intention rather than wrongfulness.
103 Mulvullha v Steenkamp 1917 CPD 571 at 573; Powell v Jonker 1959 (4) SA 443 (T) at 446.
104 Union Government (Minister of Railways) v Sykes 1913 AD 156; Johannesburg Municipality v African Realty Trust
Ltd 1927 AD 163; East London Western Districts Farmers’ Association v Minister of Education and Development
Aid 1989 (2) SA 63 (A); Simon’s Town Municipality v Dews 1993 (1) SA 191 (A); Government of the Republic of
South Africa v Basdeo 1996 (1) SA 355 (A).
105 Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 175; Sambo v Milns 1973 (4) SA 312 (T) at
320; During NO v Boesak 1990 (3) SA 661 (A) at 673.
106 Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 172–173; East London Western Districts
Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA 63 (A) at 70.
107 Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163, at 172; Simon’s Town Municipality v Dews 1993
(1) SA 191 (A) at 195–196.
108 Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 171–172.
109 See Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 172–173; Breede River (Robertson)
Irrigation Board v Brink 1936 AD 359 at 366.
110 East London Western Districts Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA 63
(A) at 72.
111 1927 AD 163.
112 1936 AD 359.
113 1938 AD 195.
114 1957 (1) SA 312 (A).
115 See East London Western Districts Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA
63 (A) at 68.
116 See Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 172–173; Bloemfontein Town Council v
Richter 1938 AD 195 at 231.
117 Simon’s Town Municipality v Dews 1993 (1) SA 191 (A) at 196.
118 Johannesburg Municipality v African Realty Trust Ltd 1927 AD 163 at 172–173; Breede River (Robertson) Irrigation
Board v Brink 1936 AD 359 at 366.
119 Matlou v Makhubedu 1978 (1) SA 946 (A) at 958; Govender v Minister of Safety and Security 2001 (4) SA 273
(SCA); Ex Parte Minister of Safety & Security: In re S v Walters 2002 (4) SA 613 (CC).
120 See May v Udwin 1981 (1) SA 1 (A) at 18, where it was held that public policy and sound administration of justice
require that a magistrate, in discharging his judicial duties ‘should be able to speak his mind freely without fear of
incurring liability for damages for defamation’.
121 See, generally, Basner v Trigger 1945 AD 22; Ingram v Minister of Justice 1962 (3) SA 225 (W); May v Udwin 1981
(1) SA 1 (A); Moeketsi v Minister van Justisie 1988 (4) SA 707 (T).
122 Minister of Justice and Constitutional Development v X [2014] 4 All SA 586 (SCA); 2015 (1) SA 25 (SCA) paras 39
–53.
123 See Moeketsi v Minister van Justisie 1988 (4) SA 707 (T) at 713; May v Udwin 1981 (1) SA 1 (A) at 19.
124 1990 (3) SA 466 (BG) at 480.
125 R v Smith (1900) 17 SC 561 at 567; S v Banda 1990 (3) SA 466 (BG) at 480.
126 R v Arlow 1960 (2) SA 449 (T) at 452; R v Van Vuuren 1944 OPD 35 at 38.
127 R v Smith (1900) 17 SC 561 at 567; S v Mule 1990 (1) SACR 517 (SWA); S v Banda 1990 (3) SA 466 (BG) at 480.
128 S v Banda 1990 (3) SA 466 (BG) at 496.
129 1972 (3) SA 1 (A).
130 Minister van Polisie v Ewels 1975 (3) SA 590 (A).
131 R v Janke and Janke 1913 TPD 382 at 385 and 392; Tshabalala v Jacobs 1942 TPD 310 at 313; Du Preez v Conradie
1990 (4) SA 46 (BG) at 51.
132 R v Scheepers 1915 AD 337 at 338; R v Schoombee 1924 TPD 481; R v Le Maitre and Avenant 1947 (4) SA 616 (C);
R v Muller 1948 (4) SA 848 (O); Hiltonian Society v Crofton 1952 (3) SA 130 (A); Du Preez v Conradie 1990 (4) SA
46 (BG) at 51.
133 R v Muller 1948 (4) SA 848 (O) at 860.
134 Du Preez v Conradie 1990 (4) SA 46 (BG) at 53.
135 R v Le Maitre and Avenant 1947 (4) SA 616 (C) at 621–622.
136 S v Lekgathe 1982 (3) SA 104 (B) at 109.
137 R v Scheepers 1915 AD 337 at 338; R v Roux 1932 OPD 59 at 61; R v Jacobs 1941 OPD 7 at 9 and 10; Du Preez v
Conradie 1990 (4) SA 46 (BG) at 53.
138 R v Janke and Janke 1913 TPD 382 at 385; Du Preez v Conradie 1990 (4) SA 46 (BG) at 51–52.
139 R v Janke and Janke 1913 TPD 382 at 388.
140 R v Janke and Janke 1913 TPD 382 at 385.
141 Section 12(1)(e) of the Constitution of the Republic of South Africa Act 108 of 1996.
142 See, generally, R v Janke and Janke 1913 TPD 382 at 385–386; Du Preez v Conradie 1990 (4) SA 46 (BG) at 51–52.
143 R v Janke and Janke 1913 TPD 382 at 385.
144 Hiltonian Society v Crofton 1952 (3) SA 130 (A).
145 For instance, where it is physically possible to prevent harm, but so costly that it would be unreasonable in terms of the
legal convictions of the community to expect preventative conduct, as in Regal v African Superslate (Pty) Ltd 1963
(1) SA 102 (A).
 Chapter 11

Liability in contract excluding action in

delict

11.1 Introduction

11.2 Why does concurrence of actions in contract and delict matter?

11.3 Exclusively delictual actions arising from a contractual relationship

11.3.1 Delictual actions arising from a pre-contractual context
11.3.2 Contractual duties giving rise to exclusively delictual actions
11.3.3 Delictual actions arising from contractual relationships

11.4 When is there concurrence, and when does liability in contract

exclude an action in delict?
|

11.1 Introduction
A delict may be described as the culpable and wrongful causation of harm to another, while breach
of contract is regarded as the wrongful causation of harm through the breach of a contractual duty.
As discussed in further detail below, the same conduct may constitute a breach of contract as well
as a delict. In Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1 the Court
referred to the possibility of the concurrence of actions as follows:
Roman law recognized the possibility of a concursus actionum, ie the possibility that
different actions could arise from the same set of facts. … The same principles were
accepted and applied in Roman-Dutch law. … In modern South African law we are of
course no longer bound by the formal actiones of Roman law, but our law also acknowledges
that the same facts may give rise to a claim for damages ex delicto as well as one ex
contractu, and allows the plaintiff to choose which he wishes to pursue.
It follows that a plaintiff may be presented with a choice between instituting a delictual or a
contractual remedy. As the Court stated, Roman law recognised a concursus actionum between
contract and delict in several instances. However, the consequences of such concurrence were
never mapped out clearly and the Roman lawyers did not identify a principled basis upon which to
choose between the different remedies.
As is evident from case law, the concurrence of contractual and delictual actions in modern
South African law is particularly relevant in cases that involve liability for the negligent
performance of professional services. The contractual relationship between a professional person
and his or her client generally involves an implied duty to act with reasonable care and skill. For
instance, a medical practitioner has, at least, an implied contractual duty to provide his or her
professional services with the necessary skill, diligence and care that is expected of someone in that
profession. At the same time, however, it may be said that a medical practitioner owes his or her
patient a similar legal duty not to cause him or her any bodily harm in an unjustifiable manner. In
the event that a medical practitioner culpably and wrongfully infringes his or her patient’s bodily
integrity, it would therefore constitute a breach of his or her contractual duty as well as a delict.
Against this background, the following questions require consideration:
• Why does concurrence of actions in contract and delict matter?
• When does an exclusively delictual action arise from a contractual relationship?
• When is there concurrence, and when does the existence of an action in contract exclude an
action in delict?

TERMINOLOGY In the broadest sense of the word, concurrency refers to the

notion that the same factual circumstances may give rise to two or
more obligations at the same time. However, it may be helpful to
remember that the concept may also be given different, narrower
meanings. A distinction may be drawn between the concurrency of
remedies and the concurrency of liability. For instance, a plaintiff
may have a concurrency of remedies when, in the same factual
scenario, one specific wrongdoer’s culpable causation of harm
results in a breach of contract as well as a delict. For instance,
where a doctor negligently and wrongfully causes harm to a
patient during the course of a medical operation. A plaintiff may
also have more than one remedy available to him or her, which he
or she may institute against two or more defendants. Seen from
the side of the defendant, concurrency of liability refers to a
situation where one defendant’s wrongful and culpable causation
of harm may expose him or her to both contractual and delictual
liability with regards to one, specific plaintiff. This would be the
case in the event that a medical practitioner negligently performs a
medical operation. Of course, the defendant may also incur
delictual and contractual liability with regards to more than one
plaintiff, that is, one plaintiff may elect to institute a delictual
remedy against him or her whereas another plaintiff may opt for a
contractual remedy.

11.2 Why does concurrence of actions in contract and

delict matter?
The fact that a plaintiff may have an action in delict as an alternative to a contractual action is a
matter of practical significance, because of the following important differences between actions for
breach of contract and actions in delict:
• The purpose of remedies based on breach of contract is either to enforce the contract or to
obtain compensation for the non-fulfillment of its terms. The purpose of a remedy based on
delict is to obtain compensation for harm caused by the infringement of a right or breach of a
duty.
• In delict, there is a right of recovery for patrimonial or non-patrimonial harm (such as pain and
suffering, or injury to reputation), whereas in contract there is no right of recovery for non-
patrimonial harm.2
• Fault is generally a requirement for delictual liability, whereas liability for breach of contract is
generally not based on fault. For example, fault is irrelevant to liability for breach of warranty.
• The Apportionment of Damages Act 34 of 1956 applies where the action for damages is based
on delict, but not where it is based on contract.3 Nowadays contracts for professional work
often incorporate the Apportionment of Damages Act by agreement, so that contributory
negligence is also taken into account in the event of a contractual claim for damages.
Thoroughbred Breeders’ Association v Price Waterhouse4
This case concerned the contractual liability of auditors for negligently failing to discover that
the financial manager of a company (their client) was involved in large-scale theft from the
company. It appeared that the company itself was careless in failing to properly supervise the
activities of the financial manager, despite being aware that the manager had previously been
convicted of theft. Both forms of negligence contributed to the company’s loss, but the Court
held that the company’s fault was comparatively negligible, and that the Apportionment of
Damages Act does not apply to contractual claims for damages. The concepts of contributory
negligence or prevalence of fault historically relate to the law of delict, and the Apportionment
of Damages Act was intended to remedy the all-or-nothing approach in delictual actions.
Therefore, the auditors were liable in full.

• The time when the extent of the harm is calculated for awarding for breach of contract
damages may be different from the time when the harm occurred for the purposes of delict.
• Claims in delict and contract may be subject to the jurisdiction of different courts.
• The extent of vicarious liability in delict is different from that in contract.
• Claims in delict and contract are subject to different rules of private international law.
• Contractual claims are actively transmissible, whereas some delictual claims are not.
• Where actions in both contract and delict are available based on the same facts, the onus of
proving or disproving negligence may differ according to the choice of action, for example,
damage done to goods left with another person.
• There is also an important difference between the values that underlie contract and delict.
Contract law is mainly based on the notion that people should be free to agree on the details of
their commercial relationship, with the result that a contracting party also accepts the risks of
the contract. The law of delict, on the other hand, imposes liability on the basis of standards
imposed by operation of law. Failing to act as a reasonable person would have done in the
circumstances can result in liability for negligence. One can regard the reasonable-person
standard as the symbolic point at which the law of delict limits freedom of action.5 Whereas
the law of contract largely deals with voluntarily assumed obligations, the law of delict
imposes obligations in accordance with the general societal norm of reasonableness.
11.3 Exclusively delictual actions arising from a
contractual relationship
In some instances, the existence of a contractual relationship can give rise to an exclusively
delictual action. This may occur in a pre-contractual context, where one person makes certain
representations to another with the view to inducing the latter to conclude the contract. Or, when a
contract has already been concluded, it may be the case that one party negligently performs his or
her contractual duty to the other, which conduct may provide the basis for delictual liability. Lastly,
a contractual relationship between two parties may also give rise to a delictual action, without the
contract having been breached by either party.

11.3.1 Delictual actions arising from a pre-contractual context

Pre-contractual liability in delict can arise where a party to a potential contract in the pre-
contractual phase negligently misrepresents facts inducing another party to enter into the contract.6
This is not a ‘true’ concurrence of contract and delict. This is because the delictual liability for
harm occurs in the pre-contractual phase and is not an alternative to contractual liability, unless the
representation is reinforced in the form of a contractual warranty.
In 1991, the Appeal Court recognised in Bayer South Africa (Pty) Ltd v Frost 7 that a negligent
misstatement that results in a contract can give rise to a delictual claim for damages. In this case,
the plaintiff purchased a new type of herbicide spray to use in his vineyards. An agent of the seller
recommended that the plaintiff spray the herbicide by helicopter and undertook to supervise the
spraying operations. During spraying the herbicide drifted to adjacent lands and caused extensive
damage to crops. The seller was held liable in delict for his negligent misstatement that wrongfully
caused the plaintiff’s harm. The Court held that negligent misrepresentation is a recognised
delictual cause of action, which can also arise where the parties are in a contractual relationship.

11.3.2 Contractual duties giving rise to exclusively delictual actions

The negligent performance of contractual duties that causes harm to a person not involved in the
contractual relationship may give rise to a delictual action by this person, for example:
• Where compliance with electrical wiring regulations is incorrectly certified in terms of a
contract with the owner or seller of the property.8
• An incorrect sworn valuation provided to the property owner is presented to a lender to apply
for a mortgage loan.9
• A retaining wall is defectively designed, causing loss for a later owner.10
• A ‘disappointed beneficiary’ does not inherit under a will, because the testator’s legal adviser
had negligently failed to ensure that the will was properly signed.11
• Negligent performance under a contract to provide security services may give rise to liability
towards a third party for loss of or damage to property.12

These are examples of purely delictual actions that arise from a contractual relationship.
Concurrence of contract and delict is not in issue.

11.3.3 Delictual actions arising from contractual relationships

A contractual relationship, such as that between an employer and employee, may also give rise to a
legal duty not to cause financial loss, even though no breach of contract is involved. Joubert v
Impala Platinum Ltd 13 is an example of this type of situation. The case involved the administrative
duties of an employer in respect of a master insurance policy entered into between the employer
and an insurance company. The policy was to provide cover for the employer’s employees. In
terms of the policy, the employer had a contractual duty towards the insurer to give notice of an
employee’s claim as soon as possible after an accident and to send the claim to the insurer as soon
as possible. The employer failed to send the claim and as a result, the employee forfeited
compensation. The Court held that the employer was liable in delict towards the employee for
negligent omission to fulfill a legal duty towards the employee.

11.4 When is there concurrence, and when does liability in

contract exclude an action in delict?
In some cases there is concurrence of contract and delict, which allows the plaintiff a choice of
action. This choice is influenced by the differences between the two actions, referred to in the
previous section. The following cases have recognised concurrence:
• In a contract between lessor and lessee, the lessor can claim for property damage in either
contract or delict.14
• In a contract between doctor and patient, the patient can claim in delict for alleged negligent
performance of professional services that caused physical harm.15
• In a contract between attorney and client, the client can claim in delict where negligent
performance of professional services caused pure economic loss to the client.16 However, it is
important to note that in the leading case concerning an attorney’s professional liability to a
client, the client’s action was treated as contractual.17
• In a contract between owner and builder, the Supreme Court of Appeal has recognised that the
builder can be liable in delict to the owner.18
• The Supreme Court of Appeal has recognised delictual liability for a negligent misstatement
by a bank manager. The bank manager incorrectly informed a client that a certain amount of
money had been deposited in his account, inducing the client to part with diamonds, under the
impression that he had been paid for them.19
• A financial services company can be liable to a client in delict for the grossly negligent advice
of an employee, on the basis of vicarious liability. This is despite the existence of a contractual
relationship between the company and the client.20
• An employer owes a common-law duty to its employees to take reasonable care of their safety.
This includes a duty to protect them from psychological harm caused by, for example, sexual
harassment by co-workers. Legal and public policy considerations require that an employer
takes reasonable steps to prevent sexual harassment of its employees in the workplace and to
compensate the victim for any harm caused in this regard should it negligently fail to do so.
The fact that there was a contractual relationship between the employee and the employer does
not alter the position. Delictual and contractual actions that arise from the same set of facts can
concur.21
• Where one company bought spices from another for use in its food business, the Court held
that the plaintiff had an alternative claim in delict for pure economic loss caused by delivery to
it of spices containing a banned colourant.22 The Court accepted that the same facts may give
rise to a claim for damages ex contractu and alternatively ex delicto, but also that the breach of
a contractual duty is not per se wrongful for the purposes of Aquilian liability.23 The negligent
causing of pure economic loss is not prima facie wrongful, but in this case, policy
considerations, such as knowledge of the potential harm, the fact that the loss was single and
determinate, and the manufacturer’s general duty to take reasonable steps to ensure that
defective products do not reach the market, indicated wrongfulness for the purposes of
delictual liability.
• If a financial investment institution, tasked with the responsibility of investing a trust’s funds,
releases the trust funds to a third party under circumstances where it should have suspected
fraud on the part of the latter, the trustees of the trust’s delictual claim will be sustainable in
law.24 The loss of trust funds was triggered by the conclusion and execution of a sale of shares
agreement in terms of which several of the trust’s shareholders sold their shareholding in the
trust to Fidentia Holdings Ltd. Following the conclusion of the agreement, Fidentia’s directors
were appointed as the new board of directors of the trust. As the new management, the board
requested the financial institution to transfer the funds to the trust which, in turn, would
transfer the funds to Fidentia. This was eventually done, whereafter, through
maladministration and misappropriation, the funds were depleted. Prior to the sale of shares
agreement being concluded, the financial institution and the trustees of the trust entered into
two contracts. The Court held that the contractual relationship between the financial institution
and the trust did not bar the trustees from instituting a delictual claim for the pure economic
loss which it had suffered. Instead, the contractual relationship was the basis for the
application of legislation which imposed certain legal duties on the financial institution. The
Court thus held that the breach of these statutory duties amounted to the wrongful and culpable
causation of the trust’s pure economic loss and that the trustees’ delictual claim was good in
law.

Courts, therefore, recognise an overlap of delict and breach of contract in cases of property damage
or bodily harm. Courts may also recognise this overlap in some cases of pure economic loss, where
the delictual action is based on a general duty to prevent loss, as opposed to a duty deriving from a
specific term of a contract. However, in some cases, contract ‘trumps’ delict, so that the existence
of a contract between the parties is a defence to an action in delict.
Two important cases where the Supreme Court of Appeal refused to recognise actions in delict
within a contractual setting are discussed below.
Lillicrap, Wassenaar and Partners v Pilkington Bros (SA) (Pty) Ltd25
The plaintiff, Lillicrap, was a firm of structural engineers. The defendant, Pilkington, was a
manufacturer of glass products. In mid-1975, Pilkington formally appointed Lillicrap as
consulting engineers to design and supervise the construction of a glass plant on a particular
site. Salanc Contractors (Pty) Ltd was employed as the building contractor to construct the
plant. In mid-1976, Pilkington assigned its contract with Lillicrap to Salanc. As a result of the
assignment, there was no longer any direct contractual relationship between Pilkington and
Lillicrap. Instead, Lillicrap’s status was changed to that of a subcontractor for Salanc. When
the completed plant was put into operation, it became apparent, because of soil instability on
the site, that there were slight movements between crucial components in the plant. These
movements rendered the plant unsuitable for manufacturing glass. Pilkington sought to
recover the cost of fixing these defects from Lillicrap, on the basis that the defects resulted
from its professional negligence in designing and supervising the construction of the plant.
Two scenarios arose from these facts:
• At first there was a direct contractual nexus between the parties.
• Later there was no such direct contractual relationship between them.

The question was whether policy considerations favoured an extension of Aquilian liability in
either case. Grosskopf AJA, writing for the majority, held that there was no need for such
extension. Recognising an action in delict in a commercial context, where the parties had
chosen to govern their relationship in terms of a contract, could create ‘a trap for the unwary’.
It could also lead to avoidance of specific contractual terms relating to, for example,
arbitration of disputes and limitation of liability. Additionally, it could create uncertainty
regarding the standard of care required (a contractual or delictual standard). A key
consideration was that this case did not involve an infringement of rights of property or
 person, but only the infringement of a contractual duty to perform specific professional work
with due diligence. There was no ‘independent’ duty for the purposes of delictual liability. In
the view of most commentators, this judgment implies that only a contractual remedy is
available where pure economic loss (as opposed to physical damage or personal injury) is
caused by negligent performance of a contractual duty, notably in the relationship between a
client and a professional practitioner.26

Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd27
The appellants were trustees of a trust that leased and operated an aquarium. They claimed
damages in delict for pure economic loss that resulted from the negligent design by the
respondent structural engineers of the exhibit tanks at the aquarium. They alleged that the
respondent’s negligence began before the contract between them was concluded, but that
even at that stage, the respondent was under a legal duty to act without negligence when
deciding on an appropriate design for the tanks. The respondent excepted to the appellants’
particulars of claim on the basis that the facts pleaded failed to establish that the alleged legal
duty existed. The facts pleaded were that:
• Prior to the trust forming, a joint venture agreement had been concluded by two potential
investors in the aquarium project with the object of investigating the feasibility of the project.
• The project was contemplated by all concerned, including the joint venture and the
respondent.
• The aquarium would be developed and operated by a trust, which was yet to be formed.
• The respondent agreed to assist in investigating the feasibility of the project with a view to its
formal appointment in the event of the project’s going ahead.

The Court held that negligently causing pure economic loss was wrongful, and therefore
actionable, only where the defendant had been under a legal duty not to act negligently. The
Court determined whether such a legal duty existed by considering relevant public or legal
policy considerations that were consistent with constitutional norms. The appellants were
asking the Court to accept that a legal duty existed that up until then was not recognised by
our law. Therefore, they were asking the Court to extend the limits of delictual liability.
Whether the Court would oblige depended upon whether there were any considerations of
public or legal policy that required the extension. The approach of our courts is not to extend
the scope of the Aquilian action to new situations unless there are positive policy
considerations that favour the extension. However, in this case, there was no need for the
extension sought because:
• It was intended from the outset that if the project proceeded at all, it would be governed by a
contractual relationship that would be created once the trust was formed.
• It was foreseen from the outset that the trust could not possibly suffer any damages through
the negligent conduct of the respondent before the contract was concluded.

The trust could have protected itself against this risk of harm by inserting, either in the
agreement between the joint venture and the respondent, or in the contract of formal
appointment of the respondent, appropriate contractual stipulations that covered conduct
occurring before the trust was formed. There was no reason to extend the Aquilian action to
rescue a plaintiff who could have avoided the risk of harm by contractual means, but who
failed to do so.

In Nashua Mobile (Pty) Ltd v GC Pale CC t/a Invasive Plant Solutions,28 the plaintiff concluded a
contract with the defendant for the provision of cellphone related services. One of the defendant’s
outlets issued a duplicate SIM card to an unknown person, who continued to use the card to
fraudulently transfer a substantial amount of money out of the plaintiff’s bank account through a
series of internet banking transactions. The plaintiff argued that, in doing so, the defendant ‘failed
to adhere to the duty of care that it owed to the plaintiff by virtue of the contractual relationship’
insofar as it had failed to verify that the person to whom the card was issued was the rightful
possessor of all rights to the cellphone number that was connected to the SIM card.29 The Court
followed the reasoning in Lillicrap and held that the:
defendant would not have owed the plaintiff any duty of care if it did not have a cellphone
contractual relationship with the plaintiff.

Furthermore, it also held that the manner in which the cause of action was couched demonstrates:
clearly that the delictual claim derives from a failure to adhere to a duty of care that is owed
by reason of a contractual relationship between the parties that is of full force and effect …
It is thus clear that the facts of this case fall more readily into the Lillicrap than into the
Holtzhausen divide.30

Similarly, in Nortje v Fakie 31 the appellant and respondent had entered into a sale agreement in
terms of which certain immovable property was sold to the respondent. The Court had to consider
the validity of the notice given by the appellant’s attorney in terms of which the appellant attempted
to place the respondent in mora. Further, the Court was asked to determine whether the appellant’s
alternative delictual claim was sustainable in law. On this point, the Court adopted the reasoning of
Grosskopf AJA in Lillicrap and held that the recognition of such claim would ‘have the effect of
eliminating provisions, in the agreement, which the parties considered necessary or desirable for
their own protection.’32
One can draw the following conclusions from these and other cases on the question of when
one can raise an action in contract as a defence, excluding an action in delict:
• Where property damage or bodily injury is concerned, as opposed to pure economic loss,
concurrence of contractual and delictual actions is allowed.
• The question of whether there is a legal duty for delictual liability within a contractual
relationship is one of wrongfulness. The general criteria of reasonableness, boni mores, or
‘legal convictions of the community’, apply. In the final instance it is a question of policy.
• An important policy consideration is whether an action in delict within the contractual setting
is needed, as an alternative to a contractual action.
• Courts will not grant a delictual action if its effect is to avoid specific contractual terms
relating to, for example, arbitration of disputes and limitation of liability.
• Courts will not grant a delictual action if its effect is to create uncertainty regarding the
standard of care required (a contractual or delictual standard).
• Courts will not grant a delictual action unless there was an ‘independent duty’ for the purposes
of delictual liability. This ‘independent delict test’ does not mean that one should ignore the
contractual relationship, but it indicates that one can infer the duty without relying on a
specific term of the contract. In the Holtzhausen case 33 the Court interpreted the Lillicrap case
as follows:
Lillicrap decided that no claim is maintainable in delict where the negligence
relied on consists in the breach of a term in a contract.

• The Court should be satisfied that the plaintiff could not have protected against the harm in
question by appropriate contractual stipulations. Courts will not extend the Aquilian action to
rescue a plaintiff who could have avoided the risk by contractual means, but who failed to do
so. This consideration is particularly important where the parties failed to govern a particular
aspect or phase of their transaction by contract, as in the Trustees, Two Oceans Aquarium
Trust case. However, one cannot apply this consideration too widely, as illustrated by the
example of delictual liability for pre-contractual misrepresentation, in which case, the action in
 delict is not excluded because the party who was misled failed to reinforce the pre-contractual
representation in the form of a contractual warranty.

1 1985 (1) SA 475 (A) at 496D–G.

2 Administrator, Natal v Edouard 1990 (3) SA 581 (A) at 595–596.
3 Thoroughbred Breeders’ Association v Price Waterhouse 2001 (4) SA 551 (SCA).
4 2001 (4) SA 551 (SCA).
5 Galligan Jr ‘Contortions along the boundary between contracts and torts’ (1994–1995) 69(2) in Tulane Law Review
457 at 460.
6 Bayer South Africa (Pty) Ltd v Frost 1991 (4) SA 559 (A) at 570; Thatcher v Katz 2006 (6) SA 407 (C) para 16. See
also NSC Carriers & Forwarding CC v Hyprop Investments Ltd 2013 (1) SA 340 (GSJ), which confirms that a
delictual claim will lie where someone’s fraudulent misrepresentation induces another to enter into a contract and it
results in harm for the latter.
7 1991 (4) SA 559 (A).
8 EG Electric Co (Pty) Ltd v Franklin 1979 (2) SA 702 (E).
9 Perlman v Zoutendyk 1934 CPD 151.
10 Tsimatakopoulos v Hemingway, Isaacs & Coetzee CC 1993 (4) SA 428 (C).
11 Pretorius v McCallum 2002 (2) SA 423 (C).
12 Compass Motors Industries (Pty) Ltd v Callguard (Pty) Ltd 1990 (2) SA 520 (W); Longueira v Securitas of South
Africa (Pty) Ltd 1998 (4) SA 258 (W). In Viv’s Tippers (Edms) Bpk v Pha Phama Staff Services (Edms) Bpk t/a Pha
Phama Security 2010 (4) SA 455 (SCA); [2011] 1 All SA 34 (SCA) the Supreme Court of Appeal refused to
recognise a legal duty on the part of the provider of security services towards the owner of a vehicle stolen from
premises protected by the security company, inter alia on account of exclusion of liability in the service contract (to
which the owner of the vehicle was not a party).
13 1998 (1) SA 463 (BH).
14 Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A) at 496E–F.
15 Van Wyk v Lewis 1924 AD 438.
16 Rampal (Pty) Ltd v Brett Wills and Partners 1981 (4) SA 360 (D).
17 Mouton v Die Mynwerkersunie 1977 (1) SA 119 (A); and see also Bouwer v Harding 1997 (4) SA 1023 (SE), where
the liability of an attorney to a client was the subject of an action for breach of contract, and the attorney was found to
be in breach of his contractual duties by negligently failing to advise the client correctly on the tax implications of
structuring a sale of property in a particular manner. See also Mlenzana v Goodrick & Franklin Inc 2012 (2) SA 433
(FB), where an attorney’s failure to obtain the requisite information and timeously lodge the plaintiff’s claim against
the Road Accident Fund, which resulted in the prescription of the client’s claim, was regarded as a negligent breach of
the attorney’s contract of mandate. See also Steyn NO v Ronald Bobroff & Partners 2013 (2) SA 311 (SCA).
18 SM Goldstein & Co (Pty) Ltd v Cathkin Park Hotel (Pty) Ltd 2000 (4) SA 1019 (SCA) para 7.
19 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA) para 6.
20 Pinshaw v Nexus Securities (Pty) Ltd 2002 (2) SA 510 (C).
21 Media 24 Ltd v Grobler 2005 (6) SA 328 (SCA) para 69.
22 Freddy Hirsch Group (Pty) Ltd v Chickenland (Pty) Ltd 2011 (4) SA 276 (SCA).
23 Para 33.
24 Living Hands (Pty) Ltd v Ditz 2013 (2) SA 368 (GSJ).
25 1985 (1) SA 475 (A).
26 Midgley ‘The nature of the inquiry into concurrence of actions’ (1990) 107(4) SALJ at 621; Van Aswegen ‘Die Toets
vir die Bestaan van ’n “Onafhanklike Delik” in die Geval van Kontrakbreuk’ (1992) 55(2) THRHR 271 at 273–274.
27 2006 (3) SA 138 (SCA).
28 2012 (1) SA 615 (GSJ).
29 2012 (1) SA 615 (GSJ) para 16.
30 2012 (1) SA 615 (GSJ) paras 27–28.
31 2013 (1) SA 577 (KZP).
32 2013 (1) SA 577 (KZP) para 13.
33 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA) para 6.
 Chapter 12

Exemption clauses

12.1 Validity and effect

12.2 Interpretation

12.3 Exemption clauses and the Constitution of the Republic of South

Africa, 1996

|
12.1 Validity and effect
A prior agreement not to claim damages if another person’s conduct causes harm (pactum de non
petendo in anticipando) can exclude the recovery of damages. Such an agreement or exemption
clause in an agreement can provide a person with a complete defence against a claim for damages
as a result of anticipated harm that is later realised. This excludes liability for negligence in breach
of contract as well as delict.1 If the defendant had contractually excluded liability for negligently
caused harm, the plaintiff cannot evade this by suing in delict. However, this is only the case if the
defendant discharges the onus of proving that the terms of the exemption exclude liability in the
particular circumstances.2 The Consumer Protection Act 68 of 2008 now limits the scope of
contractual exemption clauses. For instance, section 51(1)(c) of the Act prohibits the exemption of
a supplier of goods or services from liability for any loss directly or indirectly attributable to gross
negligence of the supplier or any person acting for or controlled by the supplier.
The requirements for a valid agreement to exclude liability are the same as for any other
contract. The requirement of authorisation to contract on behalf of another person applies. This
means that one person cannot conclude such an exemption contract on behalf of another without
authorisation. So, for example, a prior agreement by a breadwinner that no claim will be made in
the event of his or her negligently caused injury or death, even if such an agreement binds his or her
estate and dependants, will not exclude the right of his or her dependants to claim compensation for
loss of support in the event of his or her death.3 An exemption clause that would have been
effective against a deceased person is not necessarily effective against his or her dependants. It is
also questionable whether courts will consider an exemption from liability for causing death as
valid.4
Courts appear to accept that a parent or guardian can conclude a pactum de non petendo in
anticipando on behalf of a minor child, for example, where a school or sports club requires this
from the parent or guardian to allow the child to take part in an educational activity or sport,5 or
where the terms for entry into an amusement park excludes liability.6 Although participation in an
educational activity or sport is to the child’s benefit, an agreement that excludes liability for
negligently caused injury to the child is not. In this case, the child should be entitled to restitutio in
integrum, to have the effects of the agreement set aside.7 The effect of such an agreement will
depend on the precise ambit of the agreement and courts tend to interpret such agreements
restrictively.8
Courts regard an agreement to exclude liability for harm caused intentionally as invalid, but
regard an agreement to exclude liability for harm caused by negligence, or even gross negligence,
as valid.9
Courts enforce exemption clauses even where the effect appears to undermine a basic purpose
of the contract, such as to be treated with due care in a hospital,10 or to store valuables safely in a
bank vault,11 or to provide security services.12

PAUSE FOR In Viv’s Tippers (Edms) Bpk v Pha Phama Staff Services (Edms) Bpk t/a
REFLECTION Pha Phama Security13 the plaintiff concluded a lease agreement with
Lone Rock, a construction company, in terms of which it leased several
of its trucks to Lone Rock. Subsequent to the conclusion of the lease
agreement, and on the insistence of the plaintiff, Lone Rock entered into
an agreement with the defendant, a security services company,
according to which the defendant was obliged to provide security
services at Lone Rock’s construction site. The agreement included an
exclusion clause in terms of which Lone Rock indemnified the defendant
‘against any claims from loss or damage or any other claim which may
arise out of the provision of [the defendant’s] services in terms of the
[agreement]’.14 Following the theft of one of its trucks from the premises,
the plaintiff instituted a delictual claim for damages against the defendant.
The Supreme Court of Appeal regarded the theft of the truck as pure
economic loss and thus had to determine wrongfulness.15 The Court
overruled earlier High Court decisions to the effect that an exclusion
clause is ‘totally irrelevant’ in establishing the delictual liability of a
contracting party towards a non-contracting third party.16 In contrast,
Lewis JA held that an exclusion clause is a relevant policy consideration
for determining wrongfulness:17

How can the contractual arrangement between the owner of

the premises and the security provider be irrelevant to the
question whether a duty should be imposed on the security
provider to third parties whose property is stolen? … The
terms of that contract must, in my view, play a role in
assessing what the convictions of the community would be
in relation to affording a claim for compensation to a non-
contracting party.

The alarming effect of the Court’s ruling is that a contractual relationship

between two contracting parties may negatively affect the rights of an
independent third party.
 • Is such a clause, which is concluded at the expense of a third party, intelligible
or desirable in law?
• What role should an exclusion clause play when a contracting party’s delictual
liability towards non-contracting parties are placed in question?

12.2 Interpretation
Where a defendant raises a contractual term as a defence to a claim founded in delict, the defendant
bears the onus of establishing the terms of the contract.18 The defendant also bears the onus to
establish that it did everything reasonably necessary to bring the term to the attention of the
plaintiff.19 The extent of the exclusion or limitation of liability depends on the interpretation of the
exemption clause. General principles of interpretation apply, but courts will scrutinise such clauses
carefully. In Van der Westhuizen v Arnold 20 Lewis AJA said the following in this regard:
There does not, therefore, appear to be any clear authority for a general principle that
exemption clauses should be construed differently from other provisions in a contract. But
that does not mean that courts are not, or should not be, wary of contractual exclusions,
since they do deprive parties of rights that they would otherwise have had at common law.
In the absence of legislation regulating unfair contract terms, and where a provision does
not offend public policy or considerations of good faith, a careful construction of the
contract itself should ensure the protection of the party whose rights have been limited, but
also give effect to the principle that the other party should be able to protect himself or
herself against liability insofar as it is legally permissible. The very fact, however, that an
exclusion clause limits or ousts common law rights should make a court consider with great
care the meaning of the clause, especially if it is very general in its application. This requires
a consideration of the background circumstances … and a resort to surrounding
circumstances if there be any doubt as to the application of the exclusion.

If there is no specific reference to negligence in the exemption clause, for example, in the case of a
so-called ‘owner’s risk’ clause, courts must consider whether the words are extensive enough in
their ordinary meaning to cover negligence on the part of the defendant. If fault in the form of
negligence is a requirement for liability, either in the case of delict or breach of contract (where the
contract expressly or by implication provides that negligence will constitute breach of contract),
courts will interpret an exclusionary clause (for example, an ‘owner’s risk’ clause) as providing no
defence against liability for which negligence is a requirement. This is, of course, unless the clause
expressly states that such liability will also be excluded.21 In case of doubt, courts will decide the
issue against the party relying on the exemption, as the person considered responsible for drafting
the contract (that is, against the proferens).22 If the words are extensive enough in meaning to cover
negligence, courts must consider the possibility of another ground of action that does not require
proof of negligence, for example, strict liability of the owner of an animal or a common carrier.23 It
must be clear that the parties intended the cause of action in issue to be covered by the exemption.24

12.3 Exemption clauses and the Constitution of the

Republic of South Africa, 1996
Exemption clauses, like other contractual clauses, are subject to the broad principle of legality, and
will be unenforceable if they do not agree with public policy. The values enshrined in the
Constitution reflect public policy and are therefore relevant to the validity of exemption clauses.
Afrox Healthcare Bpk v Strydom25
This case considered the effect of the Constitution on exemption clauses. The plaintiff had
been admitted to a hospital that the defendant owned, for an operation and post-operative
treatment. Upon admission, the plaintiff signed an agreement that contained an exemption
clause, providing that the patient:

… absolved the hospital and/or its employees and/or agents from all liability and
indemnified them from any claim instituted by any person (including a
dependant of the patient) for damages or loss of whatever nature (including
consequential damages or special damages of any nature) flowing directly or
indirectly from any injury (including fatal injury) suffered by or damage caused
to the patient or any illness (including terminal illness) contracted by the patient
whatever the cause/causes are, except only with the exclusion of intentional
omission by the hospital, its employees or agents.

The plaintiff alleged that it was also an unspoken term of this agreement that the defendant’s
nursing staff would treat him in a professional manner and with reasonable care. After the
operation, certain negligent conduct by a nurse led to complications setting in, which caused
the plaintiff to suffer damages. The plaintiff argued that the negligent conduct of the nurse
had constituted a breach of contract by the defendant and instituted an action for damages
suffered. The defendant relied on the exemption clause to avoid liability.
The plaintiff contended that the exemption clause was contrary to the public interest, that it
was in conflict with the principles of good faith or bona fides, and that the admission clerk
had had a legal duty to draw his attention to the relevant clause, which he had not done. The
public policy grounds that the plaintiff used for his attack on the validity of the exemption
clause were:
• The alleged unequal bargaining positions of the parties
• The fact that the defendant provided medical services, which involved the duty to provide
medical treatment in a professional and caring manner
• The wide ambit of the exemption, which could encompass even gross negligence on the part
of the nursing staff.

The plaintiff also relied on section 39(2) of the Constitution, which obliges every court, when
developing the common law, to promote the spirit, purport and object of the Bill of Rights, to
take into account the fundamental rights contained in the Constitution. The plaintiff’s
argument in this regard was that the exemption clause conflicted with the spirit, purport and
object of section 27(1) (a) of the Constitution, which guaranteed each person’s right to
medical care. It was, therefore, in conflict with the public interest, or alternatively, it was
unreasonable, unfair and in conflict with the principle of bona fides or good faith.
The Court noted that exclusionary and indemnity clauses should be interpreted
restrictively. Specific exclusionary clauses could be declared contrary to public policy and as
such unenforceable. The relevant considerations of public policy were no different from those
that apply to other contractual terms. There was no evidence that indicated that the plaintiff
had occupied a weaker bargaining position than the defendant during the conclusion of the
contract. The plaintiff had not relied on gross negligence by the nursing staff. So, the
question of whether the contractual exclusion of a hospital’s liability for damages caused by
the gross negligence of its nursing staff was in conflict with the public interest, was
accordingly not relevant. Even if that were the case, it would not mean the automatic
invalidity of the relevant clause. The provisions would probably rather have been restricted to
exclude gross negligence.
With regard to the constitutional argument – that a court first had to decide whether section
39(2) of the Constitution empowered and obliged – the Court considered constitutional
provisions that were not yet in operation when the contract between the parties began. For
the purposes of the judgment, the Court accepted in favour of the respondent that the
provisions of section 27(1)(a) of the Constitution had to be taken into account. However, the
 exemption clause did not conflict with that right, and the elementary and basic general
principle was that it was in the public interest that courts enforce contracts freely entered into
and seriously by parties with the necessary capacity. The contention that the exemption
clause was contrary to the public interest therefore failed.

Johannesburg Country Club v Stott26

The Court considered an exemption clause to exclude any liability on the part of the Club for
personal injury or harm to its members or their children whilst on the club premises, however
caused. One of the members of the Club was killed by lightning whilst playing golf and his
wife instituted a dependant’s action in which she sought damages for loss of support and
funeral costs. The Court issued a directive requesting that the legal representatives of the
parties provide argument on whether the exemption clause should be unenforceable on the
grounds that it violates public policy, because it is inconsistent with the constitutionally
entrenched right to life. However, despite the directive that it issued, the Court found that it
could decide the matter without having to resort to a constitutional enquiry. The Court held
that the terms of the exemption clause did not cover liability for loss of support as claimed by
the plaintiff.

Regulation 44(3)(a) of the Consumer Protection Act now states that a clause that excludes or limit
the liability of the supplier for death or personal injury caused to the consumer though an act or
omission of that supplier will be presumed to be unfair, subject to the provisions contained in
section 61 of the Act, which is discussed in detail in Chapter 37.
Barkhuizen v Napier27
The Constitutional Court considered the constitutional validity of exemption and limitation
clauses. The case involved a constitutional challenge to a time limitation clause in a short-
term insurance contract. The clause in question required the claimant to institute court
proceedings within ninety days after the insurance company had rejected the claim.
The majority of the Court held that the proper approach to a constitutional challenge is to
determine whether the time limitation clause is contrary to public policy based on
constitutional values, in particular, those in the Bill of Rights. The question was whether the
time limitation clause was contrary to public policy, and whether it allowed an adequate and
fair opportunity to seek the assistance of a court. The period of ninety days was not
inadequate or unfair. At the beginning of the ninety-day period, the plaintiff had all the
information he needed to sue the insurance company as he had already lodged his claim
with the insurance company, which they had rejected. Also, there was no evidence that the
contract was not freely concluded between persons with equal bargaining power or that the
insured was not aware of the clause. The insured waited for two years to sue after being told
of the rejection of his claim. However, he did not provide any reasons for not observing the
clause. In the circumstances, the Court found that the clause was not unfair or unreasonable.
In a minority judgment Sachs J held that considerations of public policy, based on the
Constitution, dictated that the time-bar clause, which limited access to courts, should not be
enforced. This was not merely because it was in small print, or because it was harsh on the
insured, but because the clause was in a standard form document, and generally failed to
comply with standards of notice and fairness, which contemporary notions of consumer
protection required in open and democratic societies. In another minority judgment
Moseneke DCJ, with whom Mokgoro J concurred, agreed with Sachs J that the clause was
against public policy and should be unenforceable. According to this judgment, the proper
approach was whether the clause clashed with public norms and whether the contractual
term was so unreasonable that it was against public policy. Moseneke DCJ held that the time
clause was, based on the facts of this case, unreasonably short and inflexible, with the result
that it precluded a reasonable and adequate opportunity to seek legal redress.

Naidoo v Birchwood Hotel28

 Naidoo checked into the Birchwood Hotel and signed a registration card which contained a
clause that exempted the hotel from harm arising from a guest’s injury, ‘whether arising from
fire, theft or any cause, and by whomsoever caused or arising from the negligence (gross or
otherwise) or wrongful acts of any person in the employment of the Hotel’.29 On the morning
of his departure, the plaintiff suffered serious bodily injuries after a heavy steel gate fell on top
of him.
Having found that the hotel was negligent, the Court had to determine whether the
exemption clause was contractually binding on the plaintiff. In this regard, the Court adopted
the reasoning in Barkhuizen v Napier,30 to the effect that a court may first examine whether a
term in a contract is objectively reasonable and, if it finds that it is, decide whether to enforce
the term in the particular circumstances.31 Heaton-Nicholls J said:32

In my view exemption clauses that exclude liability for bodily harm in hotels and
other public places have the effect, generally, of denying a claimant judicial
redress. As this question was not argued before me I make no finding on the
first enquiry. This court is, however, equipped to consider whether in the
particular circumstances of this case the exemption clause should be enforced,
even if the relevant exemption clause is not contrary to public policy. I now
proceed to deal with the circumstances of this case. Naidoo was a guest in a
hotel. To enter and egress was an integral component of his stay. A guest in a
hotel does not take his life in his hands when he exits through the hotel gates.
To deny him judicial redress for injuries he suffered in doing so, which came
about as a result of the negligent conduct of the hotel, offends against notions
of justice and fairness.

In summary, the Court did not make a finding on the general, objective validity of clauses that
exclude liability for negligently caused bodily injuries or death, but rather held that, in the
particular circumstances, the enforcement of the exclusion clause would be unfair and
unjust.33 As noted above, regulation 44(3)(a) of the Consumer Protection Act states that a
clause that excludes or limits the liability of the supplier for death or personal injury caused to
the consumer through an act or omission of that supplier (such as the one in Naidoo) will be
presumed to be unfair, subject to the provisions contained in section 61 of the Act.

In each of these cases, specific exemption clauses were examined in specific circumstances. Further
constitutional challenges to such clauses may follow, in particular where they are found in standard
contracts and where personal injury is involved. The focus is likely to be on the impact of such
clauses on the right to access of courts, as guaranteed in section 34 of the Constitution. The
Consumer Protection Act also further limits the scope of contractual exemption clauses in
transactions where the Act applies.34

1 Government of the Republic of South Africa v Fibre Spinners & Weavers (Pty) Ltd 1978 (2) SA 794 (A) at 807.
2 Durban’s Water Wonderland (Pty) Ltd v Botha 1999 (1) SA 982 (SCA) at 991B–C.
3 Jameson’s Minors v Central South African Railways 1908 TS 575; and see also Payne v Minister of Transport 1995
(4) SA 153 (C).
4 Johannesburg Country Club v Stott 2004 (5) SA 511 (SCA) paras 6, 9, 12 and 14–17.
5 See Minister of Education and Culture (House of Delegates) v Azel 1995 (1) SA 30 (A), where the Court decided that
the undertaking by a parent, which limited the school’s liability for damage to property or personal injury of the child,
did not have effect where there was negligence on the part of the responsible teacher.
6 See Durban’s Water Wonderland (Pty) Ltd v Botha 1999 (1) SA 982 (SCA), where it was held that a parent had
agreed on behalf of a child to conditions that excluded an amusement park owner’s liability for injury to a person
using the park’s facilities.
7 See Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 732.
8 Minister of Education and Culture (House of Delegates) v Azel 1995 (1) SA 30 (A).
 9 Government of the Republic of South Africa v Fibre Spinners & Weavers (Pty) Ltd 1978 (2) SA 794 (A) at 807; Sasfin
(Pty) Ltd v Beukes 1989 (1) SA 1 (A) at 15; First National Bank of SA Ltd v Rosenblum 2001 (4) SA 189 (SCA); Land
and Agricultural Development Bank of SA v Ryton Estates (Pty) Ltd 2013 (6) SA 319 (SCA). As noted above, section
51(1)(c) (i) of the Consumer Protection Act prohibits an agreement exempting a supplier of goods or services from
liability for any loss directly or indirectly attributable to gross negligence of the supplier or any person acting for or
controlled by the supplier.
10 Afrox Healthcare Bpk v Strydom 2002 (6) SA 21 (SCA).
11 First National Bank of SA Ltd v Rosenblum 2001 (4) SA 189 (SCA); ABSA Bank Ltd v Fouche [2002] 4 All SA 245;
2003 (1) SA 176 (SCA).
12 Viv’s Tippers (Edms) Bpk v Pha Phama StaffServices (Edms) Bpk t/a Pha Phama Security 2010 (4) SA 455 (SCA);
[2011] 1 All SA 34 (SCA). See further Mercurius Motors v Lopez 2008 (3) SA 572 (SCA) para 33:
An exemption clause … that undermines the very essence of the contract … should be clearly and
pertinently brought to the attention of a [contracting party].
13 2010 (4) SA 455 (SCA); [2011] 1 All SA 34 (SCA).
14 Para 14.
15 Para 5.
16 Compass Motors Industries (Pty) Ltd v Callguard (Pty) Ltd 1990 (2) SA 520 (W) at 529H–530F. See also Longueira v
Securitas of South Africa (Pty) Ltd 1998 (4) SA 258 (W).
17 Paras 11 and 13.
18 See Durban’s Water Wonderland (Pty) Ltd v Botha 1999 (1) SA 982 (SCA) at 991B–C:
The respondents’ claims were founded in delict. The appellant relied on a contract in terms of which
liability for negligence was excluded. It accordingly bore the onus of establishing the terms of the
contract. (The position would have been otherwise had the respondents sued in contract. See Stocks &
Stocks (Pty) Ltd v T. J. Daly & Sons (Pty) Ltd 1979 (3) SA 754 (A) at 762E–767C.).
19 Naidoo v Birchwood Hotel 2012 (6) SA 170 (GSJ) para 7; Motowest Bikes & ATVS v Calvern Financial Services
(138/13) [2013] ZASCA 196 para 10; Mercurius Motors v Lopez 2008 (3) SA 572 (SCA) para 33; Durban’s Water
Wonderland (Pty) Ltd v Botha 1999 (1) SA 982 (SCA) at 991D–J. See also section 49(1) of the Consumer Protection
Act 68 of 2008.
20 (414/2000) [2002] ZASCA 82; [2002] 4 All SA 331 (SCA) (29 August 2002) para 21.
21 Essa v Divaris 1947 (1) SA 753 (A) at 763–764 and 766–767; Hughes NO v SA Fumigation Co (Pty) Ltd 1961 (4) SA
799 (C) at 805; Bristow v Lycett 1971 (4) SA 223 (RA) at 236.
22 Durban’s Water Wonderland (Pty) Ltd v Botha 1999 (1) SA 982 (SCA) at 989G–I:
Against this background it is convenient to consider first the proper construction to be placed on the
disclaimer. The correct approach is well established. If the language of a disclaimer or exemption
clause is such that it exempts the proferens from liability in express and unambiguous terms, effect
must be given to that meaning. If there is ambiguity, the language must be construed against the
proferens. (See Government of the Republic of South Africa v Fibre Spinners & Weavers (Pty) Ltd 1978
(2) SA 794 (A) at 804C.) But the alternative meaning upon which reliance is placed to demonstrate the
ambiguity must be one to which the language is fairly susceptible; it must not be ‘fanciful’ or
‘remote’ (cf Canada Steamship Lines Ltd v Regem [1952] 1 All ER 305 (PC) at 310C–D). See also
section 4(4) of the Consumer Protection Act 68 of 2008.
23 Essa v Divaris 1947 (1) SA 753 (A) at 763–764 and 766–767; Government of the Republic of South Africa v Fibre
Spinners & Weavers (Pty) Ltd 1978 (2) SA 794 (A) at 807.
24 Bristow v Lycett 1971 (4) SA 223 (RA) at 235–240; Government of the Republic of South Africa v Fibre Spinners &
Weavers (Pty) Ltd 1978 (2) SA 794 (A) at 803–806; Lawrence v Kondotel Inns (Pty) Ltd 1989 (1) SA 44 (D) at 53
–54.
25 2002 (6) SA 21 (SCA) at 26–27. See also, on a non-variation clause in a lease, Brisley v Drotsky 2002 (4) SA 1
(SCA).
26 2004 (5) SA 511 (SCA).
27 2007 (5) SA 323 (CC).
28 2012 (6) SA 170 (GSJ).
29 Para 37.
30 2007 (5) SA 323 (CC).
31 Paras 52–54.
32 Paras 52–53.
33 Para 54. See also Deacon v Planet Fitness Holdings (Pty) Ltd 2016 (2) SA 236 (GP), where the Court discusses the
legal position relating to exclusion clauses in a constitutional dispensation. Because it was held that the plaintiff did
not prove wrongfulness, these remarks were merely obiter dicta.
34 Section 51.
Chapter 13

Prescription

13.1 Nature and effect of prescription

13.2 Prescription period applicable to delict

13.3 Beginning of prescription

13.4 Delay of prescription

13.5 Interruption of prescription by service of process

13.6
|
Interruption of prescription by acknowledgment of liability

13.7 Waiver of prescription

13.8 The Prescription Act and other legislation on prescription or

limitation

13.9 Procedure

13.10 Onus

13.11 Prescription and the Constitution

13.12 Institution of Legal Proceedings against Certain Organs of State Act

40 of 2002

13.13 Road Accident Fund Act 56 of 1996

13.14 Apportionment of Damages Act 34 of 1956

13.15 Compensation for Occupational Injuries and Diseases Act 130 of

1993
13.1 Nature and effect of prescription
Prescription deals with the effect of the passage of time on obligations, including obligations
arising from delict. After a certain period, it is no longer possible to enforce an obligation or debt
arising from a delict. The debt is then ‘extinct’, and this effect of time on the enforceability of a
debt is commonly referred to as ‘extinctive prescription’. The Prescription Act 68 of 1969 provides
for prescription periods in respect of different types of ‘debt’. The concept of a ‘debt’ is not defined
in the Act, but courts have stated that the concept should be understood in a wide and general sense,
to include any duty side of an obligation.1 The concept ‘debt’ includes the liability that arises from
a delict, and in the context of delict it refers to the obligation to pay damages for the harm caused
by the delict.
The courts have emphasised the difference between a ‘debt’ and a ‘cause of action’ in a
number of cases.2 A ‘cause of action’ is the factual basis or set of material facts that ‘begets’ the
plaintiff’s right of action and its correlative, the defendant’s debt.3

PAUSE FOR The prescription period

REFLECTION In terms of section 10 of the Prescription Act, the ending of the applicable
prescription period extinguishes a debt. However, two principles
embodied in the Act qualify the notion that prescription extinguishes a
debt:
• Payment can discharge a debt after prescription has taken effect.
• Courts do not take note of prescription unless the debtor raises it as a
defence.4

Arguably, therefore, the ending of the prescription period does not

extinguish the debt, but rather allows the debtor to raise a complete
defence against the creditor’s claim.5

13.2 Prescription period applicable to delict

Different prescription periods apply to different types of debt, as set out in section 11 of the Act.
For any debt, there is a prescription period of three years, unless a specific other period is provided
for by section 11 or by any other Act of Parliament. Debts that arise from delict fall within this
general three-year category, unless another period is provided for by an Act of Parliament.

13.3 Beginning of prescription

Prescription, in respect of a debt, begins to run as soon as the debt is due.6 A debt that arises from
delict is due as soon as it comes into being, when all the elements of the delict are present, or when
it is ‘owing and immediately payable’, ‘immediately claimable’, ‘immediately exigible at the will
of the creditor’, or ‘enforceable’.7 In The Master v IL Back & Co Ltd 8 the Court decide that the
words ‘debt is due’ in section 12(1) mean that there must be money due, which the creditor can
claim, for which an action can be brought against the debtor. Harm is one of the elements of delict,
and only after harm has occurred is the cause of action complete and the prescription period begins
to run.9 This means that the injured party is in a position to claim payment immediately, and that
the perpetrator of the delict does not have a defence to the claim for immediate payment. In Truter
v Deysel 10 the Court held that for purposes of prescription ‘cause of action’ meant every fact that
the plaintiff had to prove in order to succeed in his claim. It did not comprise every piece of
evidence that was necessary to prove those facts.11
In terms of the ‘once-and-for-all’ rule, a person has to claim for all damages that arise from a
delict that constitutes a single cause of action at the same time, including future or prospective
damage. Therefore, prescription, in respect of the debt that arises from such a delict, begins to run
as soon as some harm is caused. However, in the case of a continuing wrong that causes continuing
harm, courts recognise that the cause of action continually exists (and the debt to pay damages
remains due) over the period of time during which the harm occurs. They do this so that the
claimant can recover damages for the harm that occurred within the applicable prescription or
limitation period.12
Slomowitz v Vereeniging Town Council13
The plaintiff claimed damages for business losses suffered as a result of the unlawful closing
of a road, from February 1960 until 17 December 1963. Summons had been served on 24
March 1964. A limitation period applicable in this case required that the claimant bring the
action within six months of the cause of action arising. The Appellate Division held that,
based on the damage caused by the unlawful closing of the road, the claimant had a cause of
action against the defendant throughout the period that the road remained closed. However,
in view of the six-month limitation period, the plaintiff was only entitled to the damages that he
could prove he sustained during the period six months before the serving of the summons on
the defendant. This was for the period 25 September to 17 December, 1963.

The Act provides, in section 12(2), that if the debtor wilfully prevents the creditor from knowing
that the debt exists, prescription will not begin to run until the creditor becomes aware that the debt
exists. Courts have held that the word ‘wilfully’ in this subsection means ‘deliberately’ or
‘intentionally’; it does not mean ‘fraudulently’.14
In terms of section 12(3) of the Act, a debt is not deemed to be due until the creditor has, or
ought to have had knowledge of who the debtor is, and of the facts from which the debt arises.15
This provision applies to debts that arise from delict or any other source of obligation.16 The
provisions of section 12(3) only require that a creditor do what could reasonably be expected in the
circumstances of a reasonable person.17
Courts will judge that a creditor has knowledge of the debtor, and the facts from which the
debt arose, if he or she could have acquired it by exercising reasonable care. The debtordefendant
must allege and prove that the creditor had, or ought to have had the required knowledge on a
particular date. On the question of when the identity of a debtor can be said to be established,
Diemont JA in Gericke v Sack 18 held as follows:
In common parlance I apprehend that the identity of an individual can be said to
be determined when one or more of his characteristics is established so that he
is definitively recognisable or known … . It may be that in some cases the debtor
is so distinguished, or notorious, that no address is necessary, or in other cases
that his name is so commonplace that a detailed address is called for. Regard
will have to be had to the particular circumstances of each case, but for practical
purposes it seems to me that there should be sufficient information for the
process-server to be able to identify the debtor by name and address.

Van Zijl v Hoogenhout19

This case concerned a claim by the adult survivor of child abuse. The abuse occurred
between 1958 and 1967, and the plaintiff attained majority in 1973. The Court found that the
abuse affected the plaintiff’s ability to attribute blame to the abuser. The plaintiff did not
 acquire ‘meaningful’ knowledge of the wrongs against her for the purposes of prescription
until what the Court described as a:

progressive course of self-discovery finally removed the blindfold she had worn
since the malign influences … took over her psyche.

The defendant failed to establish, on a balance of probability, that knowledge of the facts from
which the claim arose was acquired by the plaintiff, and so prescription commenced to run,
before 1997.

Truter v Deysel20
The plaintiff had undergone eye surgery in 1993, but it was only in 2000 that he managed to
secure medical opinion to the effect that the procedure was performed negligently. For that
reason, he instituted the claim only in April 2000. The question was when had the plaintiff
acquired knowledge of the facts from which the debt arose. The Court held that in a delictual
claim, the requirements of fault and unlawfulness were not factual ingredients of the cause of
action, but were legal conclusions to be drawn from the facts.21 For purposes of prescription,
‘cause of action’ meant every fact that the plaintiff had to prove to succeed in his claim. It did
not comprise every piece of evidence that was necessary to prove those facts.22 The expert
opinion that indicated negligence in the performance of the medical procedure was not itself a
fact, but instead, evidence.23 The presence or absence of negligence is not a fact; it is a
conclusion of law that the Court draws based on all the circumstances of the specific case.
Section 12(3) of the Prescription Act requires knowledge only of the material facts from which
the debt arises for the prescriptive period to begin running. The subsection does not require
knowledge of the relevant legal conclusions, that is, that the known facts constitute
negligence, or of the fact that an expert opinion which supports such conclusions, exists.24

13.4 Delay of prescription

In terms of section 13(1) of the Act, the ending of extinctive prescription will be delayed in certain
circumstances. Where such circumstances or ‘impediments’ exist, the completion of prescription is
delayed. Such a delay occurs if the applicable period of prescription would have been completed
before, or on, or within one year after the day on which the relevant impediment has ceased to
exist. Then, the period of prescription will not be completed before a year has elapsed after the day
on which the relevant impediment has ceased to exist.25
The ending, or completion, of prescription will be delayed if:
• The creditor is a minor, is a mentally disturbed person, is a person under curatorship, or is
prevented by superior force, including any law or any order of court, from interrupting the
running of prescription as contemplated in section 15(1) 26
• The debtor is outside the Republic 27
• The creditor and the debtor are married to each other 28
• The creditor and the debtor are partners and the debt is one that arose out of the partnership
relationship 29
• The creditor is a juristic person and the debtor is a member of the governing body of this
juristic person30
• The debt is the object of a dispute subjected to arbitration31
• The debt is the object of a claim filed against the estate of a debtor who is deceased, against the
insolvent estate of a debtor, or against the company in liquidation
• The creditor or the debtor is deceased and an executor of the estate in question has not yet been
appointed.32
The Supreme Court of Appeal in ABP 4×4 Motor Dealers (Pty) Ltd v IGI Insurance Co Ltd 33
considered the implications of section 13. The Court indicated that if a three-year extinctive
prescription period has started to run against a creditor, and one of the ‘impediments’ listed in
section 13 comes into being after two-and-a-half years, but ceases to exist after the three years have
run, the creditor will have one more year (calculated from the date when the impediment ceased to
exist) within which to sue. If the impediment ceases to exist on the day when three years have run,
the same applies. If the impediment ceases to exist when two-and-three-quarter years have run, the
result is the same. In all three instances, the creditor will have another year (calculated from the
date when the impediment ceased to exist) within which to sue. However, if the impediment comes
into being after six months of the prescription period have run, and ceases to exist six months later,
when two years of the original prescription period still remain, the creditor is not given any
additional time in which to sue.
For example, in Jonker v Rondalia Assurance Corporation of SA Ltd34 the Court held that in
respect of a minor, prescription begins running during minority, but where the relevant period of
prescription has ended before the minor attains majority, prescription will not take effect before a
year has elapsed after the attaining of majority.

Figure 13.1 Prescription in action: Beginning, taking effect within a three-year period, and delay

13.5 Interruption of prescription by service of process

The Prescription Act provides for two types of interruption of prescription:
• Judicial interruption by service of process on the debtor
• Interruption by acknowledgment of liability.

Section 15(1) of the Act provides for judicial interruption of extinctive prescription ‘by the service
on the debtor of any process whereby the creditor claims payment of the debt’. The requirements
for effective judicial interruption are set out in subsections 15(2)–(6) of the Act.
For interruption of prescription to occur, the service of process on the debtor must be legally
effective. Mere issue of such process out of court is not sufficient. The service of process on the
debtor must also begin proceedings against the debtor in a legally effective manner. Where service
of the process is premature in terms of a statutory provision, for example, legal proceedings have
not begun, prescription will not be interrupted.35 Service of process in an action instituted by, or on
behalf of a person without locus standi, for example, by a minor without the assistance of his
guardian, by a company in liquidation prior to the appointment of a liquidator, or by or on behalf of
a person who is not compos mentis, does not start proceedings against the debtor in a legally
effective manner, and will therefore not interrupt prescription in terms of section 15(1), unless such
an action is ratified by the court. A summons served timeously may interrupt the running of
prescription even if it discloses no cause of action or is otherwise excipiable, because the defect can
be rectified.36
Where two separate causes of action exist between a creditor and a debtor, there is a separate
debt corresponding to each cause of action. Service of summons on the debtor will interrupt
prescription of both debts only if the summons is an effective method of starting legal proceedings
in respect of both debts.37
In a situation where a plaintiff wishes to amend his or her claim for damages, the defendant
could rely on prescription if the new claim relates to a new debt, and the relevant prescriptive
period has run. However, this is not possible if the new claim relates to the same debt, and merely
represents a fresh quantification of the original claim by the addition of a further item of damages.38
For the purposes of interruption, ‘any process whereby the creditor claims payment of the
debt’ includes a petition, a notice of motion, a rule nisi, a pleading in reconvention, a third party
notice referred to in any rule of court, and any document that begin legal proceedings.39 In Cape
Town Municipality v Allianz Insurance Co Ltd 40 the Court held that the service of process in which
the creditor claims a declaration that a debtor is liable to indemnify the creditor, interrupts
prescription in terms of section 15(1).
The rules on interrupting prescription as set out above are subject to section 15(2) of the Act.
This states that the interruption of prescription will lapse, and courts will deem the prescription not
interrupted, if the creditor does not successfully prosecute the claim to final judgment ‘under the
process in question’,41 ‘abandons the judgment’, or if ‘the judgment is set aside’.42 For example,
should absolution be granted against a plaintiff at the end of the case, or should an exception
against the summons succeed so that the plaintiff has to issue a fresh summons, the original
‘process in question’ is deemed not to have interrupted prescription.43
If the running of prescription is interrupted by service of a process and the creditor
successfully prosecutes the claim to final judgment and does not abandon the judgment, or if the
judgment is not set aside, prescription starts to run afresh on the day on which the judgment of the
court becomes executable.44
If the running of prescription is interrupted by the service of process under section 15(1) and
the debtor then acknowledges liability, prescription commences to run afresh from the day on
which the debtor acknowledges liability.45 If at the time when the debtor acknowledges liability, or
at any time thereafter, the parties postpone the due date of the debt, prescription commences to run
afresh from the date upon which the debt again becomes due.46
If the running of prescription is interrupted, and the creditor successfully prosecutes the claim
under the process to final judgment, and further, if the interruption does not lapse in terms of
section 15(2), prescription starts to run afresh on the day on which the judgment of the court
becomes executable.47

13.6 Interruption of prescription by acknowledgment of

liability
Extinctive prescription is interrupted by the debtor expressly or tacitly acknowledging liability.
‘Acknowledgment of liability’ is not defined in section 14(1) of the Prescription Act, and in
deciding whether there has been an acknowledgement of liability interrupting prescription, the
enquiry will be factual with regard to the intention of the debtor.48 It must be clear that the debtor
intended to acknowledge that he or she had incurred the obligation, that the debt existed, and that
he or she was liable for it.49
In Road Accident Fund v Mothupi 50 the Court set out the key elements of acknowledgement of
liability: an acknowledgment of liability is a matter of fact, not a matter of law, and the intention of
the debtor is an important factor; to interrupt prescription an acknowledgment by the debtor must
amount to both an admission that the debt is in existence and of liability; the admission must cover
at least every element of the debt and exclude any defence as to its existence; and one must be able
to say when the acknowledgment of liability was made, otherwise it would not be possible to say
from what day prescription commenced to run de novo.
The debtor must make the acknowledgment of liability in person or through an authorised
agent, and it must be made it to the creditor in person or to his or her agent.51

13.7 Waiver of prescription

Whether and how a debtor can renounce the right to rely on prescription, wholly or partially,
involve questions of public policy. Courts will be reluctant to give effect to a term in a standard
form contract whereby a debtor renounces the right to rely on prescription in advance.52 However, if
the parties specifically negotiate such a renunciation, on the basis of particular circumstances or
commercial considerations, the renunciation could be valid.53

13.8 The Prescription Act and other legislation on

prescription or limitation
The provisions of Chapter III of the Prescription Act apply to any debt that arises after the Act
came into effect.54 However, in terms of section 16(1) of the Act, the provisions that relate to
extinctive prescription only apply in that they are not inconsistent with the provisions of other
legislation that prescribes a specified period within which a claim is to be made or an action is to be
instituted in respect of a debt, or which imposes conditions on the institution of an action for the
recovery of a debt.55
The provisions of the Act bind the State.56

13.9 Procedure
A court may not of its own accord take notice of prescription.57
A party to litigation who invokes prescription must do so in a document filed of record in the
proceedings, and in an action the issue of prescription is raised by way of a special plea.58 The word
‘proceedings’ has a wide meaning, as has the reference to a party to litigation, so that a person may
rely on prescription in any proceedings, not only in an action.59

13.10 Onus
The onus is on the party that invokes prescription to plead and prove the facts that indicate that
prescription has taken effect.60 This includes the date of the start of the prescription period.61
However, the onus changes if the creditor alleges ignorance of either the identity of the debtor or
the facts from which the debt arises, or of both, in terms of section 12(3). In such instances the onus
is on the debtor to show when the creditor knew or is deemed to have known of the debtor’s
identity and the facts; and this burden of proof does not change merely because the facts happen to
be within the knowledge of the creditor.62
The creditor bears the onus to allege and prove that the completion of prescription was delayed
under the circumstances set out in section 13 of the Prescription Act,63 and also that the running of
prescription was interrupted through either an express or tacit acknowledgement of liability by the
debtor,64 or by service of a legal process.65

13.11 Prescription and the Constitution

Applying special notice and limitation periods was an important feature of actions against the State
and certain state organs before a new constitution was introduced in South Africa in the last decade
of the twentieth century. Once the interim Constitution Act 200 of 1993 and the final Constitution
had come into operation, the constitutionality of these short and onerous notice and limitation
periods was challenged, in terms of the constitutional guarantees of equality and access to courts.
In Mohlomi v Minister of Defence 66 the Constitutional Court declared the notice and limitation
provisions in section 113(1) of the Defence Act 44 of 1957 invalid, on the basis of inconsistency
with the constitutionally guaranteed right of access to courts. In Hans v Minister van Wet en Orde 67
the Court reached the same conclusion in respect of the validity of provisions similar to section 32
(1) of the Police Act 7 of 1958.
The Institution of Legal Proceedings against Certain Organs of State Act 40 of 2002
subsequently reformed the law on notice and limitation periods of actions against the State and
organs of state.

13.12 Institution of Legal Proceedings against Certain

Organs of State Act 40 of 2002
This Act was introduced to regulate and harmonise the prescription periods of debts for which
certain organs of state are liable, notably by the repeal or amendment of a large number of statutes,
as listed in the schedule to the Act. In terms of this Act, prescription periods for actions for
recovery of ‘debts’ from ‘organs of state’ as defined in section 1 have been brought in line with the
provisions of the Prescription Act.
The Act also provides for a uniform notice period and procedure for intended legal
proceedings against organs of state for recovering debts.68
Section 3(2) of the Act requires notice to be given within six months of the date on which the
cause of action against the organ of state arises, but section 3(4) allows a court to condone the
failure to give notice.69

13.13 Road Accident Fund Act 56 of 1996

The Road Accident Fund Act provides for the transfer of certain forms of liability to the Road
Accident Fund that would otherwise arise for the negligent driver or owner of a motor vehicle in
terms of common law. The Road Accident Fund is liable in cases of loss or damage that result from
bodily injury or death, according to certain requirements as set out in section 17 of the Act. Claims
against the Fund are subject to prescription provisions in section 23 of the Act, which provides as
follows:
23 Prescription of claim
(1) Notwithstanding anything to the contrary in any law contained, but subject to
subsections (2) and (3), the right to claim compensation under section 17 from
the Fund or an agent in respect of loss or damage arising from the driving of a
motor vehicle in the case where the identity of either the driver or the owner
 thereof has been established, shall become prescribed upon the expiry of a
period of three years from the date upon which the cause of action arose.
(2) Prescription of a claim for compensation referred to in subsection (1) shall not
run against-
(a) a minor;
(b) any person detained as a patient in terms of any mental health legislation; or
(c) a person under curatorship.
(3) Notwithstanding subsection (1), no claim which has been lodged in terms of
section 17 (4)(a) or 24 shall prescribe before the expiry of a period of five years
from the date on which the cause of action arose.

Section 23 provides for a two-tier prescription system. A three-year prescription period, calculated
from the date upon which the cause of action arose, applies for lodging a claim against the Fund in
terms of section 24 of the Act. If a claim is not lodged within this three-year period, prescription
takes effect and the claim is lost.70 If the claim is properly lodged within the three-year period, the
second tier of the prescription system comes into operation. To enforce the claim, if the Fund
denies liability or makes an offer unacceptable to the claimant, for example, the claimant must start
action by having a summons served on the Fund before the expiration of five years from the date on
which the cause of action arose, in terms of section 23(3).
In terms of section 23(2), prescription does not run against a minor, a person detained under
mental health legislation, or a person under curatorship. The subsection does not mention other
grounds for suspending the running of prescription on the basis of incapacity, such as
unconsciousness, but common law recognises such grounds.

Figure 13.2 Prescription of claim under Road Accident Fund Act

13.14 Apportionment of Damages Act 34 of 1956

In terms of the Apportionment of Damages Act, the period of extinctive prescription in respect of a
claim for a contribution against and between joint and several wrongdoers is twelve months. This is
calculated from the date of the judgment in respect of which a contribution is claimed or, where an
appeal is made against such judgment, the date of the final judgment on appeal.
The plaintiff or any joint wrongdoer who is not sued in that action may give notice of any
action, any time before the close of pleadings in such an action. Such a joint wrongdoer may
thereupon intervene as a defendant in that action.71 A plaintiff who fails to give such notice may not
proceed against the joint wrongdoer, except with permission from a court based on showing good
cause as to why such notice was not given.72

13.15 Compensation for Occupational Injuries and

Diseases Act 130 of 1993
A claim for compensation in terms of the Compensation for Occupational Injuries and Diseases Act
must be lodged by or on behalf of the claimant, in the prescribed manner, with the commissioner,
the employer, or the mutual association concerned, within twelve months after the date of the
accident or, in the case of death, within twelve months after the death.73

1 See, generally, Loubser Extinctive Prescription (1996) at 26–31; CGU Insurance Ltd v Rumdel Construction (Pty) Ltd
2004 (2) SA 622 (SCA) para 6I; Oertel v Direkteur van Plaaslike Bestuur 1983 (1) SA 354 (A) at 369C–D. The courts
have limited this wide concept of ‘debt’ in a few cases: in Makate v Vodacom (Pty) Ltd [2016] ZACC 13 the Court
decided that ‘debt’ implies a duty in the restricted sense of an obligation to pay money, deliver goods or render
services, and that earlier judgments indicating that the duty can relate to any kind of performance due under a contract
were wrong, with the result that a reciprocal obligation to negotiate in good faith, which could not be discharged by
unilateral action, but would require the active participation of both parties, is not a ‘debt’ subject to prescription; and
in Njongi v MEC, Department of Welfare, Eastern Cape 2008 (4) SA 237 (CC) para 42, the Constitutional Court
expressed doubt that prescription could legitimately be invoked if the ‘debt’ is an obligation to pay a social grant,
because the right to a social grant is a constitutional right; and in ABSA Bank Ltd v Keet 2015 (4) SA 474 (SCA) the
Court decided that a ‘debt’ in the form of a duty to restore property, correlative to a vindicatory claim, is not subject to
extinctive prescription.
2 CGU Insurance Ltd v Rumdel Construction (Pty) Ltd 2004 (2) SA 622 (SCA) para 6, and the cases cited there.
3 CGU Insurance Ltd v Rumdel Construction (Pty) Ltd 2004 (2) SA 622 (SCA) para 6.
4 Section 17 of the Act.
5 Loubser (1996), Chapter 1.
6 Section 12(1) of the Act.
7 See, generally, Loubser (1996) at 51–52; Deloitte Haskins & Sells Consultants (Pty) Ltd v Bowthorpe Hellerman
Deutsch (Pty) Ltd 1991 (1) SA 525 (A) at 532H; Kotzé v Ongeskiktheidsfonds van die Universiteit van Stellenbosch
1996 (3) SA 252 (C) at 258H–262C.
8 1983 (1) SA 986 (A).
9 Evins v Shield Insurance Co Ltd 1980 (2) SA 814 (A) at 839C–G. In Oslo Land Corporation Ltd v The Union
Government 1938 AD 584, the spraying of excessively strong locust poison by a government agency caused the death
of cattle over the next three years, but it was held that prescription began to run when the first harm occurred. In John
Newmark and Co (Pty) Ltd v Durban City Council 1959 (1) SA 169 (N), a case concerning harm that occurred when
excavations alongside a wall caused it to collapse later, the Court held that prescription began to run when the first
subsidence occurred. In Beira v Vallet [2005] JOL 13588 (W) the Court accepted that prescription in respect of a
delictual debt for damages does not begin to run from the date of the culpable act or omission, if no loss or harm has
yet occurred (injuria sine damno). The delict does not become actionable without loss or harm. It is only when loss or
harm is suffered that the plaintiff’s cause of action becomes complete and prescription begins to run.
10 2006 (4) SA 168 (SCA).
11 Para 19 at 174H–175A.
12 On the concept of a continuing act causing continuing harm, see, generally, Loubser (1996) at 92–96; Slomowitz v
Vereeniging Town Council 1966 (3) SA 317 (A).
13 1966 (3) SA 317 (A).
14 Jacobs v Adonis 1996 (4) SA 246 (C) at 250J–251B.
15 In Geldenhuys NO v Diedericks 2002 (3) SA 674 (O) the claim arose from the collision between a car driven by the
defendant and the plaintiff’s bull, which died as a result. The plaintiff had been on holiday when the accident
occurred, but the defendant had contacted him telephonically about the accident. The defendant, at the request of the
plaintiff, sent a letter to the plaintiff setting out the details of the collision and the damage, for the plaintiff to hand
over to his insurers. The defendant also invited the plaintiff to contact him should he require any further information.
The words ‘facts from which the debt arose’ in section 12(3) do not mean every allegation made in the particulars of
claim, such as the grounds of negligence, but merely indicate the facts from which the right to institute action arose.
16 In Minister of Finance v Gore NO 2007 (1) SA 111 (SCA) paras 18–19, where knowledge of tender fraud was in
issue, Cameron JA and Brand JA stated that a mere opinion or supposition is not enough: there must be justified, true
belief. Belief, on its own, is insufficient. Belief that happens to be true is also insufficient. For there to be knowledge,
the belief must be justified. The following propositions are well established in our law: (a) Knowledge is not confined
to the mental state of awareness of facts that is produced by personally witnessing or participating in events, or by
being the direct recipient of first-hand evidence about them. (b) Knowledge includes a conviction or belief that is
engendered by or inferred from attendant circumstances. (c) Mere suspicion not amounting to conviction or belief
justifiably inferred from attendant circumstances does not amount to knowledge.
17 Gericke v Sack 1978 (1) SA 821 (A) at 832; Brand v Williams 1988 (3) SA 908 (C); Dube v Banana 1999 (1) BCLR
44 (ZH).
18 Gericke v Sack 1978 (1) SA 821 (A) at 829G–830A.
19 2005 (2) SA 93 (SCA).
20 2006 (4) SA 168 (SCA).
21 Para 17.
22 Para 19.
23 Para 20.
24 The Court referred in this regard to Loubser (1996) at 80 and the authorities cited there, in particular Evins v Shield
Insurance Co Ltd 1980 (2) SA 814 (A) at 838H–839A.
25 See, generally, ABP 4×4 Motor Dealers (Pty) Ltd v IGI Insurance Co Ltd 1999 (3) SA 924 (SCA) at 930B.
26 Hartman v Minister van Polisie 1983 (2) SA 489 (A).
27 Dithaba Platinum (Pty) Ltd v Erconovaal Ltd 1985 (4) SA 615 (T).
28 Section 13(1)(c).
29 On section 13(1)(d) see Laconian Maritime Enterprises Ltd v Agromar Lineas Ltd 1986 (3) SA 509 (D) at 513A–F;
Van staden v Venter 1992 (1) SA 552 (A).
30 Section 13(1)(e).
31 On section 13(1)(f) see Kilroe-Daley v Barclays National Bank Ltd 1984 (4) SA 609 (A).
32 Section 13(1)(h).
33 1999 (3) SA 924 (SCA) at 930 D.
34 1976 (2) SA 334 (E) at 336B–E.
35 Abrahamse & Sons v South African Railways and Harbours 1933 CPD 626 at 637.
36 Trans-African Insurance Co Ltd v Maluleka 1956 (2) SA 273 (A); Rooskrans v Minister van Polisie 1973 (1) SA 273
(T) at 274G–H; Sentrachem Ltd v Prinsloo 1997 (2) SA 1 (A) at 15H–16B.
37 Evins v Shield Insurance Co Ltd 1980 (2) SA 814 (A) at 842F–G; Park Finance Corporation (Pty) Ltd v Van Niekerk
1956 (1) SA 669 (T) at 673A–C; Erasmus v Grunow 1978 (4) SA 233 (O) at 245E.
38 Evins v Shield Insurance Co Ltd 1980 (2) SA 814 (A) at 836C–E.
39 Section 15(6).
40 1990 (1) SA 311 (C). See also Desai NO v Desai 1996 (1) SA 141 (A) at 147H–I.
41 In Chauke v President Insurance Co Ltd 1978 (2) SA 947 (W) it was held that the words ‘under the process in
question’ in section 15(2) meant ‘in terms of the Rules of Court’ governing the question (at 950E–G). See also Kuhn v
Kerbel 1957 (3) SA 525 (A) at 534A; Sieberhagen v Grunow 1957 (3) SA 485 (C) at 488–489.
42 See Van der Merwe v Protea Insurance Co Ltd 1982 (1) SA 770 (E) at 773C–E.
43 Titus v Union & SWA Insurance Co Ltd 1980 (2) SA 701 (Tks) at 704C–E; Union & SWA Insurance Co Ltd v
Hoosein 1982 (2) SA 481 (W) at 484A–E.
44 Section 15(4).
45 Section 15(3).
46 Section 15(3).
47 Section 15(4).
48 Agnew v Union & South West African Insurance Co Ltd 1977 (1) SA 617 (A) at 623A–C.
49 Estate Allie v Cape Town Municipality 1980 (1) SA 265 (C) at 268D; Eerste Nasionale Bank van Suidelike Afrika Bpk
v Vermeulen 1997 (1) SA 498 (O) at 503G–I.
50 2000 (4) SA 38 (SCA) paras 36–39.
51 Pentz v Government of the RSA 1983 (3) SA 584 (A); Markham v SA Finance & Industrial Co Ltd 1962 (3) SA 669
(A) at 676F.
52 Ryland v Edros 1997 (2) SA 690 (C) at 713H–I. The Court disagreed with the decision in Nedfin Bank Bpk v
Meisenheimer 1989 (4) SA 701 (T). See also Friederich Kling GmbH v Continental Jewellery Manufacturers, Speidel
GmbH v Continental Jewellery Manufacturers 1995 (4) SA 966 (C); ABSA Bank h/a Bankfin v Louw 1997 (3) SA
1085 (C) at 1090A–C.
53 See, generally, Loubser (1996) Chapter 8.
54 Section 16(1).
55 Meintjies NO v Administrasieraad van Sentraal-Transvaal 1980 (1) SA 283 (T) at 289B–D; Hartman v Minister van
Polisie 1981 (2) SA 149 (O) at 152A–E.
56 Section 19.
57 Section 17(1).
58 Section 17(2). See Grindrod (Pty) Ltd v Seaman 1998 (2) SA 347 (C) at 350I–J.
59 Lipschitz v Dechamps Textiles GmbH 1978 (4) SA 427 (C) at 431B–E; Rand Staple-Machine Leasing (Pty) Ltd v ICI
(SA) Ltd 1977 (3) SA 199 (W) at 201H.
60 Gericke v Sack 1978 (1) SA 821 (A) at 827 and 828C.
61 Gericke v Sack 1978 (1) SA 821 (A) at 827H–828A; Santam Ltd v Ethwar [1999] 1 All SA 252 (A); 1999 (2) SA 244
(SCA) at 256G–H.
62 Gericke v Sack 1978 (1) SA 821 (A); Mulungu v Bowring Barclays & Associates (Pty) Ltd 1990 (3) SA 694 (SWA);
Mokoena v Kraamwinkel NO [2009] JOL 24524 (GNP); Securefin Ltd v Sanlam Insurance Ltd [2006] JOL 18522 (C).
63 Regering van die Republiek van Suid-Afrika v South African Eagle Versekeringsmaatskappy Bpk 1985 (2) SA 42 (O)
at 47G; Kapeller v Rondalia Versekeringskorporasie van Suid-Afrika Bpk 1964 (4) SA 722 (T) at 728A; Naidoo NO v
Naidoo 2010 (5) SA 514 (KZP).
64 Section 14. See Pentz v Government of the RSA 1983 (3) SA 584 (A); Benson v Walters 1984 (1) SA 73 (A).
65 Du Bruyn v Joubert 1982 (4) SA 691 (W) at 695–696A.
66 1996 (12) BCLR 1559 (CC); 1997 (1) SA 124 (CC).
67 1995 (12) BCLR 1693 (C).
68 Sections 3–5.
69 Madinda v Minister of Safety & Security 2008 (4) SA 312 (SCA); Minister of Safety and Security v De Witt 2009 (1)
SA 457 (SCA).
70 Krischke v Road Accident Fund 2004 (4) SA 358 (W).
71 Sections 2(2)(a) and (b); and see Commercial Union Assurance Co Ltd v Pearl Assurance Co Ltd 1962 (3) SA 856
(E).
72 Section 2(4)(a).
73 Section 43(1)(a).
 Chapter 14

Omissions

14.1 Introduction

14.2 Wrongfulness

14.1 Introduction
Liability based on omission is more restricted
| than liability based on commission, because of public
policy. It would be socially and economically unduly restrictive, and therefore undesirable, to
enforce a wide and general duty to prevent harm to others. This approach could also involve
potentially indeterminate delictual liability, which would be against public interest.

14.2 Wrongfulness
To determine liability for an omission, one enquires whether a legal duty to prevent harm exists.
There is no general right to be protected from harm by another, and conversely, there is no general
legal duty to act positively to protect others, or to prevent harm to others. Courts recognise a duty to
act positively to prevent harm for the purposes of delictual liability only where failing to act
positively was unreasonable according to the boni mores or the legal convictions of the community.
The omission itself is not wrongful. The focus is on the whole of the causal sequence beginning
with failing to act and ending with the harmful consequences that could have been prevented by
positive action. Enquiring into wrongfulness involves applying the general criterion of
reasonableness. Courts must weigh up the interests of the persons involved and also take into
account the interests and convictions of the community at large.
In a line of cases before the decision of the Appellate Division in Minister van Polisie v Ewels
1
courts adhered to the view that they could impose liability for an omission only where the
defendant’s prior conduct created a risk of harm or a new source of danger and if the defendant
then failed to prevent the harm from occurring. Many of these cases dealt with a local authority’s
liability for injury to a person using a public road or other amenity (hence the reference to these
cases as ‘municipality cases’). The judgments in these cases often involved subtle distinctions
between situations in which the defendant simply failed to prevent harm, and situations where the
creation of a risk of harm or the introduction of a new source of danger preceded the failure.
Eventually, the Appellate Division broke away from this approach in the leading case of Minister
van Polisie v Ewels.2 In this case, the Court held that there can be delictual liability for a mere
omission, in other words, where an omission was not preceded by conduct that created a risk of
harm or that introduced a new source of danger. In Ewels the Court imposed liability in a situation
where the plaintiff was assaulted by an off-duty policeman in a police station in the presence of a
more senior policeman, who had failed to prevent the assault.
The Appellate Division’s decision in the Ewels case has been of great significance, not only in
respect of liability for omissions, but also in respect of the criteria for assessing wrongfulness
generally. The essential question is whether a legal duty exists to prevent harm to others, based on
reasonableness and public policy. In terms of the wide and evaluative criteria used by courts in this
regard, a legal duty exists where failing to prevent harm not only evokes moral indignation, but is
also regarded as so unreasonable, according to the boni mores or legal convictions of the
community, that liability should be imposed for the loss suffered. In addition to these wide and
general criteria, courts take into account the following:
• Policy considerations that indicate whether the law of delict should intervene (inter alia the
social or economic consequences of imposing liability, the availability of alternative remedies,
and the need for accountability of public bodies or officials)
• Relevant constitutional or other statutory rights and duties
• A grouping of factual circumstances that indicate a duty not to cause or to prevent harm in the
particular situation.

The ‘prior conduct’ approach to liability of local authorities for omissions was finally discarded in
Cape Town Municipality v Bakkerud.3 In this case, the Court accepted the general criterion of
reasonableness (the policy-based standard of boni mores or the legal convictions of the
community). The specific content of this general criterion is determined by taking into account a
grouping of factual considerations, such as the extent of the danger, the period of time for which it
existed, the resources of the public authority, and prior warning. The need to prove both the
existence of a legal duty to prevent harm and negligence, limits claims against public authorities.
Mostert v Cape Town City Council 4
This case involved a damaging leak from the defendant’s water main, and it illustrates the
factors that are relevant in cases that involve local authorities. The Court weighed the
technical evidence on the risk of the pipeline collapsing against the high cost of replacing it,
and judged that the council could not reasonably have been expected to replace the pipe.
The conduct of the city council was therefore not negligent. Noting that ‘an unbending
adherence to logic’ might dictate that wrongfulness is the prior enquiry, and the
reasonableness of requiring the council to replace the pipeline might also have been dealt
with under that heading, Schutz JA remarked that logic is one thing, utility sometimes
another. Both the requirements of wrongfulness and negligence must be met, and it does not
matter which of the two is determined first. One can determine wrongfulness on the
assumption that negligence will be proved, and vice versa. In many, if not most delicts that
involve property damage or injury, the issue of wrongfulness is uncontentious. This was such
a case. The essential question was whether the council was negligent in not preventing the
pipe from bursting. The requirement of wrongfulness (the general, policy-based criterion of
reasonableness, or the community’s sense of what the law ought to be) was not contentious.
The council exercised sole control over the pipeline and clearly had a legal duty to prevent
harm to people living or doing business in the vicinity of the pipeline.
 PAUSE FOR Wrongfulness
REFLECTION In Mostert v Cape Town City Council 5 Schutz JA said:

If the council was negligent in not preventing the 1990 burst

I have no doubt that the community’s sense of what the law
ought to be would demand that liability be imposed upon
the council … . After all, the council leads across densely
populated land a large volume of water under pressure, and
then exercises exclusive control over it. Whatever its
contrasted social utility, this is the equivalent of walking
one’s tiger across the forum.

Consider what the Judge means in this extract with regard to (a) the
order of determining wrongfulness and negligence, and (b) the grounds
for accepting wrongfulness.

In omission cases that involve an alleged breach of a legal duty to act positively to prevent harm to
others, the reasoning of courts often involves one or more of the following considerations:
• Prior conduct: Did the defendant’s prior conduct create a risk of harm and therefore also the
duty to prevent such harm? Prior conduct that creates a risk of harm or a new source of danger
is no longer a requirement for liability based on an omission, as shown above. However, such
prior conduct is still an important factor that indicates whether a duty to act positively to
prevent harm exists.6
• Control: Did the defendant have control over a dangerous or potentially dangerous object?
Control may be actually exercised, or the right, obligation or ability to exercise control may
flow from ownership or possession rights, or from an agreement or statutory provision. For
example, the fact that a provincial administration exercises control and supervision over all
public roads in the province, in terms of a statutory provision, is an important factor in
determining whether the administration had a legal duty to prevent a fire spreading that had
started on the side of a minor public road.7 The actual exercise of control, or the right,
obligation or ability to exercise control, is not in itself conclusive. It is, however, an important
factor that indicates whether a legal duty in respect of a particular harm occurring exists.8 One
must determine the degree and scope of the control that is required based on all the facts of the
particular case, and the harm that occurred. The essential question is whether one could
reasonably and practicably expect the person or body in control to take precautionary measures
in respect of the harm that occurred.9 The owner, occupier or person in control of property,
generally, has a duty to control fire on the property and to prevent injury to persons,10 even
trespassers on the property. The same applies to the owner or person in control of a firearm11 or
an animal.12 Both the owner of a farm and a business entity that manages a private nature
reserve on the farm are in control of the property and have a legal duty to take reasonable steps
to avoid harm to members of the public who are allowed, for a fee, to use four-wheel drive
routes that lead to dangerous areas on the property.13 Teachers in charge of schoolchildren on
an excursion have a duty to ensure that bunk-beds in a bungalow with a cement floor are safe
for young children to sleep in.14
• State departments, public bodies and officials that exercise functions in the public interest: The
South African Hang and Paragliding Association and the South African Civil Aviation
Authority exercise direction and control over paragliding in South Africa and are obliged to
ensure and promote the safety of paragliding, but it would not be reasonable to impose liability
upon them for an omission which had no direct impact on aviation safety.15 The legal
 convictions of the community demand that hospitals and health-care practitioners provide
proficient health-care services to members of the public and be held liable for failure to
provide prompt and appropriate medical treatment to a person who has suffered a spinal injury
in a rugby match.16
• Obligation to act positively in terms of common law or statute law: Did the defendant have an
obligation, in terms of a rule of common law or statute law, to act positively to prevent harm to
others? For example, the owner of low-lying land is obliged to provide lateral support for the
higher land of his or her neighbour.17 A landowner is generally required to prevent harmful
substances from flowing or otherwise escaping from the property to neighbouring land.18 The
existence of a statutory duty is determined according to the normal rules of statutory
interpretation as it appears from the wording of the statutory provision.19 For instance, a
statutory duty to provide correct information could indicate that the failure to provide
information is wrongful for the purposes of the law of delict.20
• Special relationship: Did a special relationship between the parties give rise to a legal duty for
the defendant to prevent harm to the plaintiff? The mere fact of such a special relationship is
not conclusive, but it will be an important factor in indicating a legal duty to prevent harm, and
courts will consider it along with all the other circumstances of the case. A relationship of trust
or authority, or a long-standing business relationship can give rise to such a duty.21 A
contractual relationship can also give rise to a duty to prevent harm to the other party, to the
contract, or to a third party.22 Other examples include the relationship between a policeman and
a member of the public,23 a prison officer and a prisoner,24 and an employer and his or her
employee.25
• Creating an expectation: Did the defendant create the expectation that he or she would protect
the interests of the plaintiff? For example, a security firm that provides security services for an
office building or parking garage, in terms of a contract with the owner, may have a legal duty
toward third parties who rely on the security services for protection of their person or
property.26 Creating such an expectation is not conclusive, and courts will take into account
whether the plaintiff could reasonably have relied on the security services for protection, and
also all the other circumstances of the case.27
• Knowledge: Did the defendant know or foresee that the omission would cause harm, or did the
defendant have the motive to cause harm? Knowledge or foresight of the possibility of harm
makes causing such harm unreasonable and therefore wrongful.28
• Practical measures to avert harm: What practical measures could be taken to avert the harm?
Courts consider the probable success and the relative ease and expense of practical steps that
the defendant could have taken to avert the loss. Courts also take into account the plaintiff’s
ability or lack of ability to protect himself or herself against liability, as in the case of Indac
Electronics (Pty) Ltd v Volkskas Bank Ltd,29 where the liability of a collecting bank towards the
true owner of a cheque was in issue. The Court considered the ability of the bank to protect
itself against liability by obtaining insurance cover.
• Professional duty: Did the failure to prevent harm occur while rendering professional services
(for example as an attorney, engineer, or a doctor), and was there a failure of professional
competence or skill? 30
• Public office: Did a person holding a public office, such as a notary, sworn appraiser or an
auditor, fail to prevent harm in the course of his or her official duties? 31
• Social and economic implications: What are the legal, social and economic implications of
imposing liability for the infringement? 32 For example, would imposing liability on a
policeman for dereliction of duty result in a large influx of similar claims, so that substantial
time, expense and diversion of police manpower would be required to defend such claims? 33
These policy considerations give specific content to the general criterion of reasonableness for
determining wrongfulness, providing guidelines and flexibility for further judicial development of
the law on liability for omissions.

1 1975 (3) SA 590 (A).

2 1975 (3) SA 590 (A).
3 2000 (3) SA 1049 (SCA).
4 2001 (1) SA 105 (SCA).
5 2001 (1) SA 105 (SCA) para 43.
6 Minister van Polisie v Ewels 1975 (3) SA 590 (A) at 597; Administrateur, Transvaal v Van der Merwe 1994 (4) SA
347 (A) at 349; Local Transitional Council of Delmas v Boshoff 2005 (5) SA 514 (SCA); Cape Town Municipality v
Bakkerud 2000 (3) SA 1049 (SCA) paras 25–26.
7 Administrateur, Transvaal v Van der Merwe 1994 (4) SA 347 (A).
8 Administrateur, Transvaal v Van der Merwe 1994 (4) SA 347 (A) at 359 and 360; Local Transitional Council of
Delmas v Boshoff 2005 (5) SA 514 (SCA).
9 Administrateur, Transvaal v Van der Merwe 1994 (4) SA 347 (A) at 359–360.
10 See Minister of Forestry v Quathlamba (Pty) Ltd 1973 (3) SA 69 (A); Steenberg v De Kaap Timber (Pty) Ltd 1992 (2)
SA 169 (A); Dews v Simon’s Town Municipality 1991 (4) SA 479 (C) at 485; Gouda Boerdery BK v Transnet Ltd
2005 (5) SA 490 (SCA) para 13; Minister of Water Affairs and Forestry v Durr 2006 (6) SA 587 (SCA) paras 12–13
and 17; Potgieter v University of Stellenbosch [2017] 1 All SA 282 (WCC).
11 Maylett v Du Toit 1989 (1) SA 90 (T).
12 Kruger v Coetzee 1966 (2) SA 428 (A); S v Fernandez 1966 (2) SA 259 (A); R v Eustace (2) 1948 (3) SA 859 (T);
Bristow v Lycett 1971 (4) SA 223 (RA); Zietsman v Van Tonder 1989 (2) SA 484 (T).
13 Za v Smith 2015 (4) SA 574 (SCA).
14 Hawekwa Youth Camp v Byrne [2010] 2 All SA 312 (SCA)
15 South African Hang and Paragliding Association v Bewick 2015 (3) SA 449 (SCA).
16 Oppelt v Department of Health, Western Cape 2016 (1) SA 325 (CC).
17 D and D Deliveries (Pty) Ltd v Pinetown Borough 1991 (3) SA 250 (D) at 253; John Newmark and Co (Pty) Ltd v
Durban City Council 1959 (1) SA 169 (N); Gijzen v Verrinder 1965 (1) SA 806 (D).
18 Regal v African Superslate (Pty) Ltd 1963 (1) SA 102 (A) at 109.
19 Minister van Polisie v Ewels 1975 (3) SA 590 (A) at 595–596; Administrateur, Transvaal v Van der Merwe 1994 (4)
SA 347 (A); Knop v Johannesburg City Council 1995 (2) SA 1 (A); Minister of Law and Order v Kadir 1995 (1) SA
303 (A) at 319.
20 Herschel v Mrupe 1954 (3) SA 464 (A) at 490; Da Silva v Coutinho 1971 (3) SA 123 (A) at 140; International
Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 694.
21 Jowell v Bramwell-Jones 1998 (1) SA 836 (W) (trustees); Arthur E Abrahams and Gross v Cohen 1991 (2) SA 301
(C) at 311 (professional administrators of a deceased estate); Bowley Steels (Pty) Ltd v Dalian Engineering (Pty) Ltd
1996 (2) SA 393 (T) (parties to a long-standing business relationship).
22 Greenfield Engineering Works (Pty) Ltd v NKR Construction (Pty) Ltd 1978 (4) SA 901 (N); Bayer South Africa (Pty)
Ltd v Frost 1991 (4) SA 559 (A); Cathkin Park Hotel v JD Makesch Architects 1993 (2) SA 98 (W) at 100; McCann v
Goodall Group Operations (Pty) Ltd 1995 (2) SA 718 (C) at 726 and 727; Joubert v Impala Platinum Ltd 1998 (1) SA
463 (BH).
23 Minister van Polisie v Ewels 1975 (3) SA 590 (A).
24 Mtati v Minister of Justice 1958 (1) SA 221 (A).
25 Joubert v Impala Platinum Ltd 1998 (1) SA 463 (BH).
26 Compass Motors Industries (Pty) Ltd v Callguard (Pty) Ltd 1990 (2) SA 520 (W).
27 Longueira v Securitas of South Africa (Pty) Ltd 1998 (4) SA 258 (W) at 261–262.
28 Van der Merwe Burger v Munisipaliteit van Warrenton 1987 (1) SA 899 (NC) at 908; Rabie v Kimberley
Munisipaliteit 1991 (4) SA 243 (NC); Langley Fox Building Partnership (Pty) Ltd v De Valence 1991 (1) SA 1 (A).
29 1992 (1) SA 783 (A) at 799.
30 Arthur E Abrahams and Gross v Cohen 1991 (2) SA 301 (C); Jowell v Bramwell-Jones 1998 (1) SA 836 (W);
Mukheiber v Raath 1999 (3) SA 1065 (SCA).
31 The Cape of Good Hope Bank v Fischer (1885–1886) 4 SC 368; Macadamia Finance Ltd v De Wet 1991 (4) SA 273
(T).
32 Shell and BP SA Petroleum Refineries (Pty) Ltd v Osborne Panama SA 1980 (3) SA 653 (D) at 659–660;
Franschhoekse Wynkelder (Ko-operatief) Bpk v South African Railways and Harbours 1981 (3) SA 36 (C);
Mpongwana v Minister of Safety and Security 1999 (2) SA 794 (C) at 802–803; Mukheiber v Raath 1999 (3) SA 1065
(SCA) para 28.
33 Minister van Polisie v Ewels 1975 (3) SA 590 (A) at 595–596; Minister of Law and Order v Kadir 1995 (1) SA 303
(A) at 319.
 Chapter 15

Negligent misstatements

15.1 Introduction

15.2 Wrongfulness

15.1 Introduction
Where the plaintiff acts upon incorrect | information supplied by the defendant and suffers harm,
liability depends on whether the plaintiff had a right to be given correct information and the
defendant a duty to supply such information. If no contractual relationship between the parties
existed, one cannot assume a right to information from the outset and must therefore proceed from
the duty side to assess wrongfulness. The focus of the enquiry is whether the factual situation gives
rise to policy considerations indicating that a legal duty to provide correct information exists.
Liability for negligent misstatements is an important category of liability for pure economic harm.
The development of the law towards recognising liability for a negligent misstatement causing
pure economic harm has been described as follows:1
It is clear that in our law Aquilian liability has long outgrown its earlier limitation to
damages arising from physical damage or personal injury. Thus, for instance, in
Administrateur, Natal v Trust Bank van Afrika Bpk 1979 (3) SA 824 (A) this Court held that
Aquilian liability could in principle arise from negligent misstatements which caused pure
financial loss, i.e. loss which was caused without the interposition of a physical lesion or
injury to a person or corporeal property.

Courts follow a cautious approach to liability for statements, as reflected in the following extract
from the judgment of the House of Lords in the well-known English case of Hedley Byrne & Co
Ltd v Heller & Partners Ltd: 2
Words are more volatile than deeds. They travel fast and far afield. They are used without
being expended and take effect in combination with innumerable facts and other words.
 Administrateur, Natal v Trust Bank van Afrika Bpk3
The Court applied the general criterion of reasonableness to determine whether a
misstatement causing harm is wrongful for the purposes of delictual liability. The Court
accepted that the criterion of reasonableness involves policy considerations and Rumpff CJ
quoted the following passage from Fleming’s Law of Torts as being correct also for South
African law:

In short, recognition of a duty of care is the outcome of a value judgment that

the plaintiff’s invaded interest is deemed worthy of legal protection against
negligent interference by conduct of the kind alleged against the defendant. In
the decision whether or not there is a duty, many factors interplay: the hand of
history, our ideas of morals and justice, the convenience of administering the
rule and our social ideas as to where the loss should fall. Hence, the incidence
and extent of duties are liable to adjustment in the light of the constant shifts
and changes in community attitudes.

In the Administrateur, Natal case, the defendant bank had acted on behalf of a person who
claimed compensation from the provincial authorities for expropriation of property. The
authorities eventually paid him compensation via the bank, but it turned out that he was not
the owner of the property concerned. The provincial authorities then claimed the amount that
they paid out as damages for alleged negligent misrepresentation by the bank. The action
failed because the provincial authorities themselves had initially identified the claimant as the
owner of the property. The Court held that the bank had no legal duty to verify the facts, and
that the plaintiff’s own mistake, therefore, caused the loss. However, liability for negligent
misrepresentation in instances where a duty to provide correct information existed, is now
well established.

15.2 Wrongfulness
Wrongfulness is often the main issue in determining liability for negligent misstatements, because
causing pure economic harm by negligent misstatement, is not prima facie wrongful.4 Although one
generally determines wrongfulness by looking at either the infringement of a right or the breach of
a duty, in these instances there is often no infringement of one of the settled categories of rights
(real, personal, personality, or intellectual property rights). Courts, therefore, ask whether the
defendant had a legal duty to provide correct information to the plaintiff, and whether fulfilling this
duty would have prevented harm to the plaintiff.5
The following factors are typical of what courts take into account when deciding whether a
legal duty to provide correct information to another person exists:
• Public office: Was the economic loss caused by a person holding a public office, such as a
notary, sworn appraiser or an auditor? Such a person has ‘a kind of patent of credibility and
efficiency conferred upon him or her by public authority’6 and members of the public are
‘invited and entitled to repose confidence and trust in the acts of such persons performed in
their respective capacities’.7
• Professional knowledge and competence: Was the economic loss caused while providing
professional services, and was there a failure of professional competence or skill? Where the
defendant provides professional services and professes to possess special skills, special or
exclusive knowledge, or professional competence, courts will more readily accept that it is
unreasonable to cause economic loss to a person that depends on the defendant’s professional
competence, or that relies on the correctness of information furnished in a professional
capacity.
EG Electric Co (Pty) Ltd v Franklin8
A registered electrician, on instructions of the seller of a house, had supplied a certificate that
the electrical wiring of the house complied with municipal regulations. The Court held that a
registered electrician owed a legal duty to provide a certificate with correct information to the
purchaser of the house, who had relied on the correctness of the certificate and later had to
incur costs to rectify defective wiring.

Mukheiber v Raath9
The parents of a healthy and normal child (their fourth) instituted action in delict against a
gynaecologist, alleging that he had negligently misrepresented to them that the wife had been
sterilised after the birth of their third child. Relying on this representation, they had failed to
take contraceptive measures, with the result that the fourth child was conceived and born.
The parents claimed damages from the doctor for pure economic loss, in the form of
confinement costs and maintenance of the child until he becomes self-supporting. The
Supreme Court of Appeal held that the doctor had a legal duty to stop making any
representation on the matter of sterilisation until he had taken reasonable steps to ensure the
accuracy of his representation. The factual and policy considerations indicating that such a
duty existed were the following:
• The special relationship between the doctor and the parents who consulted him
• The material risk that the situation involved, that is, the risk of the conception and birth of an
unwanted child
• The fact that this risk should have been obvious to the doctor
• The fact that it should also have been obvious to the doctor that the parents would rely on
what he told them, that the correctness of the representation was of vital importance to them,
and that they could suffer serious damage if the representation was incorrect
• The fact that the representation related to technical matters concerning a surgical procedure
about which the parents would necessarily be ignorant and the doctor should be
knowledgeable.

As far as public policy considerations are concerned, the Court held that the parents’ reasons
for wanting the sterilisation were socio-economic and family reasons, and that these reasons
were socially acceptable and not contra bonos mores. Recognising legal duty in this case
would not impose too heavy a burden on the doctor. Professional people must not act
negligently and should not make unsolicited misrepresentations. Through a misstatement, the
doctor had wrongfully caused financial loss to the parents.

Axiam Holdings Ltd v Deloitte & Touche10

The plaintiffs sued auditors for misstatement of an audited company’s financial position,
which caused the plaintiffs loss after they had purchased shares in the company. The
auditors excepted to the claim on the grounds (1) that they did not owe the plaintiffs a duty in
law and (2) that their failure to warn the plaintiffs did not constitute representation in terms of
section 20(9)(b)(ii) of the Public Accountants’ and Auditors’ Act 80 of 1991. Regarding (1), the
Court declined to decide at exception stage that it was inconceivable that the auditors who
knew of the misstatement and who knew that purchasers of shares in the audited company
would rely on the correctness of their statements, would not have a duty to speak. As to (2),
the Court found that actionable misrepresentation by silence or inaction is possible where a
duty to speak or act exists. Wrongfulness must be determined in light of the nature, context
and purpose of the statement and the relationship between parties. A court may conclude at
the trial that a reasonable person would not have kept silent, but would have expressed
reservations as to the reliability of the financial information. It was, therefore, not possible to
decide on exception (that is, without evidence) that the alleged misstatement by omission
was not wrongful.

Cape Empowerment Trust Limited v Fisher Hoffman Sithole11

The purchaser of a business relied on a certificate issued by the seller’s auditor, confirming
that the business had made a profit of R10 million. It turned out that this was entirely untrue
 and that the auditor had been grossly negligent. On the issue whether the auditor owed a
legal duty to the purchaser (who was not his client) and had wrongfully caused the
purchaser’s economic loss, the Court decided that it was impermissible to take into account
the auditor’s gross negligence as a policy consideration indicating a legal duty and
wrongfulness. This would telescope the tests for wrongfulness and negligence into one. In the
context of negligent misstatements the element of wrongfulness can exclude liability, despite
the presence of all other elements of liability, including gross negligence. In this case, the
Court took into account the following factors to determine wrongfulness:
(1) Was the representation made in a business context and in response to a serious request? (yes)
(2) Was the plaintiff dependent on the defendant to provide the information or advice? (no)
(3) Was the correct information available to the plaintiff from another source? (yes, the purchaser could
have undertaken a comprehensive due diligence investigation of the business, as provided for in the
contract of sale; and the terse, unmotivated profit statement by the defendant auditor could not be
regarded as a substitute for the protection that a comprehensive due diligence investigation by the
plaintiff’s own auditors would provide)
(4) What was the extent of the plaintiff’s ‘vulnerability to risk’ – could the plaintiff reasonably have
avoided the risk of harm by other means? (yes, the plaintiff covered itself against the risk that the
business may not have attained the profit represented by the sellers, by procuring an express
warranty from the sellers, which corresponded exactly with the representation by the seller’s auditor
– this profit warranty effectively shielded the purchaser from any adverse consequence of a
misstatement of profit by anybody, including the seller’s auditor; and yet the plaintiff, through its own
conduct in allowing the agreement to lapse, had deprived itself of this contractual remedy).

The Court weighed up these factors and decided not to impose liability on the auditor for the
grossly negligent misstatement.

• Knowledge: Did the defendant know or foresee that the misstatement would cause harm, or did
the defendant have the motive to cause harm? Such knowledge or foresight of the possibility of
harm imposes a duty on the defendant not to cause the harm. Causing such harm is
unreasonable and therefore wrongful.12
• Extent of possible liability and the economic or social consequences of imposing liability:
Where recognising a duty to prevent economic loss could lead to a situation of indeterminate
liability or ‘one fraught with an overwhelming potential liability’ or to a ‘multiplicity of
actions’ that could be ‘socially calamitous’, courts will be reluctant to accept that such a duty
rested on the defendant.13
• Ability to protect oneself against liability or loss: Courts can take into account the ability of the
person who suffered the loss to take protective measures against such loss, for example, by
verifying the information received. Courts also take into account the ability of the defendant to
protect himself or herself against liability for such loss, for example, by obtaining a contractual
warranty14 or insurance cover.15
• A special relationship: Courts will be inclined to accept that a relationship of trust or
dependence, or a fiduciary relationship, gives rise to a legal duty to provide correct
information, as between an employer and employee,16 and between a bank and its client.17
• Pre-contractual negotiations: A person in pre-contractual negotiations with another has a legal
duty not to make a misstatement about a material aspect of the contract. The breach of this
duty can lead to delictual liability for the economic loss caused by a negligent misstatement.18
• Statutory duty: Was a duty to provide correct information provided for or implied by a
statutory provision?19

These factors give specific content to the flexible criterion of general reasonableness (boni mores).
They are the basis on which courts determine whether the plaintiff had a right to be given correct
information, and whether the defendant had a duty to supply such information. If misstatement by
the defendant constitutes a breach of such a duty and causes harm to the plaintiff, the requirement
of wrongfulness is met. Courts will impose liability if the defendant acted negligently.

1 By Grosskopf in Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A) at 498.
2 1964 AC 465 (HL) at 534.
3 1979 (3) SA 824 (A) at 833–834.
4 Telematrix (Pty) Ltd t/a Matrix Vehicle Tracking v Advertising Standards Authority SA 2006 (1) SA 461 (SCA) para
13; Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 12.
5 Knop v Johannesburg City Council 1995 (2) SA 1 (A) at 27; Administrateur, Natal v Trust Bank van Afrika Bpk 1979
(3) SA 824 (A) at 832H–833A.
6 Herschel v Mrupe 1954 (3) SA 464 (A) at 488.
7 Herschel v Mrupe 1954 (3) SA 464 (A) at 488.
8 1979 (2) SA 702 (E).
9 1999 (3) SA 1065 (SCA).
10 2006 (1) SA 237 (SCA).
11 2013 (5) SA 183 (SCA) paras 24–25.
12 International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 694; Bayer South Africa (Pty) Ltd v Frost 1991
(4) SA 559 (A) at 575; Mukheiber v Raath 1999 (3) SA 1065 (SCA) at 1076.
13 Mukheiber v Raath 1999 (3) SA 1065 (SCA) para 51.
14 Cape Empowerment Trust Ltd v Fisher Hoffman Sithole 2013 (5) SA 183 (SCA) paras 29–30.
15 Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799.
16 Mukheiber v Raath 1999 (3) SA 1065 (SCA).
17 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
18 Bayer South Africa (Pty) Ltd v Frost 1991 (4) SA 559 (A) at 569C–D.
19 International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 694.
 Chapter 16

Pure economic harm

16.1 What is pure economic harm or loss, and why does it require special
attention in the law of delict?

16.2 Pure economic harm – examples

16.3 Wrongfulness

16.1 What is pure economic harm or loss, and why does it |

require special attention in the law of delict?
Pure economic harm means financial loss that is not connected to any physical injury or damage to
property of the plaintiff. It requires special attention in the law of delict mainly because it poses a
limitation problem. The economic effects of harmful conduct can be much more widespread than
the physical effects. Economic effects are not subject to the natural laws of physics and can spread
widely and unpredictably, for example, where people react to incorrect information in a news
report, or where the malfunction of an electricity network causes shut-downs, expenses and loss of
profits to businesses that depend on electricity.
Courts try not to impose indeterminate liability for causing pure economic harm (‘opening the
floodgates of liability’). Instead, they seek to contain liability within reasonably predictable limits,
so as not to stifle initiative and enterprise. They do this mainly by means of the criterion of general
reasonableness used to determine wrongfulness. However, in some cases, they apply the criterion
of reasonable foreseeability used to determine negligence, and also use a flexible standard for
determining legal causation that involves considering reasonableness, directness and fairness.

16.2 Pure economic harm – examples

Situations where courts impose liability for causing pure economic harm include the following:
• A negligent misstatement that causes the plaintiff to incur additional expenses
• Where the relationship between the parties, such as a professional or employment relationship,
involves a duty to look after the financial affairs of someone else, for example in the following
cases:
◆ A financial adviser negligently invests a client’s money in high-risk ventures1
◆ An attorney negligently fails to ensure that a will is validly executed, so that the named
beneficiaries derive no benefit from the will 2
◆ An employer negligently fails to advise its employee on a group life insurance scheme,
so that the employee’s dependants suffer loss upon his death 3
◆ An employer negligently fails to process an injury compensation claim on behalf of an
employee, so that the employee forfeits the benefit 4
◆ A bank furnishes incorrect information to its client.5
• Where someone else’s physical injury or property damage has an adverse economic effect for
the plaintiff, for example, where negligently causing the death of a breadwinner results in the
plaintiff’s loss of support,6 or where negligently causing physical damage to a facility owned
by someone else, and used by the plaintiff, results in the plaintiff having to incur additional
expenses.7

A person can suffer economic harm as a derivative of physical harm, for example, when injury
causes loss of earnings,8 or when damage to a vehicle makes it necessary to incur the expense of
hiring a substitute vehicle.9 In such cases, the primary focus is on the bodily injury or the property
damage, which courts regard as prima facie wrongful. Courts take into account the resulting
economic harm as an additional item of recoverable damages.

16.3 Wrongfulness
Causing pure economic harm is not prima facie wrongful.10 In Telematrix (Pty) Ltd t/a Matrix
Vehicle Tracking v Advertising Standards Authority SA11 Harms JA said the following in this
regard:12
When dealing with the negligent causation of pure economic loss it is well to remember that
the act or omission is not prima facie wrongful (‘unlawful’ is the synonym and is less of a
euphemism) and that more is needed. Policy considerations must dictate that the plaintiff
should be entitled to be recompensed by the defendant for the loss suffered (and not the
converse as Goldstone J once implied unless it is a case of prima facie wrongfulness, such as
where the loss was due to damage caused to the person or property of the plaintiff). In other
words, conduct is wrongful if public policy considerations demand that in the circumstances
the plaintiff has to be compensated for the loss caused by the negligent act or omission of the
defendant. It is then that it can be said that the legal convictions of society regard the
conduct as wrongful … .

One generally determines wrongfulness by looking at the infringement of a right or the breach of a
duty. In cases of pure economic harm, there is often no infringement of one of the settled categories
of rights (real, personal, personality, or intellectual property rights). Therefore, courts ask whether
the defendant had a legal duty to prevent economic harm to the plaintiff.13 The following are typical
factors that courts take into account when deciding whether a legal duty to prevent pure economic
harm to another person exists:
• Knowledge: The fact that the defendant knew or subjectively foresaw that his or her conduct
would cause damage to the plaintiff is an important and often decisive factor.14 Such
knowledge or foresight could arise from the mere fact of a contractual relationship between the
 parties, or from the fact that one party of necessity relies on the conduct, statement or
information of the other, for example, where incorrect crop insurance information is provided
by an insurance broker to a specific group of farmers.15 Knowledge that lawful steps to
liquidate a company will cause harm to its shareholders and to other shareholders a few
shareholding levels removed from the company does not establish a duty to prevent economic
harm.16
• Practical measures to avert the harm: Courts consider the probable success, relative ease and
expense of practical steps that the defendant could have taken to avert the harm.17 •
Professional knowledge and competence: Where the defendant provides professional services
and professes to possess special skills, knowledge and competence, courts will more readily
accept that he or she had a duty not to cause financial loss to others while providing
professional services.18
• Degree and extent of risk: A high degree of risk indicates that the defendant had the duty to
take preventative steps.19
• Ability to protect oneself against liability or loss: Courts take into account the ability of the
person who suffered the loss to take protective measures against such loss. They also consider
the ability of the defendant to protect himself or herself against liability for such loss by, for
example, obtaining a contractual warranty 20 or insurance cover.21
• A special relationship: Courts will be inclined to accept that a relationship of trust or
dependence, or a fiduciary relationship, gives rise to a legal duty to prevent economic loss to
others, for example, as between an employer and employee,22 between a bank and its client 23
and between an attorney and the depositor of money into his or her trust account.24
Franschhoekse Wynkelder (Ko-operatief) Bpk v South African Railways and Harbours25
The plaintiff, a wine-making cooperative, instituted an action for damages against the
defendant, whose employees has sprayed weedkiller on the undergrowth alongside one of its
railway lines. The weedkiller had contaminated the soil in the vineyards of farms next to the
railway line. The owners of these farms were members of the cooperative, and in terms of its
constitution, the members were obliged to deliver grapes to the plaintiff to make wine. The
plaintiff alleged that vines growing on the contaminated soil were destroyed or damaged and
that, as a result, the plaintiff would not receive grapes from these farms, causing the plaintiff
harm, which the plaintiff alleged was foreseeable. The defendant excepted to the claim on the
ground, inter alia, that it was not in law liable to the plaintiff for any harm, because the
defendant’s conduct was not wrongful vis-à-vis the plaintiff. The Court held that the plaintiff
had not alleged that there was any special relationship between the parties, and that no
circumstances or facts alleged by the plaintiff suggested that the defendant had a legal duty
to prevent harm to the plaintiff. There were also no considerations of public policy that
justified the recognition of such a legal duty.

• Statutory duty: Was a duty to prevent economic loss provided for or implied by a statutory
provision? Courts will determine whether such a statutory duty exists by gauging the intention
of the legislature as it appears from the wording of the statutory provision.
Knop v Johannesburg City Council 26
The Court held that a local authority charged with implementing zoning provisions did not
have a duty to prevent economic loss to a person who incurred wasted costs when his
application for permission to subdivide property was granted, but it then later appeared that
the permission was in contravention of an existing zoning plan.

Minister of Law and Order v Kadir 27

Policemen had investigated the scene of an accident that was caused by a package falling off
a delivery vehicle. The driver of the vehicle had driven on without stopping. The policemen
 failed to obtain the names and addresses of possible witnesses before these people left the
scene of the accident. The Court held that the policemen did not have a legal duty towards
the victim of the accident, who was later unable to institute a civil claim for damages against
the possibly negligent and unknown driver of the vehicle. The police had a statutory duty in
terms of section 5 of the Police Act 7 of 1958 to deter crimes, track down criminals and
protect the public against crimes. However, Hefer JA stated:28

Viewing the matter objectively, society will take account of the fact that the
functions of the police relate in terms of the Act to criminal matters and were not
designed for the purpose of assisting civil litigants.

A statutory duty to provide correct information could indicate that the failure to
provide information or providing incorrect information is unlawful.

• Fraud or dishonesty: Where one person furthers his or her own interests at the cost of harm to
another, the existence of fraud or dishonesty makes causing the harm more unreasonable. In
Minister of Finance v Gore NO 29 the Court said:
In the language of the more recent formulations of the criterion for
wrongfulness: in cases of pure economic loss the question will always be
whether considerations of public or legal policy dictate that delictual liability
should be extended to loss resulting from the conduct at issue. Thus
understood, it is hard to think of any reason why the fact, that the loss was
caused by dishonest (as opposed to bona fide negligent) conduct, should be
ignored in deciding the question. We do not say that dishonest conduct will
always be wrongful for the purposes of imposing liability, but it is difficult to
think of an example where it will not be so.

• Policy considerations: In pure economic harm cases, the plaintiff must allege in the pleadings
not only that the harm was negligently and wrongfully caused, but must also allege and prove
the facts relied upon to substantiate the considerations of policy that give rise to a legal duty on
the part of the defendant.30
◆ The first policy consideration is the law’s concern to avoid the imposition of liability in
an indeterminate amount for an indeterminate time to an indeterminate class.31
◆ Courts will consider whether recognising liability would impose an additional burden on
the defendant, which would be unwarranted or which would constitute an unjustified
limitation of the defendant’s activities.32
◆ Courts will more readily impose liability for a single loss, or loss that affects a single
identifiable plaintiff, or that occurs once, or that is unlikely to cause a large influx of
actions.33
◆ Extent of possible liability and economic or social consequences of imposing liability:
Where recognising a duty to prevent economic loss could lead to a situation of
indeterminate liability or ‘one fraught with an overwhelming potential liability’, or to a
‘multiplicity of actions’ that could be ‘socially calamitous’, courts will be reluctant to
accept that such a duty rested on the defendant.34
◆ Courts have not extended liability for negligent interference with a personal right or
personal claim to cases where the plaintiff suffers harm that results from death or injury of
another person with whom the plaintiff had a contractual relationship. In Union
Government v Ocean Accident & Guarantee Corporation Ltd 35 the Court refused the
government’s claim for loss suffered as a result of negligently inflicted injury to a
government employee (a magistrate), on the basis that recognising such a claim would
lead to a large increase in claims arising from relationships, contractual or other, between
 the physically injured person and other persons who may indirectly suffer economic harm
as a result of the injury.

Shell and BP SA Petroleum Refineries (Pty) Ltd v Osborne Panama SA36

The defendant’s employee caused damage to a mooring buoy in Durban harbour, which in
turn caused economic harm to the charterer of an oil tanker waiting outside the harbour,
because of the consequent delay in discharging the cargo and additional liability for charter
fees. The Court held that the defendant did not owe a legal duty to the plaintiff to prevent
such economic harm, because the plaintiff had not shown that the defendant should have
foreseen the possibility of such harm to the charterer specifically. The Court considered this
type of harm to be foreseeable only in relation to an undetermined class of potential victims,
namely the owners or charterers of all vessels intending to discharge at that particular
mooring buoy. Only insofar as the plaintiffcharterer was a member of that undetermined class
was his loss reasonably foreseeable. The Court was therefore unwilling to recognise a legal
duty on the part of the defendant towards one of an unknown number of potential claimants.
• In contractual relationships between the parties:
◆ Courts will enquire whether the plaintiff has a contractual remedy and whether there is a
need for a delictual remedy
◆ Courts will enquire whether the plaintiff could have been protected by contractual means,
for example, by prior agreement that the defendant will bear the risk of harm.37

Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 38

A big truck carrying asbestos dust was on its way to Durban harbour from a mine in
Zimbabwe. The driver negligently caused a collision on a toll road. The truck overturned and
spilled its cargo on and around the road. Because of the hazardous nature of asbestos
powder, the spillage required an extensive cleaning-up and decontamination operation. As a
result, the toll road had to be closed for 24 hours, and the Roads Agency lost the revenue
that would otherwise have been collected at two nearby toll plazas. The Roads Agency
instituted an action in delict against the driver and his employer (Fourway Haulage), for
damage suffered in the form of lost toll revenue for 24 hours.
Consider the following questions and the manner they were dealt with in this case:
• Did this case involve pure economic harm?
• Did this case raise the possibility of boundless liability, involving ‘the law’s concern to avoid
the imposition of liability in an indeterminate amount for an indeterminate time to an
indeterminate class’?
• Did this case involve a single loss that affected a single identifiable plaintiff, and that
occurred once (a loss unlikely to bring a large influx of actions)?
• Could the plaintiff have protected itself by contractual means, for example, by prior
agreement that the defendant will bear the risk of harm? Was the Roads Agency ‘vulnerable’
to the risk of the loss, because it could not readily protect itself against that risk by
concluding a contract with every user of the toll road?
• Would imposing liability unreasonably interfere with the defendant’s commercial freedom, or,
conversely, was there no such unreasonable interference, because the defendant was
already under a duty to the world at large to take reasonable care in transporting the cargo?
• Would it have been more appropriate to regard this as a case where the risk of loss could
have been spread among all the road users by an increase in toll fees, instead of burdening
the defendant with the Road Agency’s loss?
• Was it relevant to the question of wrongfulness that the defendant’s employee was
transporting an extremely dangerous cargo? Would the situation have been different if the
truck had contained an innocuous substance like sand?

Country Cloud Trading CC v MEC, Department of Infrastructure Development39

The Department of Infrastructure Development entered into a building contract with iLima.
When the project ran into difficulties iLima borrowed R12 million from Country Cloud. The
 loan agreement was subject to the condition that the Department would repay Country
Cloud’s R12 million out of the amount payable by the Department to iLima in terms of the
building contract. The Department later cancelled the building contract, thereby committing
breach of contract, and this resulted in iLima’s liquidation and a loss to Country Cloud, who
then sued the Department in delict. The central issue was whether the Department had
wrongfully caused harm to Country Cloud. Country Cloud’s claim was for pure economic
harm and it could not show that the Department had wrongfully infringed its rights or had a
legal duty not to cause it economic harm. Country Cloud relied on state accountability, but
this consideration does not always give rise to a private-law duty, particularly if, as in this
case, the Department did not act dishonestly or corruptly. Also, Country Cloud could have
attempted to take steps to protect itself against non-payment by iLima, by claiming repayment
from iLima’s liquidator under the loan agreement, or taking cession of iLima’s claim for breach
of contract against the Department, or by calling for payment from a surety.

1 Pinshaw v Nexus Securities (Pty) Ltd 2002 (2) SA 510 (C).

2 Pretorius v McCallum 2002 (2) SA 423 (C).
3 Aucamp v University of Stellenbosch 2002 (4) SA 544 (C).
4 Joubert v Impala Platinum Ltd 1998 (1) SA 463 (BH).
5 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
6 Evins v Shield Insurance Co Ltd 1980 (2) SA 814 (A) at 837–838.
7 Shell and BP SA Petroleum Refineries (Pty) Ltd v Osborne Panama SA 1980 (3) SA 653 (D). In Viv’s Tippers (Edms)
Bpk v Pha Phama Staff Services (Edms) Bpk t/a Pha Phama Security 2010 (4) SA 455 (SCA); [2011] 1 All SA 34
(SCA) the Supreme Court of Appeal regarded the claim for loss suffered by the owner of a vehicle stolen from the
premises protected by a security company as a claim for pure economic loss. Arguably, this is rather a case of physical
loss of property.
8 SA Eagle Insurance Co Ltd v Hartley 1990 (4) SA 833 (A) at 836; Ngubane v South African Transport Services 1991
(1) SA 756 (A) at 781.
9 Smit v Abrahams 1994 (4) SA 1 (A).
10 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 12; Country Cloud
Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 22.
11 SA 2006 (1) SA 461 (SCA).
12 Para 13 (footnotes omitted).
13 Knop v Johannesburg City Council 1995 (2) SA 1 (A) at 27; Administrateur, Natal v Trust Bank van Afrika Bpk 1979
(3) SA 824 (A) at 832H–833A.
14 Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799; International Shipping Co (Pty) Ltd v
Bentley 1990 (1) SA 680 (A) at 694; Bayer South Africa (Pty) Ltd v Frost 1991 (4) SA 559 (A) at 575; Mukheiber v
Raath 1999 (3) SA 1065 (SCA) at 1076.
15 Delphisure Group Insurance Brokers Cape (Pty) Ltd v Kotzé [2011] 1 All SA 109 (SCA) paras 23–26.
16 Itzikowitz v Absa Bank Ltd 2016 (4) SA 432 (SCA).
17 Coronation Brick (Pty) Ltd v Strachan Construction Co (Pty) Ltd 1982 (4) SA 371 (D) at 384; Arthur E Abrahams
and Gross v Cohen 1991 (2) SA 301 (C) at 312.
18 Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799; Mukheiber v Raath 1999 (3) SA 1065
(SCA); Viv’s Tippers (Edms) Bpk v Pha Phama Staff Services (Edms) Bpk t/a Pha Phama Security 2010 (4) SA 455
(SCA); [2011] 1 All SA 34 (SCA) paras 8 and 21; Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
19 Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799.
20 Cape Empowerment Trust Ltd v Fisher Hoffman Sithole 2013 (5) SA 183 (SCA) paras 29–30.
21 Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799.
22 Mukheiber v Raath 1999 (3) SA 1065 (SCA).
23 Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
24 Hirschowitz Flionis v Bartlett and Another 2006 (3) SA 575 (SCA).
25 1981 (3) SA 36 (C).
26 1995 (2) SA 1 (A).
27 1995 (1) SA 303 (A) at 319.
28 Para 321H.
29 2007 (1) SA 111 (SCA) para 87.
30 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA).
31 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 23.
32 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 26.
33 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 24.
34 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 23; Coronation Brick
(Pty) Ltd v Strachan Construction Co (Pty) Ltd 1982 (4) SA 371 (D) at 383–384 and 386–387; Arthur E Abrahams
and Gross v Cohen 1991 (2) SA 301 (C) at 312; Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A)
at 798.
35 1956 (1) SA 577 (A) at 585B–D.
36 1980 (3) SA 653 (D).
37 Fourway Haulage SA (Pty) Ltd v SA National Roads Agency Ltd 2009 (2) SA 150 (SCA) para 25.
38 2009 (2) SA 150 (SCA).
39 2015 (1) SA 1 (CC).
 Chapter 17

Interference with contractual relations

17.1 Introduction

17.2 Negligent interference with contractual relations

17.3 Intentional interference with contractual relations

17.4 Conclusion

17.1 Introduction
The interference with contractual relations amounts to the causation of pure economic loss and
therefore the plaintiff is required to institute the Aquilian action to recover the patrimonial harm
which he or she has suffered from the wrongdoer.1 To be successful, the plaintiff must prove all of
the elements of delictual liability and, because this is an instance of pure economic loss, the
plaintiff will generally have to establish that the loss was caused wrongfully.
Courts have approached the interference with contractual relations differently, depending on
whether it occurred negligently or intentionally.

17.2 Negligent interference with contractual relations

Generally speaking, courts have been consistent in their refusal to impose delictual liability for the
causation of pure economic loss through negligently interfering with another person’s contractual
rights.2 This approach may be traced back to Union Government v Ocean Accident & Guarantee
Corporation Ltd,3 where the government’s claim for loss suffered as a result of a negligently
inflicted injury to a government employee (a magistrate) was refused. The Court refused the claim
on the basis that recognising it would lead to an unlimited number of claims arising from other
relationships, contractual or other, between the physically injured person and other persons who
may indirectly suffer economic harm as a result of the injury.4
The Court thus adopted a conservative approach and declined to expand Aquilian liability for
negligently interfering with contractual relations. Schreiner JA described the dangers inherent in
such expansion as follows: 5
Once one goes beyond physical proximity and considers the possibilities that may arise out
of the relationships, contractual or other, between the physically injured person and other
persons who may suffer indirectly, though materially, through his incapacitation, one is
immediately met with the prospects of an unmanageable situation. It is easy to imagine the
absurdities that would arise if all persons contractually linked to the injured person could
sue the careless injurer for the loss suffered by them. The case was put to us of the injured
building contractor who in consequence of his injury has to discontinue his contract, so that
his employees and the building owner and the architect and his sub-contractors and their
employees are all put to some loss. Insurance companies would also be a wide class of
plaintiffs who could bring actions when persons insured by them were negligently injured
or, presumably, killed, if the extension of liability contended for were recognised. In fact it
would be a rare accident that did not give occasion for a crop of actions at the suit of
persons who had made contracts with the injured party. The question whether in any
particular case there is a liability under the lex Aquilia towards a particular person cannot
be satisfactorily answered by any such wide proposition as that contended for on behalf of
the Government.

In a subsequent case, Dantex Investment Holdings (Pty) Ltd v Brenner,6 the Appellate Division did
not reject the possibility of imposing delictual liability for the negligent interference with a
contractual relationship and left open the possibility of future expansion of liability.7 In Minister of
Safety and Security v Scott 8 the Supreme Court of Appeal indicated that it would, in principle, be
willing to consider expanding delictual liability to negligent interference with a contractual
relationship in future.
Minister of Safety and Security v Scott9
In this case, the first respondent, Scott, was a professional hunter and conducted big game
hunting enterprises. He was also the chief executive officer of Scottco, the second
respondent, which owned a game ranch and conducted hunting safaris. Scottco contracted to
advertise its safaris in an American hunting magazine (Field & Stream) from February 2004.
The parties further agreed that Scottco would host hunting trips for American hunters, co-
ordinated by the publisher of the Field & Stream. The first American hunting party arrived in
South Africa in June 2004. On the night they arrived, Scott was arrested for handling a
firearm while under the influence of alcohol and consequently detained by the police. Scott
argued that the arrest and detention was wrongful because he had not ‘handled’ the firearm
for the purposes of the then applicable Arms and Ammunition Act 75 of 1969 and therefore
did not commit a statutory offence under the Act. As a result of his detention, he was unable
to manage the hunting party and the entire hunting trip was ruined.
Consequently Field & Stream cancelled its contract with Scottco. Scott and Scottco then
instituted delictual claims against the Minister of Safety and Security. While Scott attempted
to hold the Minister vicariously liable for his alleged unlawful arrest and detention, Scottco
sued the Minister in delict, alleging that the Minister’s employees had culpably and wrongfully
caused it pure economic loss of contractual income and profits. This claim was based on the
employees’ alleged negligent interference with the contract between Scottco and Field &
Stream.
Both claims were successful in the High Court, but the Supreme Court of Appeal rejected
Scottco’s claim based on the negligent interference with the advertising and safari contract.
The Court held that this was not a case of intentional interference and, because ‘the kind of
liability now sought to be imposed does not fall within historically recognised instances’,10 the
claim should fail. However, noting the argument that the current approach to negligent
interference with contractual relations is perhaps too restrictive, it suggested that any
 negligent conduct by a third party that involves the infringement of a contractual right or the
increase of contractual obligations ought on principle to be a basis for Aquilian liability.11 The
Court added that the fear of unlimited liability could be allayed by the correct application of
the elements of delictual liability.12 Assuming that Scottco’s claim was actionable under South
African law, the Court nevertheless did not impose liability, because it did not regard the
causing of pure economic loss in this case to be wrongful, and also considered the loss to be
too remote from the harm-causing conduct to satisfy the requirements of legal causation.13
With regard to wrongfulness, the Court confirmed that the police officers had relied on a
technically wrong basis for Scott’s arrest, thereby rendering the arrest and detention
wrongful. Nevertheless, the officers could lawfully have arrested Scott for assault with intent
to do grievous bodily harm. The technically incorrect arrest and detention paled in
comparison with Scott’s reprehensible behaviour the night of his arrest. Against this
background, the Court held that it would be unreasonable to impose delictual liability on the
Minister, especially because this was likely to create an ‘unascertainable class of potential
claimants’.14
On legal causation, the Court emphasised that there was no evidence to find that the
‘police officers knew of the contract between Scottco and Field & Stream magazine’ and
further that there ‘was no evidence that the police officers knew, let alone foresaw, that
Scott’s detention would have any impact on the planned elephant hunt, lead to the
cancellation of the contract between the respondents and Field & Stream magazine and
cause financial loss to Scottco’.15

Notwithstanding the general denial of delictual liability for the negligent interference with
contractual relations, there are some exceptions where courts have recognised liability in this
context; for instance, where the defendant negligently interferes with the plaintiff’s contractual
right to property.16 However, this is applicable only in cases where, together with the contractual
right and economic interest in the property, the holder of the right is also physically in possession
of the property and bears the risk of damage; for example, the instalment-purchaser of a motor
vehicle,17 the instalment-purchaser of land,18 or the borrower of a motor vehicle.19

PAUSE FOR Negligent interference with the personal right to support: the
REFLECTION dependent’s claim for loss of support
Other than the exceptional situations where liability is imposed for the
negligent interference with contractual relations, the South African law of
delict also recognises Aquilian liability for the negligent interference with
a personal right in the form of negligently causing the death of a
breadwinner. Dependants may institute a claim for loss of support arising
from the death of the breadwinner caused culpably and wrongfully by the
wrongdoer. This type of delictual liability is based on family law principles
and the claim for loss of support was historically restricted to cases
where a duty of support derives from a valid marriage, mainly concerning
spouses and children. However, as we will discuss further in Chapter 23,
courts have extended the group of claimants to include those whose right
to support derives from a relationship akin to marriage, but may not be a
legally valid marriage; for example, ‘a solemn marriage in accordance
with recognised and accepted faith’,20 a same-sex permanent life
relationship similar in other respects to marriage,21 or the right to support
of a divorced person under a court order dealing with maintenance after
dissolution of the marriage.22
We emphasise that courts have not yet extended delictual liability for
negligent interference with a personal right to support to cases where the
right of support is based on a contract between the plaintiff and the
person whose death or injury causes harm to the plaintiff.
17.3 Intentional interference with contractual relations
While the courts generally deny liability for negligently interfering with contractual relations, the
position is different when the interference with contractual relations occurs intentionally.
Intentional interference with a contractual relationship by a third party may give rise to delictual
liability towards the contracting party who suffered loss.23 In Dun and Bradstreet (Pty) Ltd v SA
Merchants Combined Credit Bureau (Cape) (Pty) Ltd 24 the Court described this category of
delictual liability as follows:
Moreover, incorporeal property, such as a personal right flowing from contract, also enjoys
a measure of protection in that a delictual remedy is available to a party to a contract who
complains that a third party has intentionally and without lawful justification invaded his
enjoyment of such property by inducing the other party to the contract to commit a breach
thereof … .

The courts have imposed delictual liability for the intentional interference with contractual relations
in the following circumstances:
• Where the defendant induced a party to a contract to breach that contract, the courts have
awarded a delictual claim to the other contracting party (the plaintiff).25 In Atlas Organic
Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd,26 where the managing director of a
company enticed some of the employees of a competing company to leave their respective
positions, take up employment at the competing company and sell the competitor’s product,
the Court held that a ‘delictual remedy is available to a party to a contract who complains that
a third party has intentionally and without lawful justification induced another party to the
contract to commit a breach thereof.’27 The Court stated that the causation of pure economic
loss in this context may be regarded as wrongful if the aim ‘is not to benefit from their services
but to cripple or eliminate the business competitor’. 28 The Constitutional Court has held that
these types of cases ‘may arguably be prima facie wrongful’. 29 However, it also sought to
emphasise that ‘the wrongfulness enquiry is “based on the duty” not to cause harm – indeed to
respect rights – and questions the reasonableness of imposing liability’ 30 and stated that the ‘[i]
nducement without harm to or infringement of the right to goodwill, will not usually be
wrongful.’ 31 Therefore, whether or not a plaintiff would have to prove wrongfulness in a so-
called inducement case will depend on the particular circumstances of the case.
• Where the defendant interferes with the contractual relationship between the plaintiff and
another person with the result that the plaintiff does not obtain the performance to which he or
she is entitled in terms of the contract, even if there is neither breach of contract nor conduct
by the defendant that amounts to enticement or inducement to breach of contract. In Country
Cloud Trading CC v MEC, Department of Infrastructure Development 32 the Constitutional
Court referred to this category of delictual liability as follows:
◆ The delict of intentional interference with contractual relations may be brought by a party
to a contract who asserts that a third party – a stranger to the contract – has intentionally
deprived it of benefits it would otherwise have gained from performance in terms of the
contract.
◆ For example, where a defendant intentionally continues to occupy premises which the
plaintiff has leased from the owner of the premises, well knowing that the incumbent lessee
is thereby deprived of its contractual rights under the lease.33 Two of these so-called
holding-over cases, in which the courts have sought to clarify and expand on the
requirements for a delictual claim in this context, are discussed below. The Constitutional
 Court has characterised this category of cases as cases where the defendant does not
‘simply cause the plaintiff to lose its right to occupy premises’, 34 but rather ‘usurp[s] that
right, appropriating it for itself’. 35

PAUSE FOR Frustrating the conclusion of a contract

REFLECTION In Makulu Plastics & Packaging CC v Born Free Investments 128 (Pty)
Ltd,36 the respondent, the owner and landlord of a building, concluded a
lease agreement with the appellant. The respondent had requested the
municipality to terminate the electricity supply to the leased premises on
the ground that the previous occupier had been liquidated, without
mentioning that a new tenant (the appellant) was occupying the
premises. When the appellant subsequently sought to contract with the
municipality for services, the latter refused to do so on the basis that the
respondent had denied any contractual relationship with the appellant.
The Court held that the conduct of the respondent may be regarded as
the intentional interference with a contractual relationship between the
appellant and the municipality. Therefore, it granted the appellant an
order interdicting the respondent from preventing the appellant from
entering into an agreement with the municipality for the provision of
municipal services.37 In effect the Court held that the respondent’s
conduct, which was aimed at frustrating the contract between the plaintiff
and the municipality, may be regarded as the intentional interference with
a contract.
To date, this judgment has not been confirmed or applied by either
the Supreme Court of Appeal or Constitutional Court. Do you agree with
the Court’s decision in Makulu?

Dantex Investment Holdings (Pty) Ltd v Brenner38

In this holding-over case the plaintiff (appellant) instituted a delictual claim against the
defendants (respondents) on the basis that the defendants had intentionally and wrongfully
interfered with the plaintiff’s lease agreement with a third party. In its particulars of claim the
plaintiff stated that it had entered into a written lease agreement with a third party in terms of
which it became entitled to occupy the leased premises as from a certain date and that, since
then, the defendants had been in wrongful occupation thereof. Furthermore, the particulars
stated that, despite being so demanded, the defendants had refused to vacate the premises
and that, arising from such wrongful occupation, the plaintiff had suffered pure economic loss
and would continue to suffer such harm until the defendants vacated the premises. The
defendants excepted to the plaintiff’s claim, which was upheld by the Court a quo.
On appeal, the Appellate Division accepted that the interference with contractual rights
can, under certain circumstances, constitute a delict, but noted that it was not certain what
the requirements for delictual liability are in this context.39 Both parties had accepted that
intent (dolus) is a requirement for liability in this context and the Court assumed without
deciding, for the purposes of the case, that the attitude of the parties was correct in respect
of the requirement of intent. However, Grosskopf JA noted that:

… the question whether culpa might not constitute a sufficient element of fault
to ground liability for damages for an unlawful interference with contractual
relations was not raised or debated in argument. Since there was in any event no
allegation of culpa in the pleadings I need say no more about this possibility.40

The Court held that, when claiming patrimonial loss based on the intentional and wrongful act
of another, the plaintiff is required to allege and prove that the defendant intended to cause
the plaintiff’s loss.41 However, in the present case, the plaintiff failed to do so because it
merely alleged that the defendants acted with knowledge of the plaintiff’s rights and that the
plaintiff had suffered a loss.
The defendants’ exception that the pleadings of the plaintiff did not disclose a cause of
action because it did not allege the requisite intent on the part of the defendants was
therefore upheld. In other words, the Court held that the plaintiff’s allegations did not
embrace all that was meant by intent (dolus). In fact, it held that the plaintiff’s pleadings were
not inconsistent with the defendants’ belief that the plaintiff would not suffer harm by being
kept out of the leased premises, and it is easy to imagine a situation in which this belief
arose. For instance, the defendants might believe that the plaintiff required the premises only
for future expansion, or that the plaintiff has, since entering into the lease, acquired more
suitable premises and would prefer not to take occupation under the lease.42
The Court further held that it was accepted that intent encompassed not only the intention
to achieve a particular result, but also the consciousness that such a result would be
wrongful.43 Although there might be policy considerations that could explain why a plaintiff
who relied on fault in the form of intent should not have to prove consciousness of
wrongfulness, such considerations did not apply in this case.44 The plaintiff should therefore
have alleged consciousness of wrongfulness on the part of the defendants, but also failed to
do so.45

Lanco Engineering CC v Aris Box Manufacturers (Pty) Ltd 46

The plaintiff concluded a lease agreement with the owner of business premises on 14
December 1989 in terms of which it was to take occupation on 1 February 1990. The
defendant was at that time the tenant of the premises in terms of a lease agreement which
was due to expire on 31 December 1989. The defendant failed to vacate the premises on that
date, and the plaintiff could consequently only take occupation of the premises on 2 April
1990. It was this holding over by the defendant which the plaintiff argued constituted an
intentional and wrongful interference with its lease agreement and caused it to suffer pure
economic loss.
The evidence indicated that, from at least early December 1989, the defendant had been
aware that the plaintiff was negotiating a lease for the relevant premises with its owner. The
defendant’s reaction had been to send a ‘rental’ cheque for January 1990 to the owner and,
when that was rejected and returned, he had instructed his attorneys to assert in writing (for
the first time) that in October 1989 he and the owner’s representative had concluded an oral
agreement in terms of which the defendant’s lease would be extended beyond 31 December
1989 and in any event until the premises to which it hoped to move were ready for
occupation.
The plaintiff instituted the Aquilian action for damages which it suffered as a result of the
defendant’s wrongful and intentional interference with its contractual relationship with the
owner of the premises. In considering the plaintiff’s claim, the Court confirmed that a party to
a contract who seeks to recover delictual damages from a third party for interfering with his
contractual relationship has to prove the causing of harm by wrongful and culpable
interference. As far as culpability is concerned, it is clear that the action was based on intent,
and the Court therefore did not decide as to whether negligence will also suffice. As to the
interference itself, it is neither necessary that the interference should result in a breach of the
contract, nor that the interference should consist of an incentive by the defendant upon the
party to the contract to commit the breach. Nonetheless, inducement and breach are
significant factors that courts should consider when determining if the particular interference
in a contractual relationship is wrongful.
On the facts, the Court found this declaration of an oral lease to be untrue and that the
attempt to send a ‘rental’ cheque for the period in dispute had been a dishonest and
mischievous attempt to support the untruthful declaration. The Court found that the
defendant had been aware that, should he fail to vacate the premises, the plaintiff would
suffer losses. The Court decided that:
• The defendant’s failure to vacate the premises had interfered with the plaintiff’s contractual
relationship.
• The defendant had the necessary intent.
• Even though the defendant’s conduct did not constitute inducement of a breach of contract, it
had wrongfully caused the plaintiff’s harm.

Country Cloud Trading CC v MEC, Department of Infrastructure Development47

The respondent, the MEC, had contracted with a third party, iLima, for the latter to build a
hospital. To comply with its obligations under the contract iLima had borrowed R12 million
from the appellant, Country Cloud. This loan agreement was subject to the condition that Tau
Pride, the MEC’s managing agent for the building contract, would repay Country Cloud the
R12 million out of the amount payable by the MEC to iLima. Further, one Lupepe, the chief
executive officer of iLima, was listed as a party to the loan agreement and stood as surety
and co-principal debtor for the loan. The MEC subsequently cancelled its contract with iLima
on the basis that iLima had made certain misrepresentations. This resulted in the eventual
liquidation of iLima, because it was relying on the money from the construction contract. Due
to iLima’s precarious financial position, it failed to repay Country Cloud the money it had
borrowed.
Country Cloud then instituted a delictual claim against the MEC for the pure economic loss
it had suffered, on the basis that the MEC had unjustifiably cancelled the contract with iLima
in circumstances where it had foreseen that Country Cloud would suffer harm as a result.
The MEC contended that Country Cloud had failed to establish the element of wrongfulness.
The facts of this case differ from the situations described in Atlas Organic Fertilizers,
Dantex and Lanco, because there was no inducement to breach a contract or holding over.
This was a novel case of interference with contractual relations, where a stranger to a
contract had suffered economic loss as a result of the intentional repudiation of the contract
by one of the contracting parties.48
The Supreme Court of Appeal dismissed the plaintiff’s claim. It held that the case did not
fall under the established categories of intentional interference with contractual relations. It
also held that there were insufficient policy-based considerations justifying a finding of
wrongfulness for the purposes of expanding delictual liability in this context.49 On appeal, the
Constitutional Court followed the line of reasoning adopted by the Supreme Court of Appeal.
The Court said the following:50

I cannot quibble with Country Cloud’s contention that, loosely speaking, the
department intentionally ‘interfered’ with Country Cloud’s contractual relations
with iLima by causing it to lose contractual benefits to which it was entitled. But
that fact does not in itself render conduct prima facie wrongful.

The legal category is narrower. The cases where conduct may arguably be prima
facie wrongful are limited. They involve a situation where a third party, A, the
defendant, intentionally induces a contracting party, B, to breach his contract
with the claimant, C, without lawful justification for doing so. But the department
did not induce iLima’s breach in the relevant sense. In these circumstances this
would require an act of persuasion directed at iLima with the intent that it
dishonour its agreement with Country Cloud. The defendant wrongdoer thereby
becomes an accessory to the primary wrong: the breach of contract. The act of
persuasion, paired with intent, establishes this accessory liability. [… However,]
iLima’s breach of the loan agreement with Country Cloud was simply a
consequence of the department’s conduct in cancelling the completion contract.
There was no act of persuasion.

Liability has also been established in cases where A refuses to vacate premises
owned by B, which interferes with the lease agreement between B and her
tenant, C, causing C loss. Both Dantex and Lanco involved these circumstances.
 While the plaintiff’s claim in Dantex failed because fault was not alleged, the
plaintiff in Lanco succeeded. But that case is different from Country Cloud’s.
The act of interference in Lanco involved the holding over of leased premises.
The defendant there did not simply cause the plaintiff to lose its right to occupy
the premises. The defendant usurped that right, appropriating it for itself. It also
did so in a manifestly ‘dishonest and mischievous’ way. The factual matrix in
this case — where the defendant’s supposed act of interference is the
cancellation of an entirely different contract — is thus distinguishable from that
which confronted the court in Lanco. The department’s responsibility for
Country Cloud’s loss is very different.

The Court found that Country Cloud’s claim was not on a par with the cases previously considered
to amount to intentional interference with contractual relationships. It further refused to extend
delictual liability to what was recognised as a novel situation, because the MEC’s causing of pure
economic loss to Country Cloud was not considered wrongful. The main policy-based reasons for
the Court’s stance was that, although the MEC had foreseen Country Cloud’s loss and could be
regarded as having caused it intentionally, and although imposing liability for a foreseen and
intended loss would not open the floodgates to limitless liability, Country Cloud was not vulnerable
to the risk of harm.51 Country Cloud could have claimed repayment from iLima in terms of the loan
agreement, could have taken cession of iLima’s claim against the MEC, or could have called up Mr
Lupepe’s suretyship obligation.52 The MEC’s reliance on state accountability was also
unsuccessful, inter alia because there was no corruption or fraud. The Court also held that the
imposition of delictual liability would interfere with the contractual relationship that existed
between Tau Pride and Country Cloud.53
Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd54
In this case, the Court was once again requested to impose delictual liability in a novel
situation. Masstores had concluded a lease agreement with the owner of a mall, Hyprop, in
terms of which it was agreed that Masstores was not allowed to trade as a general food
supermarket. The mall owner also entered into a separate lease agreement with Pick ’n Pay,
which secured for Pick ’n Pay a right of exclusivity by providing that the lessor would not
permit any other supermarket to be operated in the mall. When Masstores decided to launch
its Foodco concept at its store, Pick ’n Pay argued that it was operating a general food
supermarket and launched an interdict application against Masstores to restrain it from
interfering with the lease agreement between Pick ’n Pay and the owner of the mall.

The Supreme Court of Appeal confirmed the requirements for delictual liability in this
context:

an unlawful act which constitutes an interference in the contractual relationship

and which is committed with some form of dolus.55

The Court concluded that, in trading in competition with Pick ’n Pay, contrary to its contractual
restraint, after it was made aware of Pick ’n Pay’s right to exclusivity, Masstores had acted
wrongfully.56 Furthermore, it held that that Pick ’n Pay’s claim was based on the intentional
deprivation of a benefit that a contracting party would otherwise have obtained from
performance under a contract, more specifically the right to exclusivity in operating a
supermarket at the mall.57 The Supreme Court of Appeal held that the Constitutional Court
had recognised this category of delictual liability for intentional interference with contractual
relationships in Country Cloud58 and therefore Masstores was not required to prove
wrongfulness.59 After confirming that Masstores’s conduct had been intentional, the Court
concluded that Pick ’n Pay had proven the requirements for an action in delict.60
On appeal, the Constitutional Court held that the Supreme Court of Appeal’s interpretation
of the Country Cloud judgment had been incorrect:
 This Court’s judgment in Country Cloud is no authority for the proposition that
the deprivation of contractual rights in delictual claims for interference with
contractual relations is prima facie unlawful.61

It also concluded that this case, like Country Cloud, did not resemble holding-over cases
where the defendants not only deprived the plaintiffs of their contractual rights, but also
usurped those rights. Instead, the defendant could simply be said to have deprived the
plaintiff of its right.62 As a result, Masstores was required to prove wrongfulness.63
On the facts, the Court held that there was no wrongfulness. It emphasised the need for
free competition as a policy consideration and held that, generally, there is no legal duty on
third parties not to infringe contractually based exclusive rights to trade.64 Furthermore, the
Court distinguished this case from Lanco 65 insofar as the nature of the interest protected by
the contractual right differed. In Lanco, the protected contractual right was a property right
which operated against the whole world, while in this case the protection sought did not
operate in the same manner.66 The Court also followed its reasoning in Country Cloud and
pointed to Pick ’n Pay’s alternative contractual remedies that were available against the mall
owner.67 It approved the statement in Country Cloud that:

the law should hesitate before scrubbing out the lines [contracting parties] have
laid down by superimposing delictual liability [because this] could subvert their
autonomous dealings.68

The fact that Pick ’n Pay could have taken alternative steps to protect itself from avoiding loss
was a further important consideration against a finding of wrongfulness.69 Drawing on English
law, the Court emphasised that, merely because Masstores’s breach of its own lease with
Hyprop may have been wrongful, does not automatically entail wrongfulness in delict against
Pick ’n Pay.70 Finally, the fact that Pick ’n Pay had an alternative remedy also meant that one
of the requirements for an interdict had not been proven.71

COUNTER The categorisation of different types of interference cases

• In Country Cloud the Supreme Court of Appeal distinguished the facts before it
POINT
from cases such as Lanco and Dantex:

In those cases a delictual remedy is afforded to a party to a

contract who complains that a third party — who is a
stranger to the contract — has intentionally deprived him or
her of the benefits he or she would otherwise have obtained
from performance under the contract. Examples include
preventing a lessee from taking occupation of the leased
property in terms of the lease (Dantex); and enticing
another person’s employees to breach their contract (Atlas
Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty)
Ltd).72

• On appeal the Constitutional Court held that, although it may be said that the
Department intentionally ‘interfered’ with Country Cloud’s contractual relations
with iLima in a loose sense, it did not fit within the established categories of
delictual liability, namely holding over or inducement, and the Court ultimately
denied liability.73 The Court seems to suggest that the holding-over and
inducement-to-breach cases may be regarded as established categories of
delictual liability and therefore prima facie wrongful. The Court noted the
following:

Country Cloud was unable to bring its claim within the

established ambit of the delict of unlawful interference in a
 contract. Had Country Cloud succeeded in doing so, the
department’s conduct would have been regarded as prima
facie wrongful.74

• In Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd75 the Supreme Court of
Appeal identified two instances where a defendant may be held delictually
liable for intentionally interfering with a contract: 76 the first instance ‘results in
the contracting party not obtaining the performance to which it is entitled on the
contract’77 and the second scenario entails ‘a contracting party’s obligations
under the contract [being] increased’.78 The Court held that the first instance
was concerned with the infringement of personal right and that it included
holding-over cases as well as situations where employees are induced to
breach their contracts.79 The Court stated that the claim was based on the
‘intentional deprivation of a benefit a contracting party would otherwise have
obtained from performance under a contract’80 and held that the Constitutional
Court had referred to this type of case as the ‘usurpation of [a] right’ in Country
Cloud.81 It concluded that there ‘are therefore two types of delictual action in
interference cases, namely those where inducement or enticement feature and
others where there is a breach of a legal duty or the infringement of a
subjective right.’82
• On appeal, the Constitutional Court rejected this categorisation. It confirmed
that its judgment in Country Cloud:

did not equate the ‘usurpation of rights’ with a ‘deprivation

of interest’. [The Constitutional Court] did not use the latter
phrase, but more importantly it characterised the Lanco
decision as one that involved both the loss of the right to
use the premises and the taking over of that right, not as
similar or alternative losses. Whatever the merits of a pure
‘deprivation’ case may be, this Court’s decision in Country
Cloud cannot serve as authority that it is a case where
wrongfulness does not need to be established positively
but can be presumed.83 It also held that the Supreme Court
of Appeal’s attempt to make a twofold classification of
delictual interference with contractual relations cases, as
consisting of one category where only inducement is
required and others where a breach of a duty or
infringement of a subjective right is involved, was
‘unfortunate and wrong’.84

Do these attempts to categorise the different instances of intentional

interference with contractual relations contribute meaningfully towards
simplifying the issues? The Constitutional Court notably reaffirmed that
the law of delict:

is not a law of separate and distinct torts; it is one where all

forms of delict must conform to the general requirements of
Aquilian liability.85

Would it not, therefore, be more useful to approach the interference with

contractual relations as simply another instance of pure economic loss? It
would then follow that the primary question is whether the defendant’s
causation of pure economic loss may be regarded as wrongful. In this
regard, the following reasoning is pertinent:86

[W]rongfulness in the context of delictual liability for pure

economic loss is ultimately dependent on an evaluation
 based on considerations of legal and public policy. The
enquiry is thus: do these policy considerations require that
harm-causing conduct should be declared wrongful and
consequently render the defendant liable for the loss, or do
they require that harm should remain where it fell, ie with
the plaintiff?

In this context the policy considerations outlined in Chapters 9 and 16 are

relevant.

17.4 Conclusion
The legal position relating to the delictual liability for interference with contractual relations can be
summarised as follows:

• Apart from a few exceptional situations, there is generally speaking no delictual liability for
the negligent interference with contractual relations. Although the Supreme Court of Appeal
has recognised its willingness to impose delictual liability in this context, courts have not yet
done so.
• Courts have imposed delictual liability for the intentional interference with contractual
relations, where such interference occurs wrongfully.
• The intentional interference with contractual relations typically takes the form of inducing a
breach of contract or holding over.
• Because the interference with contractual relations is an instance of causing pure economic
loss, wrongfulness ultimately falls to be determined by reference to legal and public policy
considerations as well as constitutional norms.
• As far as intentional interference with contractual relations is concerned, plaintiffs are required
to prove both direction of will as well as consciousness of wrongfulness in order to prove
intent (dolus).

1 See Minister of Safety and Security v Scott 2014 (6) SA 1 (SCA) para 25.
2 See, generally, Hutchison ‘Relational economic loss (or interference with contractual relations): the last hurdle’ in
Scott and Visser (Eds) Developing Delict: Essays in Honour of Robert Feenstra (2001) at 133ff; first published as
Acta Juridica 2000, Juta: Cape Town.
3 1956 (1) SA 577 (A) at 585B–D.
4 At 585–586.
5 At 585–586.
6 1989 (1) SA 390 (A) at 395.
7 See also Lanco Engineering CC v Aris Box Manufacturers (Pty) Ltd 1993 (4) SA 378 (D) at 380–381 and section 17.3
below.
8 2014 (6) SA 1 SCA.
9 2014 (6) SA 1 SCA.
10 Para 31.
11 Para 31. See also Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 326.
12 Para 31; Neethling and Potgieter (2015) at 326.
13 Paras 32–40.
14 Para 36.
15 Para 39.
16 Maraisburg Divisional Council v Wagenaar 1923 CPD 94; Refrigerated Transport (Edms) Bpk v Mainline Carriers
(Edms) Bok 1983 (3) SA 121 (A); Spolander v Ward 1940 CPD 24. See also Van der Merwe Die beskerming van
vorderingsregte uit kontrak teen aantasting deur derdes (1959).
17 Lean v Van der Mescht 1972 (2) SA 100 (O).
18 Smit v Saipem 1974 (4) SA 918 (A).
19 Spolander v Ward 1940 CPD 24.
20 Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA
421 (SCA); 1999 (4) SA 1319 (SCA).
21 Du Plessis v Road Accident Fund 2003 (11) BCLR 1220 (SCA); 2004 (1) SA 359 (SCA).
22 Santam Bpk v Henery 1999 (3) SA 421 (SCA).
23 See Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC); Masstores
(Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2017 (1) SA 613 (CC).
24 1968 (1) SA 209 (C) at 215G–H. See also Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2)
SA 173 (T) at 202.
25 See Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2017 (1) SA 613 (CC) paras 15–26; Country Cloud Trading
CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) paras 27–32.
26 1981 (2) SA 173 (T).
27 At 202.
28 At 200–201.
29 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 30.
30 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2017 (1) SA 613 (CC) para 20; Country Cloud Trading CC v
MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 21. See also Loureiro v Imvula Quality
Protection (Pty) Ltd 2014 (3) SA 394 (CC) para 53.
31 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2017 (1) SA 613 (CC) para 21.
32 2015 (1) SA 1 (CC) para 27.
33 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA) para 8.
34 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 31.
35 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 31.
36 2013 (1) SA 377 (GSJ).
37 Paras 20–22.
38 1989 (1) SA 390 (A).
39 At 395.
40 At 395.
41 At 396. See also Geary & Son (Pty) Ltd v Gove 1964 (1) SA 434 (A) at 441D.
42 At 396.
43 At 396.
44 At 396.
45 At 396–397.
46 1993 (4) SA 378 (D).
47 2015 (1) SA 1 (CC). See also Country Cloud Trading CC v MEC, Department of Infrastructure Development 2014 (2)
SA 214 (SCA).
48 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) paras 16 and 27.
49 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2014 (2) SA 214 (SCA) paras 26–27.
50 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) paras 29–31.
51 Paras 33–43 and 51–61.
52 Paras 51–55.
53 Paras 44–50 and 62–66.
54 2017 (1) SA 613 (CC). See also Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA).
55 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA) para 19.
56 Para 20.
57 Paras 21–22.
58 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2015 (1) SA 1 (CC) para 22.
59 2017 (1) SA 613 (CC) para 18.
60 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA) para 23.
61 2017 (1) SA 613 (CC) para 24.
62 Para 25.
63 Paras 30–53.
64 Paras 33 and 36.
65 Lanco Engineering CC v Aris Box Manufacturers (Pty) Ltd 1993 (4) SA 378 (D).
66 Paras 37–38.
67 Paras 42–43.
68 Para 42.
69 Para 44.
70 Para 46.
71 Para 43.
72 2014 (2) SA 214 (SCA) para 26 (references omitted).
73 Paras 29–32.
74 Paras 28–30.
75 2016 (2) SA 586 (SCA) para 8.
76 Instead of ‘intentional interference’ the Court referred to ‘knowingly deprives a person of his rights under a contract’
– see further para 8.
77 Para 8.
78 Para 8.
79 Para 8.
80 Para 22.
81 Para 22.
82 Para 22.
83 Para 18.
84 Para 19.
85 Para 21.
86 Country Cloud Trading CC v MEC, Department of Infrastructure Development 2014 (2) SA 214 (SCA) para 18. See
also Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR
577 (CC); 2011 (3) SA 274 (CC) para 122.
 Chapter 18

Unlawful competition

18.1 Aquilian liability for unlawful competition

18.2 Forms of unlawful competition

18.3 Fault

18.4 Interests protected

18.5 Wrongfulness

18.6 Damages
|

18.1 Aquilian liability for unlawful competition

Remedies for patrimonial loss as a result of unlawful competition or unlawful trading are based on
the actio legis Aquiliae. The accepted position is that there is a modern general Aquilian action for
unlawful competition or unlawful trading in South African law.1
The Appellate Division recognised Aquilian liability for the causing of harm by unlawful
competition as early as 1922. In Matthews v Young 2 the Court stated:
In the absence of special legal restriction a person is without doubt entitled to the free
exercise of his trade, procession or calling … . But he cannot claim an absolute right to do so
without interference from another. Competition often brings about interference in one way
or another about which rivals cannot legitimately complain. But the competition … must
itself remain within lawful bounds. All a person can, therefore, claim is the right to exercise
his calling without unlawful interference from others. Such an interference would constitute
an injuria for which an action under the lex Aquilia lies if it has directly resulted in loss.
In Geary & Son (Pty) Ltd v Gove 3 the essence of the delict was characterised as ‘the wrongful
interference by a competitor with its rights as a trader’. In Atlas Organic Fertilizers (Pty) Ltd v
Pikkewyn Ghwano (Pty) Ltd 4 Van Dijkhorst J stated:
that the law of South Africa recognises and grants a general action in the case of unlawful
competition, based on the principles of the lex Aquilia.5

Courts have applied the general principles of Aquilian liability to many of the forms of unlawful
competition that are recognised in South African law, for example, passing off, 6 misrepresentation
of a rival’s own performance,7 acquisition and use of a competitor’s trade secrets,8 copying and
adopting a rival’s performance,9 competition in conflict with statutory provisions,10 and boycott.11
An important implication of recognising Aquilian liability in the context of unlawful
competition is that courts may develop new forms of protection against unlawful competition
within the ambit of the principles of the Aquilian action, even in the absence of a direct precedent
in case law. It is not necessary for an aggrieved competitor to bring his action within the framework
of one of the recognised forms of unlawful competition or another particular form of delict.12 The
Appellate Division in Schultz v Butt 13 confirmed this approach, which means that courts should
reconcile the influence of English law, in particular with regard to developing passing off 14 and
protecting trade secrets and confidential information,15 with Aquilian principles.16
Courts have endorsed the general application of Aquilian principles to cases of unreasonable
competitive or trading conduct, but the tendency has been, nevertheless, to fit the cases into one of
the established categories of unlawful competition. In Payen Components SA Ltd v Bovic CC 17
Schutz JA said:
Unlawful competition should not be added as a ragbag and often forlorn final alternative to
every trade mark, copyright, design or passing off action. In most such cases it is one of the
established categories or nothing.

Courts have developed the general principles of Aquilian liability into detailed sub-rules within the
established categories of unlawful competition. The Supreme Court of Appeal in Caterham Car
Sales & Coachworks Ltd v Birkin Cars (Pty) Ltd 18 set out the sub-rules concerning passing off in
detail. The essence of passing off is the misrepresentation that a particular business or product is
the same as, or connected to a business or product with an established reputation, belonging to the
complainant. This misrepresentation concerning the trade source, or connection of a product or
service has the potential to cause customers to move from the complainant’s established product or
service to the offender’s rival product or service.

18.2 Forms of unlawful competition

The following forms of unlawful competition occur often:
• Perhaps the most common form of unlawful competition is passing off. This is when persons
represent to others that their business or product is the same as, or connected to that of
another.19
• Unlawful competition can involve misappropriation of confidential information or trade
secrets. This is using or disclosing information that has commercial value, which was imparted
or received in confidence, often in a fiduciary or employment relationships.20
Dun and Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau (Cape) (Pty)
Ltd21
Corbett J held a company liable in damages for knowingly expanding its business with credit
information compiled by its competitor who had collected it using skill and labour. The
 competitor distributed this information on a confidential basis to clients. Corbett J quoted with
approval from the judgment of the United States Supreme Court in International News
Service v Associated Press,22 where copying and commercially using news material was held
to constitute unfair competition and where it was held that:

… defendant’s conduct differs from the ordinary case of unfair competition in

trade principally in this that, instead of selling its own goods as those of
complainant, it substitutes misappropriation in the place of misrepresentation,
and sells complainant’s goods as its own.

The offending conduct in the Dun and Bradstreet case was misappropriation of the credit
information, by unlawfully ‘obtaining’ and ‘using’ it.

Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd23

Van Dijkhorst J, with reference to a number of earlier cases, accepted that it is wrongful to
appropriate knowingly confidential information of a rival who has by the exercise of his skill
and labour compiled it. He also accepted that the law will not permit a person who has
obtained knowledge of a production method evolved by the intellectual effort of another to by-
pass this intellectual effort as his own and to use the method as a springboard to obtain an
unfair advantage for himself.

• Another form of unlawful competition, sometimes categorised under the rubric ‘leaning
on’ (aanleuning),24 involves misappropriating an established competitor’s name or label,25 or
advertising image.26
Stellenbosch Wine Trust Ltd v Oude Meester Group Ltd; Oude Meester Group Ltd v
Stellenbosch Wine Trust Ltd
27
Diemont J accepted that misappropriating a secret or confidential label design that had
been developed by the skill and industry of a competitor, and using that label to the detriment
of the competitor constituted unlawful interference with the trade of another. In this regard he
said the following:28

His conduct amounts to deliberate misappropriation of a business asset which

was acquired by another’s skill and industry. It is difficult to appreciate how this
conduct differs in principle from the conduct of a man who steals goods from
the shelves of a rival’s shop.

The Court here recognised that appropriating the label design was unlawful misappropriation
or unlawful interference with the trade of another, ‘within the principles of the lex Aquilia’. In
other cases too, courts have held competitive or trading conduct to be unlawful if it essentially
amounts to ‘stealing from the shelves of another’, that is, misappropriation in the form of
copying or otherwise exploiting the product of another’s creative abilities.

Union Wine Ltd v Edward Snell & Co Ltd29

The Court accepted in respect of the wine Bellingham Johannisberger that the
Johannisberger component of the name had become distinctive of the wine over a period of
more than 30 years and that this component of the name attracted goodwill.30 The Court
further accepted that the producers of this wine would suffer loss of custom as a result of the
marketing by a competitor of a wine called Edward Snell Johannisberger.31 Yet, because
passing off could not be proved (the bottles and labels were substantially different), protection
against the appropriation of the Johannisberger component of the name was not afforded.

• Wrongful trading conduct can occur outside the sphere of competition, for example, in cases of
misappropriating, for advertising purposes, the personal image of a celebrity,32 or
misappropriation for a different use of another’s trade name.33
 Capital Estate and General Agencies (Pty) Ltd v Holiday Inns Inc34
A well-known hotel group sought to restrain a property developer from using its trade name
for a shopping centre and a complex of duplex flats. The parties were engaged in dissimilar
trading activities, but the Court held that the use of the Holiday Inn name constituted passing
off, because there was the likelihood that members of the public would be deceived or
confused by the misrepresentation that the activities of the parties were associated. As there
was no common field of activity, and therefore no potential diversion of custom between the
parties, this was not passing off in its established form. The offending conduct involved
misappropriating the Holiday Inn trade name, which involved harm or potential harm in the
form of diluting the commercial value and impact of its trade name and potential harm to its
reputation.

• Wrongful misappropriation in the course of trade can also involve using the advertising value
of a fictional character created by another.35
Lorimar Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd36
The Court refused the application of the producers of the Dallas television series to restrain
the respondents from using characters, names and logos from the series to market a
restaurant, items of clothing and other products. The Court held that the respondents’
appropriation of such characters, names and logos from the series did not constitute passing
off, because the parties were not in competition in respect of the restaurant business or
clothing sales. Therefore, the applicant suffered no loss as a result of a diversion of custom.
The alternative basis for the applicant’s case was unlawful competition, in that the
respondents had foreclosed on the applicant’s opportunities to merchandise the characters,
names and logos themselves. The Court held that the important factors in assessing
unlawfulness were that the parties were not in competition with one another,37 and that the
applicant could have obtained protection in terms of trademark legislation, but failed to do
so.38 The conduct of the respondents was not considered to be wrongful misappropriation.
This case essentially involved the respondents’ unauthorised appropriation of the
merchandising value of characters created by the applicant. If the applicant had proved that it
had incurred loss as a result of such appropriation, in the form of foreclosed merchandising
opportunities or harm to reputation through the public associating products of inferior quality
with the applicant, the requirements of Aquilian liability were present. The commercial
phenomenon of ‘character merchandising’ was not proved in this case, but later accepted in
Federation Internationale de Football v Bartlett.39 Character merchandising involves the
causing of economic loss where a trader has appropriated, for its own commercial use, the
fictional characters created by another for its own commercial benefit.

• In cases of ‘misappropriation of performance’ (prestasie-aanklamping)40 the competitor copies,

or makes use of a product or process developed by another.
Schultz v Butt41
The applicant, Butt, the designer of a twin-hulled fibre-glass boat named Butt-Cat, sought to
interdict a competitor, Schultz, from manufacturing and marketing a boat with an identical hull
design. Years of experience, expertise and effort, and substantial costs had gone into the
Butt-Cat hull design and the venture was a financial success. Schultz got hold of a Butt-Cat
hull and used it to construct a mould and then used the mould to manufacture replica hulls for
a boat named Supercat, which was sold in competition with the Butt-Cat. Schultz went further
and registered the hull as a model with the Registrar of Designs. Butt applied for an interdict
against Schultz and an order for cancellation of the registration of the design, on the basis,
inter alia, of unlawful competition. The issue was essentially whether the infringement by
Schultz of Butt’s unregistered rights in respect of the design was unlawful.
The Court held that competitive conduct may fall within ‘a category of clearly recognised
illegality’,42 as in the case of trading in contravention of an express statutory prohibition,
 fraudulent misrepresentation that concerns one’s own product, passing off one’s product as
that of a competitor, and injurious falsehoods that concern a competitor’s product. However,
unlawfulness in competition is not limited to these categories, and one can also determine
unlawfulness by criteria such as fairness and honesty in competition.43 In this case, the
question was whether it can be unlawful to copy a design that is in the public domain and not
protected by copyright, patent or design legislation. In applying the criteria of fairness and
honesty, the Court considered, on the one hand, that the design was in the public domain
and that imitation is the life blood of competition, and on the other hand, that it may be
intolerable to allow one manufacturer to appropriate the product of another’s invention and
development without license or compensation. In this case, Schultz used one of Butt’s hulls,
with a shape that was developed over a long period, with considerable expenditure of time,
labour and money. Schultz used the hull to make a mould with which to make boats in
competition with Butt. Furthermore, he ‘added impudence to dishonesty’ by obtaining a
design registration in his own name for the hull designed by Butt.44 The Court considered this
conduct unlawful competition.
The offending conduct in the Schultz case was essentially misappropriating the hull
design, by ‘obtaining’, ‘copying’ and ‘using’ it unlawfully.

• Unlawful competition can also involve disparaging the good name of a business or
entrepreneur,45 organising a boycott (instigating others to exclude someone from a certain
commercial activity),46 comparative advertising,47 refusing to do business48 and parallel
importing. In respect of parallel importing, courts have decided that a trader who has an
exclusive contractual right to distribute goods in South Africa cannot succeed with an Aquilian
action against a rival trader who engages in parallel importation of the same goods, with the
same name, mark and get-up.49 However, where the importer or distributor adds his or her own
get-up or distinctive mark to the product, thereby establishing his or her own reputation for the
product, a parallel importer who imitates the get-up or mark may be liable on the basis of
passing off.50 In Taylor & Horne (Pty) Ltd v Dentall (Pty) Ltd 51 the Court refused to afford a
monopoly to an exclusive agent on the basis of his exclusive contractual rights.
• In specific circumstances wrongfully interfering with or depriving a competitor of the benefits
of a contractual right may also be unlawful competition.
Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd52
In this case Masstores and the owner of a mall, Hyprop, were parties to a lease agreement
which stipulated that Masstores was not allowed to trade in the mall as a general food
supermarket. The mall owner also entered into a separate lease agreement with Pick ’n Pay,
which secured for Pick ’n Pay a right of exclusivity as a general food supermarket, stipulating
that Hyprop would not permit any other supermarket to operate in the mall. When Masstores
launched its Foodco concept at its store, Pick ’n Pay applied for an interdict against
Masstores, alleging that it was operating a general food supermarket and asking the court to
restrain it from interfering with the lease agreement between Pick ’n Pay and Hyprop.
The Supreme Court of Appeal stated the requirements for delictual liability in this context
as follows:

an unlawful act which constitutes an interference in the contractual relationship

and which is committed with some form of dolus.53

The Court concluded that, by trading in competition with Pick ’n Pay, contrary to its own
contractual restraint, after it was made aware of Pick ’n Pay’s right to exclusivity, Masstores
had wrongfully and intentionally deprived Pick ’n Pay of its benefit of exclusivity under its
lease with Hyprop.54 The Supreme Court of Appeal held this category of delictual liability for
intentional interference with contractual relationships was recognised by the Constitutional
 Court in Country Cloud Trading CC v MEC, Department of Infrastructure Development55 and
therefore Masstores did not have to prove wrongfulness.56
On appeal, the Constitutional Court held that the Supreme Court of Appeal’s interpretation
of the Country Cloud judgment was incorrect:

[t]his Court’s judgment in Country Cloud is no authority for the proposition that
the deprivation of contractual rights in delictual claims for interference with
contractual relations is prima facie unlawful.57

It also concluded that this case, like Country Cloud, was unlike the holding-over cases, which
involved a defendant not only depriving the plaintiff of its contractual right but also usurping
that right, whereas in the present case the defendant could simply be said to deprive the
plaintiff of its right.58 As a result, Masstores was required to prove wrongfulness.59
On the facts, the Court held that there was no wrongfulness for purposes of delict. In this
regard the Court emphasised the need for free competition as a policy consideration and
held that, generally, there is no legal duty on third parties not to infringe contractually based
exclusive rights to trade.60 Furthermore, the Court distinguished this case from Lanco
Engineering CC v Aris Box Manufacturers (Pty) Ltd,61 because in Lanco the protected
contractual right was a property right which operated against the whole world, whereas in this
case the right sought to be protected was purely contractual.62 The Court followed its
reasoning in Country Cloud and took into account that Pick ’n Pay had alternative contractual
remedies against the mall owner, Hyprop.63 It approved the statement in Country Cloud that:

the law should hesitate before scrubbing out the lines [contracting parties] have
laid down by superimposing delictual liability [because this] could subvert their
autonomous dealings.64

The fact that Pick ’n Pay could have taken alternative steps to protect itself from avoiding loss
was a further important consideration against a finding of wrongfulness.65 The Court
emphasised that, merely because Masstores’s breach of its own lease with Hyprop may have
been wrongful, this does not automatically entail wrongfulness in delict against Pick ’n Pay.66
Finally, the fact that Pick ’n Pay had an alternative remedy also meant that one of the
requirements for an interdict had not been proven.67 (See also Chapter 17 ‘Interference with
contractual relations’.)

18.3 Fault
Unlawful competition, generally, involves intentional conduct.68 Competitors will often walk a very
fine line to gain an advantage in the market. In a case of passing off, for example, it is often
difficult to distinguish between an intention to compete and an intention to deceive.69 Boberg, in an
article on the role of fault in unlawful competition, wrote that all the cases in which courts held
unfair competition also to be unlawful competition, involved deliberate dishonesty, and that the
delict of ‘unfair competition’ was ‘born and nourished in a climate of calculated depravity’.70 This
may be so, but as a matter of general principle, intention is not required. The delict of unlawful
competition may involve negligently infringing goodwill.71 In this respect also, the action for
unlawful competition is no different from the general Aquilian action for patrimonial loss.

18.4 Interests protected

The interest protected by the delictual action for unlawful competition is essentially the right to
goodwill, that is, the right to attract custom, or the totality of attributes that entice clients or
potential clients to support a particular business.72 Infringing the right to goodwill may affect only
one component of goodwill, for example, the reputation of a business, the name of a product, the
get-up of a product, a trademark, or a trade secret. In the case of passing off, for instance, the
substratum or component of goodwill that is primarily affected is reputation.73
Although one can often identify the component of goodwill that is primarily affected by a
particular form of unlawful competition or trade, such as reputation,74 or the advertising value of
trademarks or trade names,75 for the purposes of liability, the proof of actual or potential harm to
trade or the inability to attract custom is the essential requirement.

18.5 Wrongfulness
Wrongfulness in the context of unlawful competition, as elsewhere, involves the question of
whether the causing of harm offends against the legal criterion of reasonableness or boni mores, a
criterion that involves assessing relevant considerations of public policy. These considerations
include fairness and honesty,76 the interests of the competing parties, the interests of society, the
morals of the market place, and the business ethics of that section of the community where the
norm is to be applied.77 Public policy as a general criterion for determining wrongfulness in
competition and trading constitutes:
a legal standard firm enough to afford guidance to the Court, yet flexible enough to permit
the influence of an inherent sense of fair play.78

Courts may often determine wrongfulness without necessarily referring to the criterion of
reasonableness or boni mores, the legal convictions of the community, or concomitant policy
considerations. It is settled law, for example, that a representation by one person that his or her
business or merchandise is that of another, which creates a reasonable likelihood of confusion
among customers or clients, constitutes unlawful competition in the form of passing off.79 Applying
the general criterion for wrongfulness becomes necessary where the conduct complained of caused
harm to the goodwill of another, but it does not fall within one of the established categories of
unlawful competition. In such cases, courts must take account of policy considerations in
developing the law.
Unlawful competition will hardly ever turn on a breach of duty between competitors. Instead,
it generally involves infringing the subjective right to goodwill, the right to attract custom
(werfkrag), or infringing a specific component of goodwill, such as copyright, a patent, a
trademark, a design, business or product name and reputation, a trade secret, or a personal
immaterial property right in the form of the right to earning capacity or personal goodwill.80
Therefore, the interests protected by the Aquilian action for unlawful competition or trading
generally constitute goodwill or a component of goodwill, such as a name or label,81 advertising
image,82 advertising value of a fictional character,83 or the personal image of a celebrity.84
Although, strictly speaking, there is no unlawful competition where the parties involved are
not in competition with each other,85 courts also use the term ‘unlawful competition’ to refer to
wrongful trading outside the sphere of competition, for example, in cases of misappropriating
another’s advertising image or name,86 or disparaging the good name of a business or
entrepreneur.87
It is relevant whether the person who caused the harm knew that his or her conduct would
cause the harm or had the motive to cause harm. However, the existence of knowledge, foresight or
motive is not conclusive in determining wrongfulness, and so one has to take it into account
together with other factors.88 The existence of fraud or dishonesty is an important factor in
indicating wrongfulness.89
 Other factors to be considered include the freedom to imitate ideas or designs in the public
domain,90 the freedom to exploit an established market for a particular type of product established
by a competitor,91 the manner of copying,92 the distinctiveness of the name or product copied,93 the
extent of the imitation and of the alleged offender’s own contribution to the product,94 and the
possibility of obtaining statutory protection for the product.95

18.6 Damages
Courts award damages for harm caused by unlawful competition according to the normal principles
that apply to patrimonial harm. Harm often involves a loss of custom to competitors.96 In most cases
it is possible to show that some harm has been suffered,97 but the extent of the harm is often
difficult to prove precisely.98 Consequently, courts tend not to require mathematically precise proof
of harm, and make an assessment of the loss ex bono et aequo on the probabilities indicated by the
evidence.99 Often the most effective remedy is an interdict ordering the wrongdoer to cease the
unlawful competition or trading.

1 See, generally, Loubser ‘Principles and policy in unlawful competition: An Aquilian mask?’ in Scott and Visser (Eds)
Developing Delict: Essays in Honour of Robert Feenstra (2000) at 168ff.
2 1922 AD 492 at 507.
3 1964 (1) SA 434 (A) at 440–441.
4 1981 (2) SA 173 (T) at 186.
5 Schultz v Butt 1986 (3) SA 667 (A) at 678; Dun and Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau
(Cape) (Pty) Ltd 1968 (1) SA 209 (C) at 218.
6 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 201–202; Lorimar
Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd, Lorimar Productions Inc v OK Hyperama Ltd, Lorimar
Productions Inc v Dallas Restaurant 1981 (3) SA 1129 (T) at 1138 and 1152ff; Tie Rack plc v Tie Rack Stores (Pty)
Ltd 1989 (4) SA 427 (T) at 445.
7 Geary and Son (Pty) Ltd v Gove 1964 (1) SA 434 (A) at 440–441; Elida Gibbs (Pty) Ltd v Colgate Palmolive (Pty)
Ltd (1) 1988 (2) SA 350 (W) at 357; William Grant & Sons Ltd v Cape Wine & Distillers Ltd 1990 (3) SA 897 (C) at
915.
8 Dun and Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau (Cape) (Pty) Ltd 1968 (1) SA 209 (C); Atlas
Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 189–196; Meter Systems Holdings
Ltd v Venter 1993 (1) SA 409 (W) at 426–427.
9 Schultz v Butt 1986 (3) SA 667 (A) at 678 and 682; Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty)
Ltd 1991 (2) SA 455 (W) at 471–475; The Concept Factory v Heyl 1994 (2) SA 105 (T) at 115–117.
10 See Silver Crystal Trading (Pty) Ltd v Namibia Diamond Corporation (Pty) Ltd 1983 (4) SA 884 (D) at 887.
11 Hawker v Life Offices Association of South Africa 1987 (3) SA 777 (C) at 780–781.
12 Dun and Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau (Cape) (Pty) Ltd 1968 (1) SA 209 (C) at 218;
Sea Harvest Corporation (Pty) Ltd v Irvin & Johnson Ltd 1985 (2) SA 355 (C) at 359–360.
13 1986 (3) SA 667 (A) at 678.
14 Boswell-Wilkie Circus (Pty) Ltd v Brian Boswell Circus (Pty) Ltd 1984 (1) SA 734 (N) at 742 (on passing off):
South African case law on the question is thin … . Plenty can be found, on the other hand, in England.
The persuasive authority of English decisions on passing off has long been accepted in this country.
15 Meter Systems Holdings Ltd v Venter 1993 (1) SA 409 (W) at 427–428.
16 Weber-Stephen Products Co v Alrite Engineering (Pty) Ltd 1990 (2) SA 718 (T) at 734–735; Hoechst
Pharmaceuticals (Pty) Ltd v The Beauty Box (Pty) Ltd (In Liquidation) 1987 (2) SA 600 (A) at 613.
17 1995 (4) SA 441 (A) at 453G.
18 1998 (3) SA 938 (SCA).
19 Caterham Car Sales & Coachworks Ltd v Birkin Cars (Pty) Ltd 1998 (3) SA 938 (SCA) at 947I–J.
20 Coolair Ventilator Co (SA) (Pty) Ltd v Liebenberg 1967 (1) SA 686 (W); Atlas Organic Fertilizers (Pty) Ltd v
Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T).
21 1968 (1) SA 209 (C).
22 248 US 215 (1918) at 242.
23 1981 (2) SA 173 (T) at 188.
24 See Mostert ‘Aanleuning: Skending van ’n handelsmerk en die reg op werfkrag buite mededingingsverband’ (1986)
49(1) THRHR at 173.
25 Stellenbosch Wine Trust Ltd v Oude Meester Group Ltd; Oude Meester Group Ltd v Stellenbosch Wine Trust Ltd 1972
(3) SA 152 (C).
26 Mostert ‘The right to the advertising image’ (1982) 99(3) SALJ at 413; Rutherford ‘Misappropriation of the
advertising value of trade marks, trade names and service marks’ in Neethling (Ed) Onregmatige
Mededinging/Unlawful Competition (1990) at 55.
27 1972 (3) SA 152 (C).
28 At 162B.
29 1990 (2) SA 180 (D); 1990 (2) SA 189 (C).
30 Union Wine Ltd v E Snell and Co Ltd 1990 (2) SA 189 (C) at 198C.
31 Supra at 198E.
32 See Neethling ‘Persoonlike immaterieelgoedereregte: ’n nuwe kategorie subjektiewe regte?’ (1987) 50(2) THRHR at
316.
33 Capital Estate and General Agencies (Pty) Ltd v Holiday Inns Inc 1977 (2) SA 916 (A).
34 1977 (2) SA 916 (A).
35 Lorimar Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd, Lorimar Productions Inc v OK Hyperama Ltd,
Lorimar Productions Inc v Dallas Restaurant 1981 (3) SA 1129 (T).
36 1981 (3) SA 1129 (T).
37 At 1154H.
38 At 1156H.
39 1994 (4) SA 722 (T).
40 See Neethling ‘Misappropriation or copying of a rival’s performance as a form on unlawful competition
(Prestasieaanklamping)’ (1993) 110(4) SALJ at 711.
41 1986 (3) SA 667 (A).
42 At 678.
43 At 678.
44 At 683.
45 Caxton Ltd v Reeva Forman (Pty) Ltd 1990 (3) SA 547 (A) at 560–561; Woodlands Dairy (Pty) Ltd v Parmalat SA
(Pty) Ltd 2002 (2) SA 268 (E).
46 Murdoch v Bullough 1923 TPD 495; Times Media Ltd v South African Broadcasting Corporation 1990 (4) SA 604
(W); Deneys Reitz v South African Commercial, Catering and Allied Workers Union 1991 (2) SA 685 (W).
47 Post Newspapers (Pty) Ltd v World Printing & Publishing Co Ltd 1970 (1) SA 454 (W), where the alleged
misrepresentation as to the comparative exposure value to advertisers of two publications was held to be mere puffery.
48 Times Media Ltd v South African Broadcasting Corporation 1990 (4) SA 604 (W) at 607.
49 Rusmarc (SA) (Pty) Ltd v Hemdon Enterprises (Pty) Ltd 1975 (4) SA 626 (W); Taylor & Horne (Pty) Ltd v Dentall
(Pty) Ltd 1991 (1) SA 412 (A).
50 Frank & Hirsch (Pty) Ltd v Roopanand Brothers 1987 (3) SA 165 (D) at 189; Salusa (Pty) Ltd v Eagle International
Traders 1979 (4) SA 697 (C) at 704–705.
51 1991 (1) SA 412 (A) at 422.
52 2017 (1) SA 613 (CC). See also Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA).
53 Masstores (Pty) Ltd v Pick ’n Pay Retailers (Pty) Ltd 2016 (2) SA 586 (SCA) para 19.
54 Paras 20–22.
55 2015 (1) SA 1 (CC) para 22.
56 2017 (1) SA 613 (CC) para 18.
57 2017 (1) SA 613 (CC) para 24.
58 Para 25.
59 Paras 30–53.
60 Paras 33 and 36.
61 1993 (4) SA 378 (D).
62 Paras 37–38.
63 Paras 42–43.
64 Para 42.
65 Para 44.
66 Para 46.
67 Para 43.
68 See, generally, Visser ‘Die rol van opset, en die boni mores by onregmatige mededinging’ (1989) 52 THRHR at 115.
69 Reckitt & Colman SA (Pty) Ltd v S C Johnson & Son SA (Pty) Ltd 1993 (2) SA 307 (A).
70 Boberg ‘The role of fault in determining the lawfulness of competition’ (1991) 54(1) THRHR at 43, 55.
71 Elida Gibbs (Pty) Ltd v Colgate Palmolive (Pty) Ltd (1) 1988 (2) SA 350 (W) at 354G–H and 357F–G; Link Estates
(Pty) Ltd v Rink Estates (Pty) Ltd 1979 (2) SA 276 (E) at 281.
72 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 182; A Becker and Co (Pty)
Ltd v Becker 1981 (3) SA 406 (A) at 417A; Caterham Car Sales & Coachworks Ltd v Birkin Cars (Pty) Ltd 1998 (3)
SA 938 (SCA) at 947G–H.
73 Caterham Car Sales & Coachworks Ltd v Birkin Cars (Pty) Ltd 1998 (3) SA 938 (SCA) at 947I–J.
74 Caterham Car Sales & Coachworks Ltd v Birkin Cars (Pty) Ltd 1998 (3) SA 938 (SCA) at 947I–J.
75 See Rutherford (1990) at 55.
76 Corbett ‘Aspects of the role of policy in the evolution of our common law’ (1987) 104(1) SALJ 52 at 62.
77 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 188.
78 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 188.
79 Capital Estate and General Agencies (Pty) Ltd v Holiday Inns Inc 1977 (2) SA 916 (A) at 929C; Caterham Car Sales
& Coachworks Ltd v Birkin Cars (Pty) Ltd 1998 (3) SA 938 (SCA) at 947E–F.
80 Atlas Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 182.
81 Stellenbosch Wine Trust Ltd v Oude Meester Group Ltd; Oude Meester Group Ltd v Stellenbosch Wine Trust Ltd 1972
(3) SA 152 (C).
82 Mostert (1982) at 413; Rutherford (1990) at 55.
83 Lorimar Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd, Lorimar Productions Inc v OK Hyperama Ltd,
Lorimar Productions Inc v Dallas Restaurant 1981 (3) SA 1129 (T).
84 See Neethling (1987) at 316.
85 Tie Rack plc v Tie Rack Stores (Pty) Ltd 1989 (4) SA 427 (T) at 445–446.
86 Capital Estate and General Agencies (Pty) Ltd v Holiday Inns Inc 1977 (2) SA 916 (A).
87 Caxton Ltd v Reeva Forman (Pty) Ltd 1990 (3) SA 547 (A) at 560–561.
88 Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty) Ltd 1991 (2) SA 455 (W) at 474J–476A.
89 Schultz v Butt 1986 (3) SA 667 (A) at 683H–I.
90 Premier Hangers CC v Polyoak (Pty) Ltd 1997 (1) SA 416 (A) at 423H–424C; Schultz v Butt 1986 (3) SA 667 (A) at
681A–E; Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty) Ltd 1991 (2) SA 455 (W) at 474E–G.
91 Taylor & Horne (Pty) Ltd v Dentall (Pty) Ltd 1991 (1) SA 412 (A) at 421I–422A.
92 Schultz v Butt 1986 (3) SA 667 (A) at 683H–I; Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty) Ltd
1991 (2) SA 455 (W) at 474E.
93 Reckitt & Colman SA (Pty) Ltd v S C Johnson & Son (SA) (Pty) Ltd 1995 (1) SA 725 (T) at 733B–G.
94 Bress Designs (Pty) Ltd v GY Lounge Suite Manufacturers (Pty) Ltd 1991 (2) SA 455 (W) at 472I–473A.
95 Lorimar Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd, Lorimar Productions Inc v OK Hyperama Ltd,
Lorimar Productions Inc v Dallas Restaurant 1981 (3) SA 1129 (T) at 1156H; Reckitt & Colman SA (Pty) Ltd v S C
Johnson & Son (SA) (Pty) Ltd 1995 (1) SA 725 (T) at 732F–G; Premier Hangers CC v Polyoak (Pty) Ltd 1997 (1) SA
416 (A) at 424I–426A.
96 Dun and Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau (Cape) (Pty) Ltd 1968 (1) SA 209 (C) at 221;
Harchris Heat Treatment (Pty) Ltd v Iscor 1983 (1) SA 548 (T).
97 Van Heerden v Paetzold 1917 CPD 221 at 224 (concerning damages for the effect of malicious statements); Atlas
Organic Fertilizers (Pty) Ltd v Pikkewyn Ghwano (Pty) Ltd 1981 (2) SA 173 (T) at 204–206 (concerning the
assessment of loss of profit); Link Estates (Pty) Ltd v Rink Estates (Pty) Ltd 1979 (2) SA 276 (E) at 286–287.
98 Lorimar Productions Inc v Sterling Clothing Manufacturers (Pty) Ltd, Lorimar Productions Inc v OK Hyperama Ltd,
Lorimar Productions Inc v Dallas Restaurant 1981 (3) SA 1129 (T) at 1138; Hushon SA (Pty) Ltd v Pictech (Pty) Ltd
1997 (4) SA 399 (SCA).
99 International Tobacco Co (SA) Ltd v United Tobacco Co (South) Ltd (1) 1955 (2) SA 1 (W) at 17; Caxton Ltd v Reeva
Forman (Pty) Ltd 1990 (3) SA 547 (A) at 573; Hushon SA (Pty) Ltd v Pictech (Pty) Ltd 1997 (4) SA 399 (SCA) at 412
–413.
 Chapter 19

Product liability

19.1 Introduction

19.2 Aquilian liability

19.3 Conclusion

19.1 Introduction |
In South Africa, product liability in delict, within the common-law framework of Aquilian liability,
is fault-based. However, legislation introducing strict liability came into effect in 2011,1 which
brought South African law into line with the position in many other jurisdictions. This chapter deals
with Aquilian liability under common law for harm caused by defective products. Chapter 39 deals
with strict liability for harm caused by defective products introduced by section 61 of the Consumer
Protection Act 68 of 2008.

19.2 Aquilian liability

In the area of product liability, the law of delict supplements the limited scope of protection
afforded by the law of contract. For example, in a contract of sale, a dealer in goods is liable to a
purchaser for latent defects where he or she publicly professes to have attributes of skill and expert
knowledge in respect of the goods sold. The Appellate Division accepted this in Kroonstad
Westelike Boere Ko-operatiewe Vereniging Bpk v Botha 2 and in Langeberg Voedsel Bpk v
Sarculum Boerdery Bpk.3 Schutz JA commented as follows on the Kroonstad decision:
It remains to add that it is not only in this small, if important, branch of the law of purchase
and sale, but in the much wider field of product liability generally that our law may be
perceived to have lagged behind.
In Ciba-Geigy (Pty) Ltd v Lushof Farms (Pty) Ltd 4 the Appellate Division commented on the scope
of delictual liability for defective products:
If a manufacturer produces and markets a product without conclusive prior tests, when the
utilisation thereof in the recommended manner is potentially hazardous to the consumer,
such negligence on the part of the manufacturer may expose him to delictual liability to the
consumer. Where the consumer does not acquire the product directly from the
manufacturer, and the manufacturer is thus a third party, such liability amounts to what is
sometimes termed ‘product liability’. A contractual nexus between the manufacturer and
the consumer is not required. Although the historical origin of the manufacturer’s liability is
an agreement between the manufacturer and the distributor, the liability, which arises from
the manufacture and distribution of the product, extends via the other contracting party to
any third party who utilises the product in the prescribed manner and suffers damage as a
result thereof. It follows as a matter of course that a manufacturer who distributes a
product commercially, which, in the course of its intended use, and as the result of a defect,
causes damage to the consumer thereof, acts wrongly and thus unlawfully according to the
legal convictions of the community.

A delictual claim for harm caused by a defective product requires proof of both wrongfulness and
negligence. In the first instance, the plaintiff must show that the defendant has wrongfully caused
harm by producing or distributing a defective product, in breach of its duty to the consumer.
Defectiveness of the product forms part of the wrongfulness enquiry. Wrongfulness essentially
involves a breach of the legal duty of a manufacturer not to expose persons acquiring and using the
products to harm. Van den Heever JA in Herschel v Mrupe,5 with reference to the decision of the
House of Lords in M’Alister (or Donoghue) v Stevenson,6 analysed the duty as follows:7
By putting into circulation potentially harmful things … the manufacturer is not merely
exercising a legal right but encroaching upon the rights of others not to be exposed, when
going about their lawful occasions and when accepting the implied general invitation to
acquire and use such commodities, to danger without warning and without their having a
reasonable opportunity to become aware of such danger before use. In other words, it is an
encroachment upon the rights of others to set hidden snares for them in the exercise of their
own rights. To refrain from doing so is a duty owing to the world at large …

The presence of a defect is a necessary element of wrongfulness. Assessing the defectiveness forms
part of the enquiry into wrongfulness, involving consideration of the legal convictions of the
community, boni mores or general reasonableness. The position in this regard has been summarised
as follows: 8
… a manufacturer has a general duty to take reasonable steps to ensure that defective
products do not reach the market or, if they do, to withdraw them from the market, or to
take other steps to ensure that no harm ensues from the presence of the product on the
market. The criterion of reasonableness coupled with the community’s concept of what
behaviour is reasonable in given circumstances is flexible enough to take into account such
factors as the type of product, the nature of the manufacturer’s business enterprise, the
customs and practices prevailing in a particular trade or industry, the amount of knowledge
and expertise of potential purchasers and users of the product, abnormal use, and the
specific stage in the production process during which a defect originated. The last-
mentioned factor may influence the duties of a manufacturer in different ways. At the stage
of planning or design the manufacturer must take into account the most recent knowledge
available in his field.

To date, the South African law of delict has not developed detailed rules for different forms of
product defects (manufacturing, design or warning defects) and courts, therefore, apply general
principles. This general approach can be interpreted to mean that a product will be considered
defective if it is unreasonably dangerous: 9
The test is flexible enough to take into account such factors as the type of product, the
nature of the manufacturer’s business enterprise, the customs and practices prevailing in a
particular trade or industry, the amount of knowledge and expertise of potential purchasers
and users of the product, abnormal use, and the specific stage in the production process
during which a defect originated. The last mentioned factor may influence the duties of a
manufacturer in different ways. At the stage of planning or design the manufacturer must
take into account the most recent knowledge available in his field.

As well as proving wrongfulness, the plaintiff must prove negligence on the part of the
manufacturer or distributor. One determines negligence by assessing the producer’s actions against
the standard of care a reasonable person (diligens paterfamilias) would adhere to.10 The enquiry is
whether a reasonable person would have foreseen the harm and would have taken steps to prevent
it, and whether the defendant failed to take such steps.
It is not only manufacturers that may be at fault. The seller of a product may also be at fault, in
which event the negligence issue involves the question of whether, in the circumstances, the seller
had a duty to inspect the product before the sale.11 The problem of proving negligence is an obstacle
for a plaintiff who brings an Aquilian action for harm caused by a defective product, despite the
potential application of the res ipsa loquitur doctrine, which in certain cases allows the inference
that the product did not meet the appropriate standard or that the manufacturer was negligent.12
Another obstacle for the plaintiff in a product liability case is the onus to prove the causal link
between the defect and the harm suffered. This may present an even more complex task than
proving that a defect exists, because the plaintiff has to collect technical data about the product and
the accident together with evidence of the conduct of the individuals concerned as well as
environmental factors.
Wagener v Pharmacare Ltd; Cuttings v Pharmacare Ltd13
The plaintiffs alleged that the manufacturer of a local anaesthetic was strictly liable for breach
of its duty to ensure that the product was safe. The manufacturer raised an exception to this
part of the particulars of the claim on the basis that it disclosed no cause of action, for want of
the allegation that there was dolus or culpa in the manufacture, sale or distribution of the
product. The plaintiffs, referring to the development of forms of strict product liability in
Europe and America, argued that the law in South Africa had reached the stage where such
liability should be imposed.

They argued that there was a constitutional need to develop the common law in this regard, to give
proper effect to the constitutional right to bodily integrity. The Court said the following on the
distinct elements of wrongfulness and negligence:14
In deciding the issues raised by the appeal it must be accepted, as regards the facts, that the
Regibloc in question was manufactured by the respondent, that it was defective when it left
the respondent’s control, that it was administered in accordance with the respondent’s
accompanying instructions, that it was its defective condition which caused the alleged harm
and that such harm was reasonably foreseeable. It must also be accepted, as far as the law is
concerned, indeed it was not disputed, firstly, that the respondent, as manufacturer,
although under no contractual obligation to the appellant, was under a legal duty in
delictual law to avoid reasonably foreseeable harm resulting from defectively manufactured
Regibloc being administered to the first appellant and, secondly, that that duty was
breached. In the situation pleaded there would therefore clearly have been unlawful conduct
on the part of the respondent: Ciba-Geigy (Pty) Ltd v Lushof Farms (Pty) Ltd.15 The essential
 enquiry is whether liability attaches even if the breach occurred without fault on the
respondent’s part.

The Court acknowledged that it had a role to develop the common law, but declined to do so in this
case, on the basis that courts should develop the common law with circumspection and that the duty
of making law is primarily that of the legislature. The Court accepted that the manufacturer had
wrongfully caused harm by selling products that did not conform to its own specifications (a
manufacturing defect). However, the Court would not impose liability without proof of fault on the
manufacturer. The case illustrates the difficulty of proving fault in respect of a defect in a certain
batch of a pharmaceutical product. It also raises the policy question of whether the risk of such a
defect should be borne by the manufacturer.
Freddy Hirsch Group (Pty) Ltd v Chickenland (Pty) Ltd16
The Court held that the plaintiff, Chickenland, had an alternative claim in delict for economic loss
caused by delivery to it of spices containing a banned colourant. The delictual claim was based on
the defendant’s awareness that breach of contract in delivering spices containing the banned
colourant would or could reasonably cause loss to the plaintiff, including the costs associated with
the recall and replacement of the affected products. The Court accepted that, in principle, the
same facts may give rise to a claim for damages in contract and alternatively in delict, and also
that the breach of a contractual duty is not per se wrongful for the purposes of Aquilian liability.17
The negligent causing of pure economic loss is not prima facie wrongful, but in this case policy
considerations such as knowledge of the potential harm, the fact that the loss was single and
determinate, and the manufacturer’s general duty to take reasonable steps to ensure that
defective products do not reach the market indicated wrongfulness for the purposes of delictual
liability.

19.3 Conclusion
The problems of proving defectiveness and negligence were the main catalysts for reforming the
law on product liability in South Africa. Chapter 39 deals with changes to the law brought by the
introduction of strict liability for harm caused by defective products, in terms of section 61 of the
Consumer Protection Act 68 of 2008.
Because the statutory remedy introduced by the Consumer Protection Act eliminates the need
to prove negligence on the part of a manufacturer, importer, distributor and retailer of defective
products (‘goods’) which cause harm, it is likely that claims for damages involving defective
products will in future in most cases be brought in the form of the statutory remedy. However, the
common-law liability for defective products continues to exist as an alternative basis for recovery
of damages for harm caused by defective products. In unusual circumstances not covered by the
provisions of section 61 of the Consumer Protection Act, claims will still be brought by way of an
Aquilian action.

1 The Consumer Protection Act 68 of 2008.

2 1964 (3) SA 561 (A).
3 1996 (2) SA 565 (A) at 572F–H.
4 2002 (2) SA 447 (SCA); [2002] 2 All SA 525 (A) (29 November 2001) paras 64 and 66.
5 1954 (3) SA 464 (A) para 487; see Combrinck Chiropraktiese Kliniek (Edms) Bpk v Datsun Motor Vehicle
Distributors (Pty) Ltd 1972 (4) SA 185 (T).
6 [1932] AC 562, 1932 SC (HL) at 31.
7 1954 (3) SA 464 (AD) at 486F–487A.
8 Van der Merwe and De Jager ‘Products liability: a recent unreported case’ (1980) 97(1) SALJ 83 at 88–89. Reprinted
by kind permission of Juta and Company Ltd.
 9 Van der Merwe and De Jager (1980) 97 SALJ 83 at 88–89.
10 A Gibb & Son (Pty) Ltd v Taylor & Mitchell Timber Supply Co (Pty) Ltd 1975 (2) SA 457 (W).
11 A Gibb & Son (Pty) Ltd v Taylor & Mitchell Timber Supply Co (Pty) Ltd 1975 (2) SA 457 (W).
12 Bayer South Africa (Pty) Ltd v Viljoen 1990 (2) SA 647 (A) at 662.
13 2003 (4) SA 285 (SCA).
14 2003 (4) SA 285 (SCA) para 7.
15 2002 (2) SA 447 (SCA).
16 2011 (4) SA 276 (SCA).
17 Para 33.
 Chapter 20

Breach of a statutory duty

20.1 Liability for breach of a statutory duty: A question of statutory

interpretation and wrongfulness

20.2 Typical factors that indicate wrongfulness

20.1 Liability for breach of a statutory duty: A question of

statutory interpretation and wrongfulness
|
A person who suffers harm from the culpable breach of a statutory duty or exercise of a statutory
power may claim in delict for damages. To succeed with such a claim, the plaintiff must prove the
normal elements of a delict, including the wrongful causing of harm. This means that the plaintiff
has to prove that compliance with a particular statutory duty is of direct interest to him or her and
that breach of that duty has caused him or her harm. It is not enough for the plaintiff to say that it is
in the common interest of all citizens that statutory duties are complied with.1 Also, on proper
interpretation, the statute in question must allow or at least be consistent with a delicual action for
damages by a person with the requisite interest who suffers harm that results from a breach of the
statutory duty.
To determine whether they should grant such an action, courts examine the content and
purpose of the statutory duty or power, and the implications of imposing delictual liability.
Following the rules of statutory interpretation, courts assess the intention of the legislature to
determine whether the statute was intended to allow the possibility of delictual liability for breach
of the duty it creates. If it did intend this, then the statute imposes a delictual duty on the defendant
towards the plaintiff. It follows that causing harm by breaching such a duty may be regarded as
wrongful and may involve delictual liability, provided that the other requirements for liability are
met.
Courts look at the wording and purpose of the statute as a whole to determine:
• The objects sought to be achieved by the legislature
• The nature of the powers conferred by the statute
• The duties and procedures involved in the exercise of such powers
• The remedies available to persons harmed by a breach of duty, both in terms of the statute
itself and the general principles of administrative law.2

Courts focus on the content and purpose of the statutory duty, and the implications of breach of the
duty, rather than on the infringement of the plaintiff’s right. The mere fact that the defendant acted
in breach of a statutory duty is not conclusive of wrongfulness for the purposes of delictual
liability.3 In South African Hang and Paragliding Association v Bewick 4 the Supreme Court of
Appeal said:
Even on the assumption that the appellants had failed to perform a duty imposed upon them
by statute, the question remains whether their omissions were wrongful in the delictual
sense. To the uninitiated it may sound contradictory to say that omissions to comply with
statutory obligations are not wrongful. But that impression loses sight of the special meaning
attributed to the element of wrongfulness in the context of delictual liability. As I have said
by way of introduction, wrongfulness in this context means that, in accordance with judicial
determination, considerations of public and legal policy dictate that it is reasonable to
impose delictual liability on the defendant for the harm caused by the omission involved.

To this one may add that the mere breach of a statutory duty itself does not necessarily constitute
negligence. Whether or not the defendant was negligent is ultimately dependent upon the
reasonableness of his or her conduct, that is, whether or not a reasonable person in the position of
the defendant would have foreseen the harm occurring and, if so, whether such reasonable person
would have taken steps to prevent the harm.5
Knop v Johannesburg City Council6
The city council made an error in granting an application for subdivision of a property. It then
later informed the applicant that the approval could not stand, because it was in conflict with
an existing town planning scheme. The Court held that the process of interpretation to
determine whether a delictual remedy arises from the breach of a statutory duty or negligent
exercise of a statutory power comes down to the following:

The essence of the plaintiff’s claim is that the Council caused him loss by the
negligent exercise of a statutory power. The source of the power being statutory,
it is necessary to examine the legislation by which it was brought into being, for
it is self-evident that the intention of the legislature is an important, and may
possibly be a decisive, feature of the circumstances material to the
determination of whether or not a legal duty existed. The legislative intention is
to be ascertained with reference to the nature of the powers conferred, the
nature of the duties involved in their exercise, the procedures prescribed for
their exercise and for persons aggrieved by it to obtain redress, and the objects
sought to be achieved by the legislature.

In this case, the applicant claimed damages for losses incurred as a result of the delayed
development of the property. The Court took into account, as a matter of policy, that there
was an administrative procedure for an aggrieved person to obtain redress in the event of
refusal of his application. It also considered that potential liability in delict could unduly
hamper the local authority in carrying out its statutory duty to consider and dispose of such
applications expeditiously. The application in this case was defective, because it did not
comply with the existing town planning scheme, and the applicant should have ensured such
compliance. In the circumstances, it would be contrary to the objective criterion of
reasonableness to hold the local authority liable for damages, and it would also offend the
legal convictions of the community.
Olitzki Property Holdings v State Tender Board7
In this case, a provincial government invited tenders for office accommodation. The appellant
obtained an option to purchase a building and tendered to provide office space in it to the
provincial government. When its tender was not accepted, the appellant instituted a claim for
damages for lost profit against the State Tender Board and the provincial government. The
alternative bases of the claim were:
• That there was a breach of the appellant’s right to a fair, public and competitive system of
tendering as required by section 187(2) of the Constitution of the Republic of South Africa
Act 200 of 1993, in that the provincial government, an interested party, had influenced the
Tender Board in its decision to award the tender
• That the interfering conduct was an infringement of the appellant’s fundamental right to
administrative justice enshrined in section 24(a), (b) and (c) of the Interim Constitution,
entitling it to damages for its lost profit.

The Supreme Court of Appeal held that liability for breach of a statutory duty is essentially a
matter of wrongfulness, determined by the general criterion of reasonableness. In turn, this
criterion is based on considerations of morality and policy, the legal convictions of the
community, and constitutional norms, values and principles. The other main issue is statutory
interpretation: on determining whether one can infer that the statute creates a legal duty, and
therefore confers a right of action, courts must consider the statute as a whole, taking into
account its objects and provisions, the circumstances in which it was enacted, and the type of
mischief it was designed to prevent.8
The Court held that the Interim Constitution prescribed a minimum content for future
legislation on tender procedure. Until such legislation was adopted, parties had to apply the
State Tender Board Act 86 of 1968. Neither the provisions of this Act, nor section 187 of the
Interim Constitution, justified the conclusion that a tenderer could claim lost profits for breach
of the applicable statutory provisions. Tender legislation could limit remedies for breach of
such legislation to interdicts, review, or out-of-pocket losses caused by actual expenses
arising from the breach complained of. In the absence of a specific provision in the State
Tender Board Act or the Interim Constitution, judicial interpretation did not allow recognition
of a claim for lost profits. Its contentious nature means that such a claim is a matter for the
national and provincial legislatures, not for judicial inference. A claim for lost profits is also
not ‘appropriate relief’ in terms of section 7(4)(a) of the Interim Constitution. In the
circumstances of this case, applying for an interdict would have been ‘appropriate relief’.
Such an application would have both anticipated the dispute and eliminated the loss claimed
by the appellant. In the circumstances of the case, and in view of the availability of alternative
remedies, the claim for lost profit was not an appropriate constitutional remedy.

Steenkamp NO v The Provincial Tender Board of the Eastern Cape9

This case concerned a claim for damages by a successful tenderer whose tender award was
later set aside on account of negligence on the part of the tender board. The Constitutional
Court summarised the factors relevant to assessing wrongfulness in the context of negligent
performance of a statutory duty:10

Our courts – Faircape, Knop, Du Plessis and Duivenboden – and courts in other
common law jurisdictions readily recognise that factors that go to wrongfulness
would include whether the operative statute anticipates, directly or by inference,
compensation of damages for the aggrieved party; whether there are alternative
remedies such as an interdict, review or appeal; whether the object of the
statutory scheme is mainly to protect individuals or advance public good;
whether the statutory power conferred grants the public functionary a discretion
in decision-making; whether an imposition of liability for damages is likely to
have a ‘chilling effect’ on performance of administrative or statutory function;
whether the party bearing the loss is the author of its misfortune; whether the
harm that ensued was foreseeable. It should be kept in mind that in the
 determination of wrongfulness foreseeability of harm, although ordinarily a
standard for negligence, is not irrelevant. The ultimate question is whether on a
conspectus of all relevant facts and considerations, public policy and public
interest favour holding the conduct unlawful and susceptible to a remedy in
damages.

20.2 Typical factors that indicate wrongfulness

The typical factors that courts take into account when deciding whether the breach of a statutory
duty is wrongful for the purposes of delictual liability, include the following:
• Could the plaintiff have averted the harm? Did the breach of duty materially contribute to the
harm?
Premier, Western Cape v Faircape Property Developers (Pty) Ltd11
The Supreme Court of Appeal confirmed that state officials, including employees of local
authorities and members of government at every level, are accountable for their decisions.
They must perform their statutory duties without negligence. Failure to exercise due care, in
circumstances where they owe a duty to members of the public to act responsibly and without
causing loss or harm, should involve liability for the damage caused. However, in this case,
the Court held that the legislation governing an application to remove restrictive conditions of
title from land placed no duty on the responsible Minister to check that the applicant had
complied with all the procedures laid down. The Minister only had to satisfy himself that it was
desirable in the interests of the area to remove the restrictive conditions of title. Therefore, the
Court could not hold the Minister liable for the applicant’s losses where an application had
first been granted and later overturned because of non-compliance with the prescribed
procedures.

• Was the statutory duty intended to benefit the plaintiff as one of a specific class of persons?
Where the statute was enacted in the public interest and not to protect individual interests, a
court may refuse to recognise an action for damages, but may grant an interdict. In the context
of unlawful competition, the ‘licence’ cases deal with the situation where one trader alleges
that he or she is incurring losses as a result of competition from another trader, who does not
hold the necessary statutory licence or permit to trade, or who otherwise contravenes a
statutory duty or prohibition. For example, in Patz v Greene and Co 12 a trader who conducted
business near a mining compound applied for an interdict against a rival trader who traded on
claim land, in contravention of a specific statutory prohibition. The Court accepted, on
principle, that it is wrongful to cause loss to a rival trader through conduct expressly prohibited
by statute. (However, the Court did not grant the interdict, because the application did not
contain all the necessary allegations to found the cause of action.) In Lascon Properties (Pty)
Ltd v Wadeville Investment Co (Pty) Ltd,13 where a duty in terms of a regulation to prevent
contaminated water from escaping was in issue, the Court held:14
In my view, the regulation was prima facie enacted for the benefit of the owners
of land which might be polluted as a result of the actions of a mining company.
… This having been the purpose of the regulation, it seems to me that the
Legislature would not have imposed an obligation to prevent the escape of
noxious water without intending persons harmed thereby to be entitled to be
compensated by the person permitting the water to escape. ‘For, if it were not
so, the statute would be but a pious aspiration’ … The regulation does not
impose a standard of care but imposes a duty in absolute terms. In my view, it
must be inferred from such imposition of a duty that those who have suffered as
 a consequence of a failure to comply with such duty should be entitled to
compensation therefor.

• Was the harm suffered by the plaintiff of the type that the statute intended to guard against? In
Da Silva v Coutinho 15 the Court held that the owner of a motor vehicle was liable in delict for
not complying with a statutory duty (at that time, compensation for road accident victims was
based on a system of compulsory insurance) to provide the plaintiff with a declaration of
insurance in respect of the motor vehicle. Compliance with this duty would have enabled the
plaintiff, who was injured in an accident involving the motor vehicle, to claim damages from
the insurance company concerned. In effect, the defendant had therefore wrongfully caused the
plaintiff’s loss.
• Are there alternative remedies or sanctions, including criminal sanctions, for breach of the
statutory duty? According to the judgment in the leading case of Knop v Johannesburg City
Council 16 the wrongfulness of causing harm partly depends on whether alternative redress for
persons harmed by the breach of a statutory duty exists.17
• What are the implications of recognising a delictual remedy for breach of the statutory duty?
Would imposition of liability for damages have a ‘chilling effect’ on performance of
administrative or statutory function? For example, would potential liability hamper a statutory
body in carrying out its statutory duties, or the police in carrying out their normal duties? 18

The question of whether causing harm through breach of a statutory duty is wrongful for the
purposes of delictual liability involves a process of statutory interpretation, according to the normal
rules of interpretation. It also involves a value or policy judgement on wrongfulness based on
factual and policy considerations as set out above.

1 Roodepoort-Maraisburg Town Council v Eastern Properties (Prop) Ltd 1933 AD 87 at 101; Bophuthatswana
Transport Holdings (Edms) Bpk v Matthysen Busvervoer (Edms) Bpk 1996 (2) SA 166 (A) at 173.
2 Knop v Johannesburg City Council 1995 (2) SA 1 (A) at 28.
3 South African Hang and Paragliding Association v Bewick 2015 (3) SA 449 (SCA) para 23; contra Neethling and
Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 78.
4 2015 (3) SA 449 (SCA) para 23.
5 Alves v LOM Business Solutions (Pty) Ltd 2012 (1) SA 399 (GSJ) para 29; Naidoo v Minister of Police [2015] 4 All
SA 609 (SCA) paras 20–25.
6 1995 (2) SA 1 (A) at 28.
7 2001 (3) SA 1247 (SCA).
8 See also Alves v LOM Business Solutions (Pty) Ltd 2012 (1) SA 399 (GSJ) para 28.
9 2007 (3) SA 121 (CC).
10 Para 42 (footnotes omitted).
11 2003 (6) SA 13 (SCA).
12 1907 TS 427.
13 1997 (4) SA 578 (W) at 583C–D.
14 At 583 (references omitted).
15 1971 (3) SA 123 (A).
16 1995 (2) SA 1 (A) at 28.
17 See also Olitzki Property Holdings v State Tender Board 2001 (3) SA 1247 (SCA); Steenkamp NO v The Provincial
Tender Board of the Eastern Cape 2007 (3) SA 121 (CC); South African Hang and Paragliding Association v Bewick
2015 (3) SA 449 (SCA).
18 Minister of Law and Order v Kadir 1995 (1) SA 303 (A) at 319.
 Chapter 21

Public authorities

21.1 Introduction

21.2 Widening liability

21.3 General trends in South Africa: Effect of the Constitution

21.4 State Liability Act 20 of 1957

21.4.1 Section 1
21.4.2 Section 2
21.4.3 Section 3
21.4.4 Section 4 |

21.5 Section 35 of the General Law Amendment Act 62 of 1955

21.6 State liability for public schools

21.7 Local authorities

21.1 Introduction
Delictual liability of public authorities is based on the ordinary general principles of delict. This
chapter aims to:
• Indicate the international trend of widening liability of public authorities
• Indicate the effect of the Constitution on widening this form of liability in South Africa
• Refer to legislation governing State liability1
• Refer to legislation governing liability of the State for public schools
• Indicate briefly the application of certain general principles to liability of local authorities.
21.2 Widening liability
The liability in delict of the State or government, and of public authorities generally, has been
extended in South Africa during the last part of the twentieth century and into the twenty-first
century. One can attribute this development to both external and internal factors.
Worldwide, the relationship between State administrations and their citizens changed during
the twentieth century. The range of public services has widened considerably, and the culture has
also changed, in that members of the public are seen much more as active ‘clients’ or ‘customers’
whose needs have to be met.2 The disappointed ‘customer’ of a public service feels justified in
complaining if the service provided or procured by the State on his or her behalf is deficient, and
readily claims compensation for injury or loss. The liability of the State or a public authority
requires that there is a balance between protecting the interests of the citizen and preserving the
ability of the administration to function in the public interest. Generally, the trend has been towards
what Markesinis has called a ‘consumerist vision of public liability’, under which ‘compensating
the damages suffered by citizens because of administrative activities can never be a wrong use of
public money’. 3

21.3 General trends in South Africa: Effect of the

Constitution
In South Africa, the Constitution of the Republic of South Africa, 1996, with its extensive Bill of
Rights, has supported the trend towards expanding the liability of public authorities. This has
opened the way to a wider interpretation of wrongfulness in the law of delict. Criminality and the
absence of a victim-compensation scheme are among the factors that have led to a number of cases
in which victims of crime have successfully claimed damages from the State, based on the
wrongful failure of the police to provide protection.4 In line with the trend towards widening
liability, the Constitutional Court held in Minister of Defence and Military Veterans v Thomas 5 that
a provincial government employee who had been seconded to work for a national government
department was entitled to claim statutory compensation from the provincial government as her
employer 6 and also to claim delictual damages against the national government.7 The
Constitutional Court rejected the notion of the State as a single employer 8 and stressed the
importance of vindicating the employee’s fundamental right to freedom and security of the person.9
However, the courts have also been concerned not to cast the net of liability too wide. In
Minister of Safety and Security v Scott 10 the Court held that holding the Minister for Safety and
Security responsible for the patrimonial losses suffered by the second respondent would open the
door to indeterminate police liability, and for this reason wrongfulness had not been established. As
regards causation, the Court was concerned not to ‘cast the net too wide and to land the police with
liability for loss that is too remote’. 11 Accordingly, the State was not held delictually liable for the
second respondent’s patrimonial loss. In Minister of Safety and Security v Janse Van der Walt 12 the
Court held that the State could not be held vicariously liable for the negligent conduct of a
magistrate while performing his or her judicial functions.13 In MEC for the Department of Public
Works, Roads and Transport v Botha 14 the Court held that no evidence had been placed before the
Court to enable it to find that a general legal duty rested on the State to maintain a public road by
removing trees that constantly grow and cause a potential danger to the road’s users.15
The Constitution guarantees just administrative action,16 which is given detailed content in
legislation.17 This has widened the scope for liability of public bodies based on negligent failure of
administrative justice. The Constitutional Court has affirmed that ‘(i)n our constitutional
dispensation, every failure of administrative justice amounts to a breach of a constitutional duty’,
which might, but will not necessarily, give rise to delictual liability.18 This has manifested itself in a
number of cases that deal with irregular tender procedures.19

21.4 State Liability Act 20 of 1957

The State Liability Act 20 of 1957 restates the abolition of immunity of the State against liability
for the acts of its employees performed under statutory authority. The result of this is that, within
the limits imposed by the Act, the liability of the State is co-extensive with that of the individual
citizen.20
The Act does not define the concept of ‘State’, and so one should refer to the definitions of
‘organ of state’ in section 239 of the Constitution, and in section 1(b) of the Institution of Legal
Proceedings against Certain Organs of State Act 40 of 2002.

21.4.1 Section 1
Section 1 of the State Liability Act confirms that a person can bring an action against the State in
any competent court, ‘whether the claim arises out of any contract lawfully entered into on behalf
of the State or out of any wrong committed by any servant of the State acting in his capacity and
within the scope of his authority as such servant’. In terms of this section, the State is liable for a
‘wrong’ committed by an employee or ‘servant’ acting ‘in his capacity and within the scope of his
authority as such servant’. 21 To an extent, State liability is qualified ‘by express mention of
‘‘contract’’ and ‘‘wrong’’, [but] it is settled law that these are not intended to be the sole grounds of
liability’. 22 State liability is not confined to breaches of contracts entered into by the State or delicts
committed by employees of the State acting in the course and scope of their employment. Courts
will not decline redressing a violation of individual rights that results from an administrative act
simply because the act is performed ‘in the course of implementing a general policy’. 23 In effect,
the State is vicariously liable for the acts of its employees on the same basis as an ordinary
employer.24 The law imposes liability on the State not only for acts committed by its employees, but
also in cases of strict liability for damage done by animals (pauperies).25
In some earlier cases, courts held that the State is not liable where the statutory duty imposed
on the employee allows the employee to exercise his or her personal discretion independently of the
State.26 The State must then prove that the employee was exercising his or her personal discretion.27
However, such instances, if they still occur at all, would be exceptional.28
In Mhlongo v Minister of Police 29 the Court summarised the liability for the conduct of the
police in terms of the State Liability Act as follows:
The liability of the State for such a wrong is founded on the provisions of s1 of the Act …
The remedial legislation, of which Act 20 of 1957 is the modern offspring, was intended as
the sole foundation of State liability in our law. The particular provisions of s1 of the Act
relevant to the liability of the State for the delict of a policeman are those to be found in the
words ‘… any wrong committed by any servant of the State acting in his capacity and within
the scope of his authority as such servant ...’ the view that all members of the police force
are prima facie servants of the State is too well entrenched in decisions of this court of the
Appellate Division. Although [s1 of Act 20 of 1957] speaks only of a State servant acting
‘within the scope of his authority’, the Courts appear to have treated this as embracing the
concept ‘within the scope of his employment …’ It has never been suggested that the State
escapes liability for a wrongful act committed by a servant in his capacity as such simply
because the act fell outside the ‘scope of his authority’, when it was clearly within the ‘scope
of his employment’. All members of the South African Police Force are prima facie servants
of the State and consequently, when a wrongful act is committed by a member of the Force
in the course or scope of his employment, the State is prima facie liable. It is then for the
 State to show that in committing the wrongful act, the policeman was engaged upon a duty
or function of such a nature as to take him out of the category of servant pro hac vice. In
order for the duty or function to take him out of the category of servant it must be one
which is personal to the policeman in the sense that from its very nature [it is divorced from
the State].

Therefore, to escape liability for harm caused by a public servant, the State must show that the
servant was engaged in a personal activity that is divorced from the duties of a State employee.30

21.4.2 Section 2
Section 2 of the State Liability Act is a procedural measure. It provides that one should cite the
political head of a state or provincial department in proceedings.

21.4.3 Section 3
Section 3 of the Act prohibits execution, attachment or like process for enforcing judgments against
the State.31 Courts have held that section 3 is no bar to obtaining an Anton Piller order against the
State, in a case where the order was aimed at the search of a police facility to look for torture
apparatus allegedly kept there.32 The Constitutional Court held in Nyathi v The MEC, Department of
Health, Gauteng 33 that this section was in conflict with the Constitution, because it unjustifiably
limits the right to equal protection of the law contained in section 9(1) of the Constitution and was
inconsistent with the constitutional protection of dignity and the right of access to courts. The Court
also held that section 3 violated the principles of judicial authority, as well as the principle that the
public administration be accountable. Accordingly, the Court confirmed a declaration of
constitutional invalidity, but suspended the order for 12 months to allow parliament to pass
legislation that provides for an effective means of enforcement of money judgments against the
State. This was followed by the passing of the State Liability Amendment Act 14 of 2011, which
allows the issue of a writ of execution or a warrant of execution against movable property owned
by the State and used by the department concerned, but only after various notification procedures
have been followed.

21.4.4 Section 4
Section 4 of the Act provides that limitations of liability, prescribed periods and conditions for
claims against the State that are contained in other legislation are not affected by the Act. The
application of special notice and limitation periods was an important feature in actions against the
State and certain State organs prior to the introduction of a new Constitution in South Africa in the
last decade of the twentieth century. Once the interim Constitution and the final Constitution had
come into operation, the constitutionality of these short and onerous notice and limitation periods
was challenged, in terms of the constitutional guarantees of equality and access to courts.34 The
Institution of Legal Proceedings against Certain Organs of State Act 40 of 2002 subsequently
reformed the law on notice and limitation periods for actions against the State and organs of State.

COUNTER Should the courts impose direct or vicarious liability on the State?
POINT While claims in delict against the State are brought on the basis of
vicarious liability in terms of the State Liability Act, some judges and
academic commentators have argued that claims in delict against the
State should be brought on the basis of direct State liability.35
In terms of this approach, the plaintiff would proceed on the basis that
the employees or servants of the State who engaged in wrongful,
 culpable conduct that resulted in the plaintiff suffering harm were
extensions of the State itself, making the State directly liable for the
harm. This is not merely a semantic or technical distinction: if plaintiffs
take the direct-liability approach, they will not have to prove that the State
employee was acting in the course and scope of employment when the
State employee engaged in the conduct that resulted in the harm. The
Court will be entitled to take a broader view of the matter and decide
whether the State should in the circumstances be held liable in delict for
the harm suffered by the plaintiff (the wrongfulness enquiry), as well as
whether the State, when considering all the facts and circumstances of
the matter, was negligent in failing to foresee and prevent the harm.
The fault enquiry often proves conceptually problematic in claims
brought in delict against the State, as in order to be successful in a claim
brought on the basis of vicarious liability, the plaintiff first needs to prove
that the employee himself or herself personally committed a delict (which
involves a finding that the employee acted negligently in the
circumstances), and then to prove that the delict was committed in the
course of the employee’s employment. As regards the fault enquiry, it is
sometimes the case that the employee in question did not personally act
negligently, but overall the State was negligent in failing to put in place
appropriate systems and safeguards to prevent the harm. For example,
in Minister of Safety and Security v Hamilton 36 the police officers
processing a firearms licence application arguably performed their
employment duties by ensuring that the applicant applied for a firearm in
the prescribed manner, and then processing it in accordance with their
employer’s instructions. They themselves were not personally negligent
in issuing the psychologically unstable woman the firearm, but the State
arguably was for not putting in place systemic measures to ensure that
psychologically unfit persons were not issued with firearms.
In the circumstances, it would seem more logical and appropriate to
impose liability directly on the State rather than through trying to establish
vicarious liability on the basis of the personal negligence of the
employees in question. In addition, it has been argued that the vicarious-
liability enquiry gives rise to complications and confusions when it is
applied in the context of delictual claims against the State, in which the
Court is required to integrate constitutional rights and obligations into the
private-law schema of the vicarious-liability enquiry.37

21.5 Section 35 of the General Law Amendment Act 62 of

1955
Section 35 of the General Law Amendment Act 62 of 1955 is a procedural provision. It sets an
obligatory notice period of 72 hours for obtaining a rule nisi as an interim interdict against the
State. The Act does not define the concept of ‘State’, and one should seek assistance from the
definition of ‘organ of state’ in section 239 of the Constitution. This provision sets procedural
conditions for interdicts against the State, but not for an Anton Piller (search) order against the
State.38

21.6 State liability for public schools

The State is liable for harm or loss for which a public school would otherwise be liable in terms of
the provisions of section 60(1) of the South African Schools Act 84 of 1996. Under the heading
‘Liability of State’, this section provides:
The State is liable for any damage or loss caused as a result of any act or omission in
connection with any educational activity conducted by a public school and for which such
public school would have been liable but for the provisions of this section.

MEC for Education, Western Cape Province v Strauss39

The plaintiff, a public-school teacher, had been struck on the head by a discus while coaching
discus-throwing on the school’s athletics field. She suffered extensive injuries and was
permanently disabled. She sought to hold the defendant liable under section 60(1) of the
South African Schools Act for the school’s failure to enclose the discus circle with safety nets.
The plaintiff had been employed by the governing body of the school in terms of section 20(4)
of the Act.
The defendant raised the special plea that section 60(1) of the Act did not apply, because
an employer’s liability is generally excluded by section 35(1) of the Compensation for
Occupational Injuries and Diseases Act 130 of 1993 (COIDA). The Court upheld the plea. It
held that COIDA provides compensation for employees and section 35(1) expressly excludes
the employer’s liability for damages in respect of any occupational injury or disease that
results in disablement or death. Since the school as the employer was not liable to the
plaintiff in terms of section 35(1), the Court could not attribute liability to the State in terms of
section 60(1).

Section 60(1) of the Act has been described as an ‘umbrella provision’. It establishes general State
liability for harm or loss that arises from any educational activity undertaken by a public school, on
the basis that education is primarily the responsibility of the State.40 In Technofin Leasing &
Finance (Pty) Ltd v Framesby High School 41 Pickering J said that section 60(1) ‘is couched in the
broadest of terms and the State’s liability is expressed in the most general language’. There is
nothing in the section to indicate that it is restricted to delictual liability.

21.7 Local authorities

For some purposes, local authorities are regarded as organs of State. In terms of section 239 of the
Constitution, an ‘organ of state’ includes ‘(a) any department of state or administration in the
national, provincial or local sphere of government’. Also, in terms of section 1(1)(vii) (b) of the
Institution of Legal Proceedings against Certain Organs of State Act 40 of 2002, an ‘organ of state’
includes ‘a municipality contemplated in section 151 of the Constitution’. However, courts have
held that a municipality is not included in the word ‘State’ for the purposes of immunity against
attaching and executing the State in terms of section 3 of the State Liability Act.42
The delictual liability of a local authority is governed by the same principles as the liability of
an individual. However, in determining wrongfulness where local authorities are involved, courts
are guided by policy considerations that are concerned with, for example, the need for
accountability of public bodies or officials, the proportionality of the risk of harm and the cost of
prevention, control over a dangerous object or situation, awareness of danger, prior conduct
creating danger, and a relationship imposing responsibility.
Courts discarded the view that liability of public authorities for an omission can only be
imposed where the defendant’s prior conduct had positively created a risk of harm or a new source
of danger. This had the effect of lowering the threshold for considering the failure of a local
authority to prevent harm as wrongful and widening the liability of local authorities for omissions.
Courts finally discarded the ‘prior conduct’ approach to liability of local authorities for omissions
in Cape Town Municipality v Bakkerud.43 The case dealt with liability of the Cape Town
Municipality for the injuries of a woman who had stepped into a pothole in a pavement. The Court
took into account the area where the injury occurred, the period of time for which the dangerous
situation had existed, and the resources needed to maintain the pavement in a safe condition. These
factors were relevant to both the enquiry into wrongfulness and negligence. Based on the facts, the
Court came to the conclusion that there was a legal duty on the part of the municipality to prevent
the harm (failure to prevent the harm was wrongful), there was negligence on the part of the
municipality, and there was contributory negligence on the part of the plaintiff for failing to be
reasonably observant when walking on the pavement.
To determine the negligence of local authorities (acting through officials or employees), courts
apply the ordinary principles regarding reasonable foreseeability and preventability of harm. They
take into account factors such as the extent of the risk, the opportunity and efforts to become aware
of risks, and the cost of prevention. This is illustrated in Mostert v Cape Town City Council,44 a case
that involved a damaging leak from the city council’s water main. Weighing the technical evidence
of the likelihood that the pipe would burst against the high cost of replacement, the Court decided
that the Council could not have been reasonably expected to replace the pipe, and held that the
Council had therefore not been negligent.

1 The State Liability Act 20 of 1957 and section 35 of the General Law Amendment Act 62 of 1955.
2 Fairgrieve, Andenas and Bell (Eds) Tort Liability of Public Authorities in Comparative Perspective (2002) at xvii.
3 Fairgrieve, Andenas and Bell (2002) at xviii.
4 Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies Intervening) 2001 (4) SA 938 (CC);
Van Eeden v Minister of Safety and Security 2001 (4) SA 646 (T); Minister of Safety and Security v Van Duivenboden
2002 (6) SA 431 (SCA); Alves v LOM Business Solutions (Pty) Ltd 2012 (1) SA 399 (GSJ) paras 20–21; Minister of
Justice and Constitutional Development v X [2014] 4 All SA 586 (SCA); 2015 (1) SA 25 (SCA) paras 13–18 and 33
–34; Naidoo v Minister of Police [2015] 4 All SA 609 (SCA) para 33; Dlanjwa v Minister of Safety and Security 2015
JDR 2094 (SCA) paras 24 and 25; Oppelt v Department of Health, Western Cape 2016 (1) SA 325 (CC) paras 51–68;
Mashongwa v Passenger Rail Agency of South Africa 2016 (3) SA 528 (CC) paras 16–20, 22, 24–27 and 29; South
African Hang and Paragliding Association v Bewick 2015 (3) SA 449 (SCA) paras 23–25.
5 2016 (1) SA 103 (CC).
6 Paras 5 and 23–25. The claim would be in terms of the Compensation for Occupational Injuries and Diseases Act 130
of 1993.
7 Para 39.
8 Para 18.
9 Paras 38–39.
10 2014 (6) SA 1 (SCA).
11 Para 40.
12 2014 JDR 2454 (SCA).
13 Para 23.
14 2016 JDR 0514 (SCA).
15 Paras 8–13.
16 Section 33.
17 Promotion of Administrative Justice Act 3 of 2000.
18 Steenkamp NO v The Provincial Tender Board of the Eastern Cape 2007 (3) SA 121 (CC) para 37.
19 Steenkamp NO v The Provincial Tender Board of the Eastern Cape 2007 (3) SA 121 (CC); Olitzki Property Holdings
v State Tender Board 2001 (3) SA 1247 (SCA); Minister of Finance v Gore NO 2007 (1) SA 111 (SCA).
20 East London Western Districts Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA 63
(A) at 69–70.
21 Interpreted in Mhlongo v Minister of Police 1978 (2) SA 551 (A) at 566ff; Minister of Police v Rabie 1986 (1) SA 117
(A) at 132; Tshabalala v Lekoa City Council 1992 (3) SA 21 (A) at 31ff; Minister of Law and Order v Kadir 1995 (1)
SA 303 (A).
22 East London Western Districts Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA 63
(A) at 69–70.
23 East London Western Districts Farmers’ Association v Minister of Education and Development Aid 1989 (2) SA 63
(A) at 69–70.
24 Mhlongo v Minister of Police 1978 (2) SA 551 (A) at 567; Minister van Polisie v Gamble 1979 (4) SA 759 (A) at 765;
Minister of Police v Rabie 1986 (1) SA 117 (A) at 132.
25 South African Railways and Harbours v Edwards 1930 AD 3 at 9. See Chapter 32 on strict liability.
26 Sibiya v Swart NO 1950 (4) SA 515 (A) at 520ff; Mazeka v Minister of Justice 1956 (1) SA 312 (A) at 317; Sizani v
Minister of Police 1980 (3) SA 1205 (SEC) at 1210. It has been held, however, that a policeman who arrests another is
carrying out a statutory duty and not a ‘personal duty’: Naidoo v Minister van Polisie 1976 (4) SA 954 (T) at 957;
Mhlongo v Minister of Police 1978 (2) SA 551 (A) at 567.
27 Sibiya v Swart NO 1950 (4) SA 515 (A); Minister of Police v Mbilini 1983 (3) SA 705 (A) at 711.
28 Minister of Police v Rabie 1986 (1) SA 117 (A) at 132; Mtetwa v Minister of Health 1989 (3) SA 600 (D) at 606.
29 1978 (2) SA 551 (A) at 567.
30 See also Chapter 33 on vicarious liability. In Minister of Defence v Von Benecke 2013 (2) SA 361 (SCA) the Supreme
Court of Appeal held that the State employee in question had not entirely dissociated himself from the business of his
employer, justifying the imposition of vicarious liability. This case illustrates how constitutional norms that have been
incorporated into the enquiry into vicarious liability have widened the potential for the imposition of this form of
liability on the State (paras 13–14).
31 Jayiya v Member of the Executive Council for Welfare, Eastern Cape 2004 (2) SA 611 (SCA) paras 15–16; Minister
of Health v Brückner [2007] JOL 18898 (LAC).
32 Shoba v Officer Commanding, Temporary Police Camp, Wagendrift Dam; Maphanga v Officer Commanding, South
African Police Murder and Robbery Unit, Pietermaritzburg 1995 (4) SA 1 (A) at 20C–E.
33 2008 (5) SA 94 (CC).
34 Mohlomi v Minister of Defence 1997 (1) SA 124 (CC); Hans v Minister van Wet en Orde 1995 (12) BCLR 1693 (C).
35 See the separate concurring minority judgment of Froneman J in F v Minister of Safety and Security 2012 (1) SA 536
(CC), in which the judge makes the case for imposing delictual liability directly on the State. Also see Boonzaier
‘State Liability in South Africa: A More Direct Approach’ (2013) 130(2) SALJ at 330–368. The direct approach to
state liability is endorsed by Botha and Millard ‘The Past, Present and Future of Vicarious Liability in South
Africa’ (2012) 45(2) De Jure at 225 as a valid alternative to a claim in delict brought against the State on the basis of
vicarious liability. However, Neethling and Potgieter ‘Deliktuele staatsaanspreeklikheid weens
polisieverkragting’ (2012) 9(2) LitNet Akademies at 73–99 are of the view that it is questionable whether a radical
deviation from the traditional vicarious-liability approach is justified, although they accept that this course could be
adopted.
36 2004 (2) SA 216 (SCA).
37 See Fagan ‘The Confusions of K’ (2009) 126(1) SALJ at 156–205 and Wagener ‘K v Minister of Safety and Security
and the Increasingly Blurred Line between Personal and Vicarious Liability’ (2008) 125(4) SALJ at 673–680. Also see
the arguments made in this regard in the judgment of Froneman J in F v Minister of Safety and Security 2012 (1) SA
536 (CC) paras 100–108. In the context of the enquiry into factual causation, it has been argued that the law of delict,
having been designed to allow private litigants to recover compensatory damages from other private litigants, is
inherently unsuited for claims against the State involving alleged infringements of fundamental rights. Such rights
should be vindicated by means of a claim brought directly against the State for constitutional damages. In such an
action, the litigants would not be circumscribed by the private-law ethos and principles of the law of delict, and the
Court would be able to hand down a damages award to vindicate the rights at stake and deter further infringements of
them. See Price ‘Factual Causation after Lee’ (2014) 131(3) SALJ at 491–500.
38 Shoba v Officer Commanding, Temporary Police Camp, Wagendrift Dam; Maphanga v Officer Commanding, South
African Police Murder and Robbery Unit, Pietermaritzburg 1995 (4) SA 1 (A) at 20C–E. 39 2008 (2) SA 366 (SCA);
39 [2008] 1 All SA 440 (SCA).
40 Louw v LUR vir Onderwys en Kultuur, Vrystaat 2005 (6) SA 78 (O) para 13.
41 2005 (6) SA 87 (SE) at 92I–93C.
42 See Mateis v Ngwathe Plaaslike Munisipaliteit 2003 (4) SA 361 (SCA).
43 2000 (3) SA 1049 (SCA).
44 2001 (1) SA 105 (SCA).
 Chapter 22

Professional liability

22.1 Introduction

22.2 Medical practitioners

22.3 Legal practitioners

22.3.1 Attorneys
22.3.2 Advocates

22.4 Engineers, builders and architects

22.5 |
Accountants and auditors

22.6 Banks

22.1 Introduction
Delictual liability of professionals is based on the ordinary general principles of delict. This chapter
illustrates the application of certain general principles to liability of certain professionals.
A person who engages in a profession, trade, calling, or any other activity that demands special
knowledge and skill must not only exercise reasonable care, but must also measure up to the
standard of competence that is reasonable to expect from a person professing such knowledge and
skill. Therefore, the general test for negligence is adapted to accommodate situations in which
special knowledge or skill is required. In assessing liability of a professional or expert, courts have
to consider the general level of knowledge and skill of members of the relevant branch of the
profession or area of expertise.
The relationship between a professional person and his or her client is primarily a contractual
one, which gives rise to potential liability for breach of contract. However, some concurrence of
contractual and delictual liability can occur within this relationship. Courts accept that the
contractual relationship between a professional person and his or her client, in addition to a duty to
act with reasonable care and skill as an implied term of the contract, can under certain
circumstances also give rise to a delictual duty with a similar content. Where the performance of
professional services constitutes either a breach of contract or a delict, the plaintiff can choose
whether to proceed either in delict or in contract against the same defendant.
The relationship between a professional person and a client can also give rise to an exclusively
delictual action, for example, where one party in the pre-contractual phase negligently or
intentionally misrepresented facts to the other, to induce entry into the contract.1 Also, a
professional person or expert may, under certain circumstances, incur delictual liability towards
third parties in cases where misrepresentation, negligent advice, or defective workmanship affects
third parties, for example, in the ‘disappointed beneficiary’ cases, where a beneficiary does not
benefit from a will, because a professional advisor was negligent in drafting or executing the will.2
Both in contract and in delict, the liability of professional persons is based on the expectation
of a reasonable degree of specialised knowledge, competence and skill. This standard is applied to:
• Members of established and organised professions, such as medical doctors,3 legal
practitioners,4 and accountants-auditors5
• Other specialised occupations such as bankers,6 investment advisors,7 and sworn appraisers of
land8
• Technicians, such as electricians 9
• Others with a special competence for performing certain tasks or handling certain equipment,
such as a petrol station attendants,10 farmers using special equipment,11 persons engaged in
burning fire-belts,12 and drivers of motor vehicles.13
It is negligent to engage voluntarily in any potentially dangerous activity unless a person has the
skill and knowledge usually associated with the proper discharge of the duties connected with such
an activity. In Durr v ABSA Bank Ltd 14 the Court said the following in this regard:
It is not negligent not to be a lawyer. But those who undertake to advise clients
on matters including an important legal component do so at their peril if they
have not informed themselves sufficiently on the law.

The test for negligence in cases where professional knowledge or expertise is involved has two
components: the possession of the necessary knowledge or skill, and the exercise of necessary care
and diligence. One replaces the ordinary reasonable person, or diligens paterfamilias, with the
reasonable expert and, when assessing the attributes required, a court will have regard to the
general level of skill and diligence possessed and exercised at the time by members of the branch of
the profession to which the practitioner belongs.15 For example, in the medical profession courts
will determine the level of knowledge and skill required not only by the type of practice (general
practitioner or specialist), but also by the nature of the duties the practitioner is carrying out at the
time (specialist duties or general duties, for example, as the casualty officer on duty at a hospital).
An error of judgement may indicate negligence, but some errors of judgement are reasonable, for
example, in emergencies or in situations fraught with risk, such as a complicated medical
procedure. However, an error of judgement caused by a lack of the required degree of knowledge,
skill and diligence will constitute negligence.16
The test for negligence requires that one assesses the defendant’s conduct in the particular
circumstances prevailing at the time. In that sense, locality considerations are also relevant to the
enquiry into professional negligence. However, as far as the level of knowledge and skill expected
of a professional is concerned, it is not likely that a lower level of competence will be required of a
professional practising in a particular locality, for instance a rural area. In Van Wyk v Lewis 17 Innes
CJ applied a general standard when assessing the competence of a surgeon, without considering the
circumstances or practices in the particular locality. Wessels JA, in the same case, indicated that the
level of skill and care to be expected could differ according to locality.18 Courts are bound to take
into account the surrounding circumstances (such as lack of facilities) when assessing care and
diligence, but are unlikely to adapt the level of competence expected of a professional according to
locality.
As indicated in the chapter on fault, one determines the standard of care required by
objectively evaluating whether the conduct complied with normal and generally accepted practice.
Generally accepted practice in a particular profession or area of expertise indicates the nature and
extent of care and precautions in particular situations.19 However, the generally accepted practice is
not necessarily conclusive. The overriding standard in every instance is that of a reasonable person
in the particular situation.

22.2 Medical practitioners

A medical practitioner engages in a profession that demands special knowledge, skill and care, and
should measure up to the standard of competence that is reasonably expected of a person of that
profession.20 The relationship between a medical practitioner and patient is usually, but not
necessarily, contractual; for example, where medical services are rendered to an unconscious
person. The relationship involves the duty to act with reasonable care and skill, as an implied term
of the contract, or as a duty imposed by the law of delict. Courts recognise concurrence of contract
and delict in respect of the relationship between doctor and patient in cases where the harm was
physical and not purely economic.21
In Van Wyk v Lewis 22 Wessels JA said the following on the standard of competence and care
expected of a medical practitioner (a surgeon):
… the surgeon (must perform) the operation with such technical skill as the average medical
practitioner in South Africa possesses and (must) apply that skill with reasonable care and
judgment … (he) is not expected to bring to bear on a case entrusted to him the highest
possible professional skill but is bound to employ reasonable skill and care and is liable for
the consequences if he does not.

Courts also consider the branch of the medical profession to which the practitioner belongs. They
will not only determine the level of knowledge and skill required by the type of practice (general
practitioner or specialist), but also by the nature of the duties the practitioner is carrying out at the
time. A general practitioner may be required to carry out specialist duties, for example in radiology,
when acting as the casualty officer on duty at a hospital. Courts will then assess competence at the
level of a general practitioner, but this should include the knowledge and judgement of when it is
necessary to call upon the services of a specialist.
In this regard, Bekker J said the following in Esterhuizen v Administrator, Transvaal: 23
… a general practitioner is not expected to have the same degree of knowledge
and skill and experience as a specialist has … . The question is what is the
common knowledge in the branch of the profession to which the accused
belongs.

The standard that the Court applied in Buls v Tsatsarolakis 24 was that of an average general
practitioner, carrying out his duties as a casualty officer in a public hospital, and not that of a
specialist orthopaedic surgeon.
Liability for medical practitioners may arise not only from clinical work, such as prescribing
medicine or performing a procedure, but also from failure to inform patients properly about risks of
medicine or medical procedures. Patients must be informed of the material aspects of the risk of
harm.25 In Castell v De Greef 26 the Court comprehensively reviewed the South African law in this
regard and held that, for a patient’s consent to constitute a defence, the doctor is obliged to warn a
patient of a material risk inherent in the proposed treatment, and that a risk is material if, in the
circumstances of the particular case:
• A reasonable person in the patient’s position, if warned of the risk, would be likely to attach
significance to it
• The medical practitioner is or should reasonably be aware that the particular patient, if warned
of the risk, would be likely to attach significance to it.27

If a medical procedure involves the risk of harmful side effects, even if performed competently, the
doctor must inform the patient of the material aspects of the risk of harm. Questions of
wrongfulness, causation and negligence arise in this regard. If, for example, a surgeon operates
without first fully informing the patient of the risks, the operation is performed without the patient’s
informed consent. If harmful side effects occur, and the patient would not have undergone the
operation if informed of the risk, the surgeon will have caused the harm wrongfully. If failing to
inform the patient was also negligent in terms of the standard of a reasonable practitioner in the
relevant branch of the medical profession, the practitioner will be liable. If, on the other hand, the
operation is completely successful and no harmful side effects occur, there is arguably no wrongful
causing of harm, despite the fact that the operation was performed without the patient’s informed
consent.

PAUSE FOR Consent

REFLECTION In Broude v McIntosh28 Marais JA referred to the situation where an
operation is performed without proper consent, but the operation turns
out to be completely successful. He expressed the view that there is no
principle of law by which objectively characterising, as wrongful or lawful,
an intentional act that involves causing bodily harm to another, could be
postponed until its consequences are known. Is this correct? In the event
of a completely successful operation, arguably there is no harm for the
purposes of the law of delict, despite the fact that the patient might have
chosen not to undergo the operation with knowledge of the risks. In the
law of delict, courts decide wrongfulness based on the causal sequence,
not on the act by itself. They can only do this once the consequences of
the act are known.

In cases of harm that arise from childbirth, the medical practitioner can be liable towards the
parents in cases of unwanted pregnancy 29 or failure to warn of the risk of a child being born
handicapped.30 However, a handicapped child does not have an action against a medical practitioner
for negligently failing to warn the parents that their child may be born disabled.
Stewart v Botha31
The Supreme Court of Appeal declined to recognise a ‘wrongful life’ claim by a child born
disabled. The child alleged that his mother, had she been properly advised, would have
chosen to have an abortion and that she would have been able to do so in terms of the
governing legislation. The Court refused the action on the basis that the child’s disability was
congenital and not in any way caused by the conduct of the medical practitioners. The
practitioners had merely failed to warn the mother of the risk of the child being born disabled.
From the child’s point of view, the failure to warn his mother caused him to be born, and the
Court refused to regard the causing of a life, as opposed to no life at all, as a wrongful
causing of harm.
 H v Fetal Assessment Centre32
The plaintiff instituted a claim on behalf of her minor child who suffered from Down’s
syndrome. She alleged that the defendant had acted negligently by failing to inform her that
there was a high risk of her child being born with Down’s syndrome. She alleged that, if she
had known of this risk, she would have had an abortion. The High Court disallowed the claim,
holding that, while South African law recognises a claim by a mother for damages for harm
caused by a pre-natal misdiagnosis, it does not recognise such a claim brought by the child.
The Constitutional Court unanimously decided that the existing position, which deprives a
child of the right to claim damages for a pre-natal misdiagnosis, may not be in the child’s best
interests and found that, in theory, such a claim may exist. The Court referred the matter
back to the High Court, since a decision on the validity of the child’s claim might involve
complex factual and legal considerations which the High Court was best-placed to assess.
The High Court would have to decide whether the elements of a delict had been satisfied, or
whether a new remedy should be developed to allow a child to recover compensation for any
harm he or she may have suffered.

Mukheiber v Raath33
The parents of a healthy and normal child (their fourth) instituted action in delict against a
gynaecologist. They alleged that he had negligently misrepresented to them that the wife had
been sterilised after the birth of their third child. Relying on this representation, they had failed
to take contraceptive measures. This resulted in their fourth child being conceived and born.
The parents claimed damages from the doctor for pure economic loss, in the form of
confinement costs and maintenance of the child until he became self-supporting. The
Supreme Court of Appeal held that the doctor had a legal duty not to make any
representations on the matter of sterilisation until he had taken reasonable steps to make
sure the representations were accurate. The factual and policy considerations that indicated
that such a duty existed were:
• The special relationship between the doctor and the parents who consulted him
• The material risk that the situation involved, that is, the risk of the conception and birth of an
unwanted child
• The fact that this risk should have been obvious to the doctor
• The fact that it should also have been obvious to the doctor that the parents would rely on
what he told them, and that the correctness of the representation was of vital importance to
them, as they could suffer serious damage if the representation was incorrect
• The fact that the representation related to technical matters that concerned a surgical
procedure about which the parents would necessarily be ignorant and the doctor should be
knowledgeable.

Oppelt v Department of Health, Western Cape34

The plaintiff suffered a serious spinal cord injury in a rugby match. He was taken to a
government hospital for treatment. The medical professionals failed to provide him with
timeous appropriate medical treatment (a ‘closed reduction’) and the delay meant that he
became paralysed from the neck down.
The plaintiff sued the provincial government for damages in delict. The trial court allowed
his claim, but this decision was overturned by the Supreme Court of Appeal. In a further
appeal, the Constitutional Court found that the medical professionals had been negligent and
that the plaintiff’s harm had been caused wrongfully. Regarding wrongfulness, the Court held
that there was a legal duty in the circumstances for the medical professionals to provide the
plaintiff with timeous medical treatment, in part because the Constitution requires that
emergency medical treatment be provided to those who require it.35 The defendant could not
justify the delay through citing a need for it to adhere to hospital protocols regarding the
transfer of patients to other health-care facilities.36

Medical practitioners can be held liable not only for physical harm to patients, but also for
economic loss. When deciding on liability, courts take into account all the circumstances, including
the nature of the work and the nature of the alleged failing (for example, error of judgement, or lack
of required knowledge or care).

22.3 Legal practitioners

The two branches of the legal profession in South Africa are based on a division of work. Broadly
stated, attorneys deal directly with clients, whereas advocates take instructions from attorneys, who
act on behalf of a client. Advocates have the right to appear in the higher courts. The Right of
Appearance in Courts Act 62 of 1995 has bridged the division between the two branches of the
profession in terms of which attorneys under certain conditions also have the right to appear in the
higher courts in the Republic (including the Supreme Court of Appeal and Constitutional Court).
Generally, the legal profession demands specialised knowledge, skill and care from members of
both branches of the profession. A practitioner must measure up to the standard of competence of a
reasonable person professing such attributes.37

22.3.1 Attorneys
The client-attorney relationship is a contractual one. The relationship imposes the duty, by way of
an implied term, to perform professional services with the knowledge, competence, skill and care
that can reasonably be expected of a member of the profession. Failure to exercise this standard of
care is a breach of this implied term of the contract.38 An attorney will not be guilty of negligence
merely because he or she committed an error of judgement, whether on matters of discretion or
law.39 However, an error of judgement may be indicative of negligence – in particular, where the
practitioner should have realised that the work required a degree of expertise that he or she did not
possess. Also, an error of judgement caused by a lack of the required degree of knowledge, skill
and diligence will constitute negligence.40 Where an attorney does not exercise the required skill
and care, and the client suffers loss, the damages will be those that normally flow from the breach
or may reasonably be supposed to have been contemplated by the parties as likely to result from it.41
Apart from liability for damages that arise in contract, an attorney may also be held liable in
delict. In one case, the Court recognised the concurrence of contract and delict in the attorney-client
relationship in respect of pure economic loss.42 However, in a leading case concerning an attorney’s
liability to his client, the Court treated the client’s action as a contractual action.43
There can also be liability in delict towards a third party outside the contractual relationship –
for example, in a ‘disappointed beneficiary’ case, where a beneficiary does not benefit from a will,
because the attorney was negligent in drafting or executing the will.44 An attorney conducting
litigation on behalf of a client has a duty to act with integrity and care, not only towards his or her
client, but also towards the Court and the opponent. However, courts are likely to consider this duty
a matter of professional ethics rather than a basis for delictual liability.45

PAUSE FOR The widespread increase in the use of the internet and the subsequent
REFLECTION explosion in cybercrime poses a new risk to attorneys. Attorneys must be
aware of the new dangers posed by the internet – ‘phishing’ scams; risks
posed by the use of email, including compromising the duty of
confidentiality to the client; the possibility of altering electronic
documents; and the possibility of receiving mandates from persons other
than the clients. The latter risk is especially encountered in conveyancing
transactions, where fraudsters may attempt to intercept email
communications between the attorney and the client and attempt to divert
money into fraudulent accounts.
 Would the legal convictions of the community dictate that an attorney
who does not exercise proper care when interacting online or via email
be held liable in delict for damages? What are the policy considerations
that would favour or militate against the imposition of liability in such
cases?

Roestoff v Cliffe Dekker Hofmeyr Inc46

Fraudsters managed to access the plaintiff’s bank account by internet ‘phishing’ and stole
R350 000 from it. They then transferred R200 000 of this amount into the defendant
attorneys’ trust account. The defendant firm was not aware that the money had been stolen
and subsequently transferred the money to a third party. The plaintiff sued the firm for loss of
the money, claiming that it, as a firm of attorneys, had a legal duty to ascertain the source of
the money before allowing the money to be paid out to a third party. The Court held that,
although the plaintiff had been unknown to the firm, there was a duty on it not to be negligent
in respect of deposits made into its trust account. This means that attorneys are generally
obliged to verify the source of unidentified trust account deposits before dealing with them.
However, in this instance the firm had not been negligent, as there had been nothing to alert
it to the fact that the deposit had been obtained fraudulently. In addition, the fraudsters had
misled the firm as to the source of the deposit, and the plaintiff himself had been negligent in
failing to heed the bank’s warnings on its internet banking website about the prevalence of
internet fraud.

Steyn NO v Ronald Bobroff & Partners47

The respondent was a firm of attorneys alleged by the appellant to be a ‘pre-eminent
specialist’ in the field of personal injury claims. The appellant contracted the firm to lodge a
claim with the Road Accident Fund in respect of personal injuries suffered by her minor son in
a motor vehicle accident. The appellant alleged that it was a tacit term in the contract of
professional services between them that the respondent would execute the mandate with
reasonable diligence, care and skill; and further, that the firm had been negligent in executing
the mandate, in that it had failed to prosecute the claim timeously, which resulted in the Road
Accident Fund paying out the appellant’s claim some 14-and-a-half months later than it
otherwise would have done. The appellant’s claim was for damages in the form of the interest
that would have been earned on the capital sum over the 14-month period, had it been
invested.
The Court found that the appellant had failed to adduce clear evidence proving that the
respondent had acted negligently and, in particular, evidence of how a reasonable attorney
would have prosecuted the claim in the circumstances.48 Accordingly, her claim for damages
failed. In a separate concurring judgment , Brand JA held that the appellant was not entitled
to mora interest in circumstances where the respondent did not owe her a principal debt.49

22.3.2 Advocates
In South African common law, there are some who hold the view that advocates are not liable for
an error, lack of expertise or mistake of fact or law, particularly where this concerns handling a
case in court.50 This immunity from liability is rooted in the policy consideration that it is not in the
interests of the administration of justice that a court should have to conduct what effectively
amounts to a retrial of a case that has been heard in another court to determine whether the alleged
error or lack of expertise was negligent, and the cause of harm to the client. This also used to be the
approach to liability of barristers in English law, but the position has changed.51
There is no authoritative South African case that deals generally with the liability of advocates
for negligence. However, courts are likely to impose the same standard of knowledge, competence,
skill and care that is expected of other legal professionals, rather than maintain a historical
immunity based on a questionable distinction between handling a case in court and other
professional work.
When deciding the liability of legal practitioners, courts take into account all the
circumstances, including the nature of the work (advisory, or the conduct of litigation), and the
nature of the alleged failing (error of judgement, or lack of required knowledge or care).

22.4 Engineers, builders and architects

Professionals involved in the construction industry, such as engineers, builders and architects, must,
like other professionals, exercise reasonable care in their work. They also have to measure up to the
standard of competence that one can reasonably expect of a person professing such knowledge and
skill. Liability could be based on the contractual relationship with clients or on breach of a duty in
delict to third parties whose legally protected interests have been harmed by the performance of
professional services.
Lillicrap, Wassenaar and Partners v Pilkington Bros (SA) (Pty) Ltd52
This case dealt with an engineer and client relationship in which professional negligence
caused pure economic loss. The testing of the ground structure prior to building a glass
factory was inadequate. The result was that the levels inside the factory were defective, and
extensive repair work had to be done. The engineering firm was initially in a direct contractual
relationship with the client, but later became a subcontractor. The Supreme Court of Appeal
held that the client had a contractual remedy and no concurrent action in delict. A key
consideration was that this case did not involve an infringement of rights of property or
person. It only involved the infringement of a contractual duty to perform specific professional
work with due diligence and care. There was no independent duty for the purposes of
delictual liability. In the view of most commentators, this judgment implies that only a
contractual remedy is available where negligent performance of a contractual duty causes
pure economic loss (as opposed to physical damage or personal injury), notably in the
relationship between a client and a professional person.53

Tsimatakopoulos v Hemingway, Isaacs & Coetzee CC54

In this case, the Court held that a firm of engineers was liable in delict for negligently
designing a defective wall. The engineers were contracted by their client to design a retaining
wall on the client’s property, and had negligently designed the wall with inadequate strength,
so that it leaned over. The client had sold the property to the plaintiff, who had to incur the
cost of restoring the stability of the wall. The principle on which the judgment was based was
the same as those accepted in cases that involve pure economic loss, namely that a party to
a contract can have a legal duty and be liable in delict to a person who is not a party to the
contract. Recognising such a legal duty is:

… the outcome of a value judgment that the plaintiff’s invaded interest is

deemed worthy of legal protection against negligent interference by conduct of
the kind alleged against the defendant.55

There was no contractual connection between the plaintiff and the engineers. When
considering the question whether the plaintiff had an independent claim in delict against the
engineers, it was irrelevant that the original client might have had a claim in contract against
the engineers, and that the plaintiff might have been able to take cession of such a claim. The
real issue was how far the engineers’ legal duty should be extended in any given situation. A
professional engineer should have foreseen that the wall would not remain stable and that the
property on which it was built could be sold to another person, and that this negligence was
likely to cause harm to the subsequent owner.
 SM Goldstein & Co (Pty) Ltd v Cathkin Park Hotel (Pty) Ltd56
This case involved a delictual action for damages by the owner of a hotel against the building
contractor responsible for building the hotel to the architect’s design. The contractor built a
fireplace for an open fire in the lobby of the hotel as designed by the architect. However, the
design created a fire hazard and did not comply with the installation instructions of the
Jetmaster firebox built into the fireplace. In due course, a fire broke out because of the faulty
design of the fireplace, and the hotel was extensively damaged. Although the building
contractor had built the fireplace according to the architect’s design (as required in the
building contract), he had failed to comply with the installation instructions of the Jetmaster
firebox as the building contract also required. It was common cause that it would have been
obvious to any builder that constructing the fireplace according to the architect’s design was
unsafe. The Court held that a builder generally has a legal duty to both the owner and third
parties to refrain from building something that is obviously unsafe. The foreseeability of the
harm was also a relevant consideration in deciding whether a legal duty and wrongfulness
existed. Therefore, the building contractor had acted both wrongfully and negligently, and was
liable in delict for the owner’s loss.
Although the SM Goldstein case, like the Tsimatakopoulos case, did not deal with pure
economic loss, the approach to determining whether a legal duty and wrongfulness exists
corresponds to the approach adopted by courts in cases that involve pure economic loss.
The Court in this case did not even refer to the question of concurrence of delictual and
contractual actions, accepting that in cases involving foreseeable harm, a builder has a legal
duty towards the owner and third parties to build safely, with the apparent implication that this
duty arises independently from his contract with the owner. This case also confirms that
recognising delictual liability for pure economic loss within a contractual relationship is
subject to the qualification that the contracting party who elects to sue in delict should not be
able to circumvent contractual immunities, waivers, or other limitations of liability to which he
agreed. The defendant can invoke such clauses against a delictual claim by way of a
defence, as happened in this case. However, on the facts of this case, and on the wording of
the building contract, the Court decided that the builder could not rely on the liability
exemption clause in the contract.

Pienaar v Brown57
The test for liability of a building contractor as employer of an independent subcontractor, for
harm negligently caused by the subcontractor, consists of three broad questions: (1) Would a
reasonable person have foreseen the risk of danger in consequence of the work he employed
the contractor to perform? If so, (2) would a reasonable person have taken steps to guard
against the danger? If so, (3) were such steps duly taken in the case in question? The liability
in these cases is personal, not vicarious, and it is not a question of the liability of the
employer being passed to the independent contractor, and thence to any subcontractor, but a
question of the respective individual liability of each of them. Only where the answer to the
first two questions is in the affirmative does a ‘legal duty arise, the failure to comply with
which can form the basis of liability’.58
On a proper reading of the National Building Regulations and Building Standards Act 103
of 1977 there is nothing to suggest that a failure to comply with its requirement to submit
plans for a building in terms of section 4(1), read with section 7, would necessarily lead to
liability for damage caused by collapse of part of the building work. There must be a causal
link between the failure to submit the plans and the collapse of the building work.59

When deciding the liability of professionals in the construction industry, courts take into account
all the circumstances, including the nature of the work and the nature of the alleged failing (error of
judgement, or lack of required knowledge or care).
22.5 Accountants and auditors
The liability of accountants and auditors to their clients is primarily contractual and is governed by
the same principles that apply to attorneys, engineers and other professionals. The Supreme Court
of Appeal extensively considered the duties of auditors towards their clients in Thoroughbred
Breeders’ Association of South Africa v Price Waterhouse.60 In this case, the Court held that
contributory negligence, in terms of the Apportionment of Damages Act 34 of 1956, does not apply
in the context of the contractual relationship between auditor and client.
Legislation governs the delictual liability of accountants and auditors to third parties (non-
clients), who rely on a negligently formulated opinion, certificate, report or statement. This
legislation essentially recognises an auditor’s or accountant’s duty to a third party if the auditor
knew or could reasonably be expected to know that the third party would act on the information
furnished to him or her.61
A potential investor or lender to a company should be entitled to rely on the company’s audited
financial statements, that is, the statements should give a correct and clear picture of the company’s
financial position. In the event of incorrect or misleading statements, an investor or lender may
claim delictual damages from the auditor. Section 46(2) and (3) of the Auditing Profession Act 26
of 2005 describes when an auditor will be liable for an opinion, report or statement:
(2) In respect of any opinion expressed or report or statement made by a registered auditor
in the ordinary course of duties the registered auditor does not incur any liability to a
client or any third party, unless it is proved that the opinion was expressed, or the
report or statement made, maliciously, fraudulently or pursuant to a negligent
performance of the registered auditor’s duties.
(3) Despite subsection (2), a registered auditor incurs liability to third parties who have
relied on an opinion, report or statement of that registered auditor for financial loss
suffered as a result of having relied thereon, only if it is proved that the opinion was
expressed, or the report or statement was made, pursuant to a negligent performance of
the registered auditor’s duties and the registered auditor–
(a) knew, or could in the particular circumstances reasonably have been expected to know,
at the time when the negligence occurred in the performance of the duties pursuant to
which the opinion was expressed or the report or statement was made–
(i) that the opinion, report or statement would be used by a client to induce the third
party to act or refrain from acting in some way or to enter into the specific
transaction into which the third party entered, or any other transaction of a similar
nature, with the client or any other person; or
(ii) that the third party would rely on the opinion, report or statement for the purpose
of acting or refraining from acting in some way or of entering into the specific
transaction into which the third party entered, or any other transaction of a similar
nature, with the client or any other person; or
(b) in any way represented, at any time after the opinion was expressed or the report or
statement was made, to the third party that the opinion, report or statement was
correct, while at that time the registered auditor knew or could in the particular
circumstances reasonably have been expected to know that the third party would rely
on that representation for the purpose of acting or refraining from acting in some way
or of entering into the specific transaction into which the third party entered, or any
other transaction of a similar nature, with the client or any other person.

In terms of these provisions, an auditor will not incur liability to a client or any third party in
respect of a report or opinion expressed by the auditor unless it was given maliciously, fraudulently
or pursuant to negligent performance of the auditor’s duties. Section 46(3) provides that if the
report was made negligently, the auditor will be liable to a third party who relied on it. However,
this is only if the third party can prove that the auditor knew or could, in the particular
circumstances, reasonably have been expected to know that the client would use the opinion or
report to induce the third party to act or refrain from acting, or that the third party would rely on the
opinion or report.
Arguably, every auditor knows or ought reasonably to foresee that the annual financial
statements of a company are widely used by investors and lenders for investment and business
decisions. Also, one would reasonably expect an auditor to know that the client or the third party
would act as contemplated in section 46(3)(a)(ii). However, for the purposes of this section, the
requisite knowledge or foreseeability depends on the circumstances of each case.
There are different frameworks for auditing in South Africa. In terms of the Companies Act 71
of 2008 62 the auditors have a duty to ensure that the presentation of a company’s financial position
in the financial statements complies with the financial reporting standards applicable to the type of
company. If an auditor does not ensure compliance with such practice, this constitutes a breach of
the auditor’s statutory duties and is likely to amount to negligence. The SA Institute of Chartered
Accountants (SAICA), in conjunction with the Accounting Practices Board, also produces
frameworks for auditing. A company must disclose which auditing framework it has adopted. An
auditor’s failure to ensure that the financial statements conform to the applicable standards is a
breach of his or her statutory duty in terms of the Companies Act and is likely to amount to
negligence.
International Shipping Co (Pty) Ltd v Bentley63
In 1977, a financing company made facilities available to a group of companies. Thereafter,
Bentley was appointed auditor to the group. From 1979 he issued unqualified financial
statements that were to some extent false or misleading. The company alleged that it had
relied on the statements to review its facilities to the group, and when deciding whether to
increase the facilities. The group was liquidated in 1981 and the financing company suffered
a loss. The Court found that Bentley’s conduct was wrongful and blameworthy, and a conditio
sine qua non of the company’s loss. However, a number of other factors had intervened and
the Court regarded the ultimate loss as being too remote for Bentley to be held liable.

NPC Electronics Ltd v S Taitz Kaplan & Co64

The plaintiff had relied on unqualified reports by the defendant firm of accountants and
auditors on the financial statements of certain companies, prior to selling and delivering
goods to these companies and making additional credit facilities available to them. The
companies were subsequently liquidated. There was no contractual or other relationship
between the parties. The Court held that the auditor in this case did not conduct a proper
auditing procedure and failed to carry out its duties in terms of the Companies Act. However,
the defendant did not know, and could not reasonably be expected to foresee, that the
plaintiff would rely on the reports for this purpose. In the circumstances, the defendant did not
have a legal duty to prevent the plaintiff’s loss, and there was consequently no wrongfulness.
Even if there were such a duty and wrongfulness, the plaintiff did not prove that the defendant
had factually or legally caused its loss. The loss suffered by a third party reading financial
statements and using them in a manner unknown to the auditors should not be the auditors’
responsibility. Imposing liability in this case would place an unreasonable burden on the
auditors.

Axiam Holdings Ltd v Deloitte & Touche65

In this case, the plaintiff alleged that the auditors had owed a duty to third parties to warn
them that certain financial statements and their audit opinion were incorrect. Alternatively, it
was their duty to warn them that they had not conducted the audit properly, and that they
should not rely on the statements and the audit opinion. The plaintiff alleged that their failure
in this regard was negligent and constituted a misrepresentation within the meaning of
section 20(9)(b)(ii) of the Public Accountants’ and Auditors’ Act 80 of 1991 (which applied at
that time). The auditors raised an exception on the basis that the allegations in the pleadings
did not justify the conclusion that the auditors owed a legal duty to the third parties. Navsa JA
said the following on liability of auditors for negligent misstatement:

It is true that decisions by courts on whether to grant or withhold a remedy for

negligent misstatement causing economic loss are made conscious of the
importance of keeping liability within reasonable bounds. It is universally
accepted in common law countries that auditors ought not to bear liability
simply because it might be foreseen in general terms that audit reports and
financial statements are frequently used in commercial transactions involving
the party for whom the audit was conducted (and audit reports completed) and
third parties. In general, auditors have no duty to third parties with whom there
is no relationship or where the factors set out in the Standard Chartered Bank
case are absent.66

The majority of the Court held that in this case they could not decide on exception about the
wrongfulness of the harm caused by omission to speak. The matter had to go to trial. It was
premature to decide on exception whether a legal duty existed.

Stols v Garlicke & Bousfield Inc67

S sued GB, a firm of attorneys, on a contract of deposit. In terms of this contract, S had
deposited money with GB, and this money was meant to be repaid with interest on a
specified date. GB had been represented by C, an executive consultant who the firm had
employed. C did not deposit S’s payments into GB’s trust account, but instead paid the
money into an account belonging to PKF, a firm of chartered accountants. When the money
was not repaid on the specified date, S sued GB for payment. GB denied any knowledge of
the investment scheme, and claimed C had not been authorised to represent the firm in
concluding any contracts of this nature. GB also filed a third-party notice seeking to hold PKF
liable for the loss, alleging PKF was aware that GB had employed C as an executive
consultant, and PKF therefore had a legal duty to alert GB to the fact that C had been using
its accounts for an investment scheme in an irregular manner. PKF excepted to the third-
party notice, in part on the basis that the notice did not contain the averments necessary to
sustain an action for damages in delict based on a legal duty to speak.
The Court held that GB’s claim against PKF was delictual, based on an omission causing
pure economic loss.68 The legal duty to speak on PKF which GB claimed to have existed did
not fit into a recognised category, and the law is always cautious in extending liability for pure
economic loss to new situations.69 However, the Court held that the averments in the third-
party notice indicated prima facie that a legal duty to speak existed; or, at the very least, it
was not possible to decide whether or not such a duty existed on the pleadings alone.70 The
Court considered the following factors to have been indicative of a legal duty on PKF to
speak in the circumstances:
• Although no formal legal relationship came into being between GB and PKF, PKF believed
itself to be providing facilities for GB in dealings between GB and GB’s clients.71
• There had been prior conduct on the part of PKF, in that it had been involved in the
authorisation of payments from the accounts under its control.72
• PKF knew, or ought to have known, that the accounts were conducted irregularly, as the
deposits were not paid into GB’s trust account.73
• PKF knew that C had refused to use GB’s trust banking account, and C’s use of accounts
held by others would probably have led PKF to conclude that C wanted to avoid trust-
account standards and the scrutiny which accompanies the exclusive use of GB’s trust
account.74
• Even if PKF did not draw these inferences, it knew that GB’s trust account should have been
used, and that C had deliberately avoided doing this.75
• PKF probably knew that firms of attorneys would not function in the way C had conducted
the transactions, and evidence may show that this would have raised a flag in PKF’s mind
that C was on a frolic of his own and that GB was unaware of how C was functioning.76
 Having viewed the above factors objectively, the Court concluded that PKF had a legal duty
to warn GB about C’s conduct:77

In the light of these factors, can it then be said that public or legal policy
considerations require that the failure to speak on the part of the third party
should be actionable? Put another way, do the legal convictions of the
community demand that a legal duty be imposed on the third party to speak? It
should be borne in mind that no one factor is decisive. However, there are
cumulative aspects averred which weigh on the overall outcome: prior conduct
in terms of which the danger to the defendant was caused; the belief of
contractual proximity with the defendant on the part of the third party; the
knowledge that the accounts were being operated in a way which circumvented
the checks and balances normative for attorneys’ trust accounts; and society’s
recognition of the need for increased vigilance to ensure that people who
purport to open and operate accounts on behalf of another are authorised to do
so. In the light of these and the other factors dealt with above, both questions
must be answered in the affirmative. I conceive that it would be contrary to
public policy to exonerate the third party from speaking when it allowed its
facilities to be used in what it believed to be an operation run by the defendant
which was clearly being conducted in a manner inimical to the strictures of the
legal profession of which the defendant is a part and where the defendant was
placed at risk. Viewed objectively, society will take account of these factors and
require such a legal duty to be imposed.

When deciding on the liability of accountants and auditors, courts consider all the circumstances,
including the nature of the work (advisory, or auditing) and the nature of the alleged failing (error
of judgement, or lack of required knowledge or care).

22.6 Banks
Banking is an occupation that demands special knowledge and skill, and bankers must exercise
reasonable care and measure up to the standard of competence that one can reasonably expect of a
person professing knowledge and skill in banking. The relationship between bank and client is
contractual, with express or implied terms, or both, in respect of the banking service contracted for.
Also, the bank may have a legal duty to provide the client with information, from which arises the
potential liability for misinformation. Liability could be based on the contractual relationship with
clients or on a duty in delict towards third parties harmed by the performance of banking services to
clients.
Banking encompasses a wide variety of activities, which include assisting clients in
commercial transactions, dealing with collecting cheques, conducting foreign currency transactions,
reporting on the creditworthiness of clients, reporting on the state of accounts, dealing with
insurance matters, and offering investment advice. The following cases illustrate the potential
liability of banks in these areas.
Indac Electronics (Pty) Ltd v Volkskas Bank Ltd78
The Court held that a collecting banker can be liable under the extended lex Aquilia for
negligence to the true owner of a cheque. In this case, the owner was specified as the payee
and the cheque was crossed and marked ‘not negotiable’, but the cheque was nevertheless
collected for and the proceeds credited to another person. A delictual action for damages
would be available to a true owner of a cheque who can establish that:
• The collecting banker received payment of the cheque on behalf of someone who was not
entitled to it
• In receiving such payment, the collecting banker acted (a) negligently and (b) unlawfully
• The conduct of the collecting banker caused the true owner to sustain loss
• The damages claimed represent proper compensation for such loss.

When deciding on the principle in the Indac case that a collecting bank can be liable to the
true owner of a cheque for loss suffered if the cheque was paid to someone else who was not
entitled to it, the Court made a value judgment and referred to a number of policy
considerations relevant to the existence of a bank’s legal duty towards its client. The
objection of limitless or indeterminate liability does not arise in this case, because the extent
of the potential loss is finite (the face value of the cheque) and the potential claimants are
easily predictable and are limited to the drawer or the payee (or someone holding title under
him). Also, each potential claim will arise separately from any other and will be related to a
specific act on the part of the collecting banker. The collecting banker provides a professional
service to collect cheques payable to his or her client, and should be aware that failure to
take reasonable care may result in loss to the true owner of the cheque. The collecting
banker, by virtue of his or her calling, possesses or professes to possess special skill and
competence in his field and can, or ought to appreciate the significance of instructions on a
cheque. The bank is thus able to reduce, if not avoid, loss to the true owner by exercising
reasonable care when collecting cheques.
If there were no legal duty to take reasonable care, a collecting banker would not have to
examine, or even look at the cheque to ascertain to whom it is payable. The crossing of a
cheque would be of little consequence if there were no legal duty on the part of the collecting
banker. The collecting banker is the only person who is in a position to know whether a
cheque is being collected on behalf of a person who is entitled to receive payment. The
drawee bank has to rely on the collecting banker to ascertain this fact. The collecting bank is
fully aware of this position and therefore has a duty to ensure that it only presents a cheque
for payment on behalf of a client who is entitled to receive payment of the cheque. Whereas
the drawer or true owner of a cheque cannot protect himself or herself against loss if the
collecting banker negligently collects payment on behalf of the wrong person, the collecting
banker does have a right of recovery. The bank would have a claim for reimbursement
against a customer who was wrongly credited with the proceeds of the cheque. If that
customer is unable to pay, it would be more appropriate to allocate the loss to the banker
who chose to accept the customer’s business, than to the innocent true owner of the cheque.
Furthermore, a collecting bank could relatively inexpensively protect itself against such loss
by obtaining insurance cover.

Siman & Co (Pty) Ltd v Barclays National Bank Ltd79

The plaintiff had approached the defendant bank on a Friday afternoon to procure immediate
forward cover for its foreign currency commitments incurred for overseas purchases, as it had
heard that the Rand unit of currency would probably be devalued over that weekend. The
defendant bank informed the plaintiff that it was then too late for the cover to be procured that
afternoon and that it could only be procured on the following Monday. The Rand was in fact
devalued during that weekend and the plaintiff, when subsequently paying for its foreign
commitments, had to pay much more in Rands than it would have cost it if it had procured the
immediate forward cover on the Friday afternoon. The statement by the bank official on that
Friday, that it was too late to obtain forward cover that day, was in fact incorrect.
Consequently, the plaintiff instituted action against the bank for damages in respect of the
excess amount that it had to pay. The plaintiff alleged the following: that it had been misled by
the defendant’s statement that it was too late to procure the required cover that Friday
afternoon; that such statement was false because the cover was still procurable; and that the
statement had been made negligently and had caused the plaintiff loss. The Appellate
Division held that the action must fail, deciding that the bank official’s conduct involved two
relevant acts:
• A firm decision or refusal, taken and persisted in, not to procure any forward cover for the
plaintiff that afternoon
• A misstatement relating to the official’s reason for the refusal, that it was too late to do
anything that afternoon.

The Court had to distinguish between the two acts, because the plaintiff had for the purposes
of his action selected and relied only on the second act, on which he had founded his cause
of action. The plaintiff had not relied on the first act, the refusal to obtain the forward cover, at
any stage of the proceedings. Based on the facts, the plaintiff’s loss was due to the
defendant’s refusal to procure the cover that afternoon, when there was still time to do so.
The defendant’s reasons for refusing, wrong though they were, were then of no
consequence. (In a dissenting minority judgment, Corbett JA found that the bank official,
acting in the course and scope of his employment, made a misstatement to the client, and in
doing so acted negligently and wrongfully, and that the misstatement had caused the client’s
loss.)

Durr v ABSA Bank Ltd80

The appellant-plaintiff sought to recover from the respondent-defendant (the bank) damages
for the loss of money that she and other members of her family had invested in certain
companies that went insolvent shortly after recommendation by the regional manager of the
bank’s broking division. The manager had described the investments as ‘safe’, ‘very solid’,
‘very secure’, and ‘very sound’, and had strongly recommended them. The plaintiff alleged
that she had suffered loss as a result of the manager’s negligent failure to exercise the
degree of care and skill that she was entitled to expect from someone in his position, and for
which failure the bank should be held vicariously liable. Her evidence was that she would
never have thought of investing in the companies were it not for the advice of the manager,
and that she believed the bank had financial experts who would examine prospective
investments. The companies were represented in publicity material as having a sound
financial base and brokers were offered commission at a rate substantially higher than that
for comparable investments. Brokers were invited to marketing conferences, and were given
glossy brochures and dossiers that contained laudatory press cuttings. No prospectuses or
financial statements were provided. The companies concerned did not have the audited
financial statements or prospectuses required by law, and did not make any financial
statements available to brokers. The manager had obtained his information about the
investments from the directors of the companies themselves and from attending their
marketing conferences. The manager’s real skills were confined to advising clients on
planning their affairs with regard to income tax, estate duty, and investments such as
endowments, retirement annuities, unit trusts, fixed deposits, life assurance and the like. He
at no stage sought information or advice about investing in the companies from people who
were better placed to evaluate them.
The Court held that two issues arose: namely, in general, what was the level of skill and
knowledge required, and whether the standard against which that skill and knowledge should
be judged was that of an ordinary or average broker or that of a regional manager of the
brokering division of a bank professing investment skill and offering expert investment
advice. For the first issue, the Court held that:

in deciding what is reasonable the Court will have regard to the general level of
skill and diligence possessed and exercised at the time by the members of the
branch of the profession to which the practitioner belongs.

But that:

the decision of what is reasonable under the circumstances is for the Court; it
will pay high regard to the views of the profession, but it is not bound to adopt
them.
 For the second issue, the Court held that, as a matter of law set in the factual context of the
case, the relevant standard in this instance was not that of the average or typical broker of
modest accomplishments, because accepting this standard would allow a definition chosen
by a witness for his own purposes to dictate the result, making the enquiry of what was
required of a particular type of broker pointless. The appropriate standard was that of the
regional manager of the brokering division of a financial institution professing investment
skills and offering expert investment advice. As to the manager’s negligence, the basic rule
was as follows:

[L]ack of skill or knowledge is not per se negligent. It is, however, negligent to

engage voluntarily in any potentially dangerous activity unless one has the skill
and knowledge usually associated with the proper discharge of the duties
connected with such an activity.

In bringing the investments to the plaintiff’s attention and assuring her that he had investigated it
and strongly recommended it, he had advised the plaintiff to embark upon what was, in effect, the
potentially dangerous activity of money lending. The managers had a duty to make enquiries about
the companies, and the bank, given that it had held itself out as an expert investment advisor, could
not argue that it had not been under a duty to investigate the creditworthiness of the companies. The
plaintiff had been entitled to see the manager as a man skilled to advise her on such matters and as
one backed by a major financial institution, not as someone devoid of skills in assessing
creditworthiness and unready to seek help. The manager had the duty to investigate the
creditworthiness of the companies and should either have forewarned the appellant and her family
where his skills ended, to allow them to appreciate the dangers of accepting his advice, or he should
not have recommended the investment. The manager had, therefore, not performed his duties
adequately and had negligently caused the plaintiff’s loss. The bank was vicariously liable.

As to the basis of liability imposed, Schutz JA held:81

The claim pleaded relied upon contract, alternatively delict, but as the case was
presented as one in delict, and as nothing turns upon the precise cause of
action, I shall treat it as such.

BOE Bank Ltd v Ries82

The Supreme Court of Appeal refused to recognise a legal duty on the part of an insurance
broker who worked for a bank towards an intended beneficiary under a life insurance policy.
The broker had failed to ensure that the policy holder signed the necessary form to nominate
the beneficiary. The Court took into account that the broker’s conduct did not amount to
assuming any professional responsibility regarding signing the form, because he became
involved in the process in passing, by doing a favour for a colleague. In effect, he was acting
as a messenger, without undertaking any professional responsibility. Although it was
foreseeable that the intended beneficiary would obtain no benefit from the policy if the holder
failed to sign the necessary nomination form before his death, foreseeability of loss was not in
itself enough to indicate a legal duty.

When deciding the liability of banks, courts consider all the circumstances, including the nature of
the services rendered to the client and the nature of the alleged failing (error of judgement, or lack
of required knowledge or care).

1 Bayer South Africa (Pty) Ltd v Frost 1991 (4) SA 559 (A) at 570.
2 Pretorius v McCallum 2002 (2) SA 423 (C); BOE Bank Ltd v Ries 2002 (2) SA 39 (SCA).
 3 Van Wyk v Lewis 1924 AD 438; Blyth v Van den Heever 1980 (1) SA 191 (A) at 221; Castell v De Greeff 1993 (3) SA
501 (C) at 509; Broude v McIntosh 1998 (3) SA 60 (SCA); Mukheiber v Raath 1999 (3) SA 1065 (SCA).
4 Bruce NO v Berman 1963 (3) SA 21 (T); Honey & Blanckenberg v Law 1966 (2) SA 43 (R); Mouton v Die
Mynwerkersunie 1977 (1) SA 119 (A); Slomowitz v Kok 1983 (1) SA 130 (A). As to the duty of a lawyer towards an
opponent in litigation, see Barlow Rand Ltd t/a Barlow Noordelike Masjinerie Mpy v Lebos 1985 (4) SA 341 (T);
Pretorius v McCallum 2002 (2) SA 423 (C); and see, generally, Midgley Lawyers Professional Liability (1992) at 120
–132.
5 International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A); Thoroughbred Breeders’ Association v Price
Waterhouse 2001 (4) SA 551 (SCA); Axiam Holdings Ltd v Deloitte & Touche 2006 (1) SA 237 (SCA).
6 Siman & Co (Pty) Ltd v Barclays National Bank Ltd 1984 (2) SA 888 (A); BOE Bank Ltd v Ries 2002 (2) SA 39
(SCA); Holtzhausen v ABSA Bank Ltd 2008 (5) SA 630 (SCA).
7 Durr v ABSA Bank Ltd 1997 (3) SA 448 (SCA); Pinshaw v Nexus Securities (Pty) Ltd 2002 (2) SA 510 (C).
8 Perlman v Zoutendyk 1934 CPD 151.
9 EG Electric Co (Pty) Ltd v Franklin 1979 (2) SA 702 (E).
10 Brown v Hunt 1953 (2) SA 540 (A).
11 Wessels v Bouwer 1971 (1) PH J9 (NC) (involving the use of a tractor-driven hammer-mill to crush maize cobs).
12 Van Wyk v Hermanus Municipality 1963 (4) SA 285 (C); Simon’s Town Municipality v Dews 1993 (1) SA 191 (A).
13 R v Du Toit 1947 (3) SA 141 (A); Beswick v Crews 1965 (2) SA 690 (A) at 705; Griffiths v Netherlands Insurance Co
of SA Ltd 1976 (4) SA 691 (A) at 695; Butt v Van den Camp 1982 (3) SA 819 (A) at 826.
14 1997 (3) SA 448 (SCA) at 462 F–G.
15 Van Wyk v Lewis 1924 AD 438 at 444. See also Blyth v Van den Heever 1980 (1) SA 191 (A) at 221; Randaree v W H
Dixon & Associates 1983 (2) SA 1 (A) at 4; Castell v De Greeff1993 (3) SA 501 (C) at 509.
16 Mouton v Die Mynwerkersunie 1977 (1) SA 119 (A) at 143.
17 1924 AD 438 at 444.
18 1924 AD 438 at 457.
19 Colman v Dunbar 1933 AD 141 at 157; Van Heerden v SA Pulp & Paper Industries Ltd 1946 AD 382.
20 See Mitchell v Dixon 1914 AD 519 at 525; Van Wyk v Lewis 1924 AD 438; Blyth v Van den Heever 1980 (1) SA 191
(A) at 221; Castell v De Greeff 1993 (3) SA 501 (C) at 509; Broude v McIntosh 1998 (3) SA 60 (SCA); Mukheiber v
Raath 1999 (3) SA 1065 (SCA).
21 Van Wyk v Lewis 1924 AD 438.
22 1924 AD 438 at 456.
23 1957 (3) SA 710 (T) at 723H.
24 1976 (2) SA 891 (T) at 894.
25 See, generally, Esterhuizen v Administrator, Transvaal 1957 (3) SA 710 (T) at 719; Santam Insurance Co Ltd v
Vorster 1973 (4) SA 764 (A) at 781; Castell v De Greeff1994 (4) SA 408 (C) at 425.
26 1994 (4) SA 408 (C).
27 At 426G.
28 1998 (3) SA 60 (SCA).
29 Administrator, Natal v Edouard 1990 (3) SA 581 (A).
30 In Premier, Kwazulu-Natal v Sonny 2011 (3) SA 424 (SCA) a doctor failed to inform a pregnant patient of the
meaning of an ultrasound scan which indicated possible Down’s syndrome, or of the importance of the patient
returning for a second scan. This, coupled with a nurse’s later misdirection, resulted in the patient not returning for the
scan, and not having further tests to diagnose the syndrome within the time allowed to terminate the pregnancy. The
child was later born with Down’s syndrome, and the patient and her husband sued successfully for the costs of
maintaining their daughter.
31 2008 (6) SA 310 (SCA).
32 2015 (2) SA 193 (CC).
33 1999 (3) SA 1065 (SCA).
34 2016 (1) SA 325 (CC).
35 Paras 51–68. Section 27(3) of the Constitution provides that no one may be refused emergency medical treatment.
36 Para 68.
37 See, generally, Midgley (1992) at 120–132.
38 Honey & Blanckenberg v Law 1966 (2) SA 43 (R); Bruce NO v Berman 1963 (3) SA 21 (T) at 23; Mouton v Die
Mynwerkersunie 1977 (1) SA 119 (A); Groom v Crocker [1938] 2 All ER 394 (CA) at 401–402.
39 Honey & Blanckenberg v Law 1966 (2) SA 43 (R).
40 Mouton v Die Mynwerkersunie 1977 (1) SA 119 (A) at 143.
41 Lavery & Co Ltd v Jungheinrich 1931 AD 156 at 169; Whitfield v Phillips 1957 (3) SA 318 (A) at 325; Saif Ali v
Sydney Mitchell & Co & P [1978] 3 All ER 1033 (HL); Mlenzana v Goodrick & Franklin Inc 2012 (2) SA 433 (FB).
42 Rampal (Pty) Ltd v Brett Wills and Partners 1981 (4) SA 360 (D).
43 Mouton v Die Mynwerkersunie 1977 (1) SA 119 (A); and see also Bouwer v Harding 1997 (4) SA 1023 (SE), where
the liability of an attorney to a client was the subject of an action for breach of contract, and the attorney was found to
 be in breach of his contractual duties by negligently failing to advise the client correctly on the tax implications of
structuring a sale of property in a particular manner.
44 See Pretorius v McCallum 2002 (2) SA 423 (C).
45 Barlow Rand Ltd t/a Barlow Noordelike Masjinerie Mpy v Lebos 1985 (4) SA 341 (T).
46 2013 (1) SA 12 (GNP).
47 2013 (2) SA 311 (SCA).
48 Paras 27 and 30–31.
49 Paras 32–33, 35–36 and 38.
50 See, generally, Midgley (1992) at 1ff.
51 See Rondel v Worsley [1966] 3 All ER 657 (CA) and Rondel v Worsley [1967] 3 All ER 993 (HL); Saif Ali v Sydney
Mitchell & Co [1978] 3 All ER 1033 (HL).
52 Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A).
53 Midgley ‘The nature of the inquiry into concurrence of actions’ (1990) 107(4) SALJ at 621; Van Aswegen ‘Die Toets
vir die Bestaan van ’n “Onafhanklike Delik” in die Geval van Kontrakbreuk’ (1992) 55 THRHR 271 at 273–274.
54 1993 (4) SA 428 (C).
55 At 431H–I.
56 2000 (4) SA 1019 (SCA).
57 2010 (6) SA 365 (SCA).
58 Para 11 at 368H–369B.
59 Para 16 at 370D–E, read with para 7 at 367H–I.
60 2001 (4) SA 551 (SCA).
61 Section 46 of the Auditing Profession Act 26 of 2005.
62 Section 84(4)(b).
63 1990 (1) SA 680 (A) at 700.
64 [1998] 1 All SA 390 (W).
65 2006 (1) SA 237 (SCA).
66 Para 18.
67 2012 (4) SA 415 (KZP).
68 Para 18.
69 Para 17.
70 Para 30.
71 Para 18.
72 Para 19.
73 Para 20.
74 Para 21.
75 Para 22.
76 Para 23.
77 Para 29.
78 1992 (1) SA 783 (A) at 797–801.
79 1984 (2) SA 888 (A).
80 1997 (3) SA 448 (SCA).
81 At 453.
82 2002 (2) SA 39 (SCA) paras 13–26.
 Chapter 23

Injury or death of another person

23.1 No general right

23.2 Claims based on injury of a dependant

23.3 Claims by parents and employers for loss of services

23.4 Contracting parties

23.5 Claims by heirs and legatees for reduced inheritance

23.6 Claims by heirs or family members for funeral expenses

23.7 Claims by executor for funeral and other expenses

|
23.8 Claims based on death of support provider (action of dependants)
23.8.1 History
23.8.2 Nature of, and requirements for, the action
23.8.3 Who can sue?
23.8.4 Damages

23.9 Claims for loss of support based on injury to the support provider

23.1 No general right

There is no general right to recover damages for patrimonial harm suffered as a result of the injury
or death of another person. Here one is concerned with pure economic harm, and causing such
harm is not prima facie wrongful. However, there are exceptional circumstances under which the
injury or death of another person may give rise to delictual liability. These exceptions are cases in
which the harm derives from a recognised duty of support, mostly based on a family relationship.
23.2 Claims based on injury of a dependant
A person with a duty of support may claim damages for patrimonial harm suffered as a result of the
injury of a dependant, such as a spouse1 or child.2 The harm may consist of actual medical
expenses, prospective future medical expenses, or the costs of increased duties of maintenance.3
Where the spouse or child is partly responsible for the injury or death, he or she is treated as a joint
wrongdoer under sections 2(1A) and 2(1B) of the Apportionment of Damages Act 34 of 1956.

23.3 Claims by parents and employers for loss of services

In Roman-Dutch law parents had an action for loss of income suffered as a result of injury to a
child.4 The master of a domestic servant was also afforded an action for harm that resulted from
loss of services due to the injury of a servant.5 The modern relationship between an employer and
domestic worker is now completely different. The services of domestic workers are no longer
regarded as contributions in kind to the household of their employer. Therefore, employers cannot
recover damages for pure economic harm suffered as a result of injury to employees, domestic or
otherwise.6 Similarly, the common-law action for loss of services caused by the injury of a
domestic servant is no longer recognised.7

23.4 Contracting parties

As discussed in Chapter 17, there is no general claim in South African law for pure economic harm
suffered as a result of negligent interference with contractual relations, or what is sometimes called
‘contractual relational economic loss’. 8 Courts do not recognise claims for damages by those who
suffer harm as a result of the death or injury of another person solely because of a contractual
relationship with the injured or deceased person. In Union Government v Ocean Accident &
Guarantee Corporation Ltd 9 the Court declined to expand Aquilian liability beyond the historical
scope of the action of dependants, for policy reasons. Schreiner JA pointed out the danger of
indeterminate and unlimited liability inherent in expanding this action. He referred to an
‘unmanageable situation’ and ‘absurdities’ that could arise from affording an action to, for
example, property owners, sub-contractors, and employees who were in a contractual relationship
with a building contractor, for losses suffered if a third party causes the injury or death of the
building contractor.10

23.5 Claims by heirs and legatees for reduced inheritance

Heirs and legatees have no claim for damages based on the possibility of a reduced inheritance
caused by the premature death of the testator. The heirs and legatees may inherit less than they
would have done had the testator lived longer, but only if the value of the estate would have grown
on account of the testator’s earning capacity. However, the future value of the estate is uncertain.
For policy reasons, the law does not recognise that the wrongdoer who caused the death of the
testator owes a duty to the heirs and legatees, because their loss, if any, is indeterminate and
speculative.11

PAUSE FOR Reduced inheritance

REFLECTION
 Why is this situation different from that of a ‘disappointed beneficiary’,
who obtains no benefit from a will, because the testator’s legal adviser
had negligently failed to ensure that the will was properly signed?12

23.6 Claims by heirs or family members for funeral expenses

Heirs or immediate family members who have incurred funeral expenses have an action (actio
funeraria) to recover reasonable funeral expenses from the wrongdoer who had caused the death.
The action is based on the duty to bury or cremate the deceased, and includes, for example,
reasonable costs of burial or cremation, transport and a tombstone.13 The action is anomalous in the
sense that these costs would eventually have been incurred in any event, because death is
inevitable. However, this anomaly is outweighed by the fact that the wrongdoer had caused
premature death.

23.7 Claims by executor for funeral and other expenses

The executor of a deceased estate, who has incurred funeral expenses out of the estate, has an
action to recover reasonable funeral expenses from the wrongdoer who caused the death. The
executor can also recover costs incurred by the estate for the medical treatment of the deceased
prior to death. There is no claim for the loss of income that the deceased could have earned but for
the premature death, because the earning capacity of the deceased is not an asset in his or her own
estate.14

23.8 Claims based on death of support provider (action of

dependants)
A delictual claim (actio legis Aquiliae) for the loss of support or maintenance arising from the
culpable and wrongful killing of a breadwinner (support provider) is recognised at common law as
the dependants’ action.15 In Fortuin v Road Accident Fund 16 the Court held that:
[In] our law ‘maintenance’ or ‘support’ not only includes food, clothing and shelter, but also
medical care and, in respect of children, education.

This is a claim for patrimonial harm in the form of pure economic harm.17
The claim by dependants for loss of support that results from the death of a breadwinner has a
long history and has been expanded in modern law. It is a claim often regarded as being in a
category of its own (sui generis), because the claimant (the dependant) derives the claim both from
the death of the breadwinner, negligently and wrongfully caused by the wrongdoer, and also
independently, as a result of personal harm suffered through loss of support.

23.8.1 History
The action of dependants for loss of support resulting from the death of a breadwinner was
unknown in Roman law, but was recognised in Roman-Dutch law, probably under the influence of
customary Germanic law. Most Roman-Dutch writers did not attempt a systematic classification of
the action, but apparently considered it to be an independent delictual action, available to any
dependant who was legally entitled to, and actually received, support from the deceased. The action
of dependants as developed in Roman-Dutch law forms part of modern South African law and has
undergone further expansion. In Paixão v Road Accident Fund 18 the SCA stated:
Although the precise scope of the dependants’ action is unclear from the old Roman-Dutch
jurists, there is a strong suggestion that it was not confined only to those classes of persons to
whom the breadwinner had a legal obligation to support, but was also available to those
whom the deceased ‘was accustomed to support from a sense of duty.’

23.8.2 Nature of, and requirements for, the action

The basic requirements for the action by dependants, as set out in Brooks v Minister of Safety and
Security,19 are the following:
• A legal duty of support on the part of the deceased
• A legal right to such support on the part of the dependant(s).20

The action is a direct rather than a derivative action. In this regard, Innes CJ said the following in
Jameson’s Minors v Central South African Railways: 21
Our law, while recognising no right of action on behalf of the deceased’s estate, gives to
those dependent on him a direct claim, enforceable in their own names, against the
wrongdoer. This is a right not derived from the deceased man or his estate, but
independently conferred upon members of his family.

As discussed in Chapter 10, it follows that a waiver of action agreed to by the deceased
breadwinner before death (pactum de non petendo in anticipando) does not bind the dependant(s).22
If the dependant or the deceased breadwinner was partly responsible for the death, then the
breadwinner’s estate or the dependant is treated as a joint wrongdoer under section 2(1B) of the
Apportionment of Damages Act.
An essential and unusual feature of the action, as pointed out by Corbett JA in Evins v Shield
Insurance Co Ltd,23 is that one determines fault and wrongfulness with reference to both the death
of the breadwinner and the consequent harm suffered by the dependant:
… while the defendant incurs liability because he has acted wrongfully and negligently (or
with dolus) towards the deceased and thereby caused the death of the deceased, the claimant
(the dependant) derives his right of action not through the deceased or from his estate but
from the facts that he has been injured by the death of the deceased and that the defendant
is in law responsible therefore. Only a dependant to whom the deceased was under a legal
duty to provide maintenance and support may sue and in such action the dependant must
establish actual patrimonial loss, accrued and prospective, as a consequence of the death of
the breadwinner.

23.8.3 Who can sue?

The dependants’ claim for loss of support from the wrongful and culpable causing of the death of a
breadwinner is founded in family law. The claim was historically restricted to cases where a duty of
support derives from a valid, civil marriage, mainly concerning spouses and their biological
children. However, courts have developed the scope of the action and extended the group of
claimants, on the basis of the constitutional imperative of non-discrimination, to include those
whose right to support derives from a relationship which is similar to marriage, but does not
constitute a legally valid marriage between a man and a woman. The action has been developed in
the following cases:
• In Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality
Intervening) 24 the Supreme Court of Appeal had to determine whether the appellant, married
 to her deceased husband in terms of Islamic law, could institute the dependants’ action for loss
of support. The Court held that the correct approach was not to ask whether the marriage was
lawful at common law, but to establish if the deceased had a legal duty to support the appellant
during the subsistence of the marriage and, if so, whether the right of the widow was in the
circumstances a right which deserved protection for the purposes of the dependants’ action.25
The Court required the appellant to prove that the deceased had a legally enforceable duty to
support her, that it was a duty that arose from a solemn marriage in accordance with the tenets
of a recognised and accepted faith, and that it was a duty that deserved recognition and
protection for the purposes of the dependants’ action.26 The appellant was successful in proving
these requirements and the common law was developed accordingly.
• The primary question in Du Plessis v Road Accident Fund 27 was whether the plaintiff, who
had entered into a same-sex union with the deceased, was entitled to claim damages for loss of
support from the defendant. The Supreme Court of Appeal followed the reasoning in Amod,
and set out to determine whether the deceased had a legally enforceable duty to support the
plaintiff and, if so, whether the plaintiff’s right to such support was worthy of legal
protection.28 The Court answered both questions affirmatively and extended the dependants’
action for loss of support to include partners in a same-sex permanent life relationship that is
similar in other respects to marriage, who had a contractual duty to support one another.29
• The action for dependants was further extended in Santam Bpk v Henery 30 when the Supreme
Court of Appeal extended the action to divorced women who have a right to support under a
court order dealing with maintenance after dissolution of the marriage, and whose former
husbands were wrongfully and negligently killed.
• In Paixão v Road Accident Fund 31 the Supreme Court of Appeal considered whether the
common law should be developed to extend the dependants’ action to include permanent
heterosexual relationships. The appellants were Maria Paixão, who had a permanent
heterosexual relationship with the deceased, as well as her daughter. The deceased had been
living with Maria and her children at the time of his death, planned to marry her and executed
a joint will with her. When considering the merits of the case, the Court noted the practical
problems of proving a heterosexual life partnership that gives rise to a legal duty of support: 32
Proving the existence of a life partnership entails more than showing that the
parties cohabited and jointly contributed to the upkeep of the common home. It
entails, in my view, demonstrating that the partnership was akin to and had
similar characteristics — particularly a reciprocal duty of support — to a
marriage. Its existence would have to be proved by credible evidence of a
conjugal relationship in which the parties supported and maintained each other.
The implied inference to be drawn from these proven facts must be that the
parties, in the absence of an express agreement, agreed tacitly that their
cohabitation included assuming reciprocal commitments — ie a duty to support
— to each other. Courts frequently undertake this exercise without much
difficulty …

The Court also emphasised that the extension of the dependants’ action did not apply only to permanent
heterosexual relationships where there is agreement to marry.33 Instead, it confirmed that the primary
question is whether the deceased had a legally enforceable duty of support arising out of a relationship
akin to marriage.34 Ultimately, the Court developed the law so that the dependants’ action may also be
instituted by unmarried persons in ‘heterosexual relationships who have established a contractual
reciprocal duty of support’. 35
• In Engela v Road Accident Fund 36 the Court decided that the legal duty to support also applied
to the relationship between a daughter (plaintiff) and her mother’s ex-husband, who was not
her biological father. The deceased had been divorced from the plaintiff’s mother, but,
 following a reconciliation, had been in a permanent heterosexual relationship with the
plaintiff’s mother. Although the case differed from Paixão insofar as the parties did not agree
to remarry and did not draw up a joint will, the Court decided that the relationship between
them was ‘unquestionably akin to marriage’ 37 and that there existed a tacit agreement in terms
of which the deceased assumed a legal duty to support the plaintiff as his own child.38
• In JT v Road Accident Fund 39 the plaintiff was the mother of the deceased, who had adopted
his biological daughter (her granddaughter) and instituted a claim for her granddaughter’s loss
of support. Despite her adoption by her grandmother, her father had nevertheless voluntarily
continued to support her. The Court decided that the deceased’s legal duty to support his child
had not been extinguished by the adoption and his voluntary assumption of the duty to support
his daughter conferred on her an enforceable right correlative to his duty of support.40
• In Fortuin v Road Accident Fund 41 the Court held that a foster mother, acting on behalf of her
foster child, may claim damages for the child’s loss of support as a result of the death of her
foster father. The Court decided that the deceased owed the foster child a legally enforceable
duty of support worthy of the law’s protection.42
• Customary unions enjoy statutory protection43 and, ‘subject to compliance with certain
statutory formalities, found a claim for loss of support arising out of the unlawful killing of a
partner’. 44 In Chitima v Road Accident Fund 45 the Court held that the surviving partner in an
unregistered customary marriage concluded in Zimbabwe has a delictual claim in South Africa
against the Road Accident Fund for loss of support arising out of the wrongful death of her
partner in a motor vehicle accident.46

The recent development of the dependant’s action highlights the flexibility of the remedy as well as
the courts’ willingness to adapt it to modern conditions.47 The rationale for the remedy is to afford
relief to those whom the deceased had a legal duty to support, even if the duty arose out of natural
law.
Brooks v Minister of Safety and Security48
In this case, the Supreme Court of Appeal refused to extend the action for loss of support
where the breadwinner was rendered incapable of supporting his son, because he was
imprisoned for the murder of his wife and daughter. The son alleged that the police had
negligently failed to deprive his father (Brooks sr.) of his firearms, despite their knowledge of
repeated incidents where Brooks sr. had threatened his family when drunk. In another case
based on the same events, the Supreme Court of Appeal had held that the police had
negligently and wrongfully failed to prevent Brooks sr. from shooting and injuring a
neighbour.49 However, the Court refused to extend the action for loss of support to this
situation, where the breadwinner was still alive and had rendered himself unable to provide
support, by committing murder and being sentenced to lengthy imprisonment.
Consider the public policy considerations for, and against, the extension of the action in
favour of the plaintiff in this case.

Minister of Safety and Security v Madyibi50

The claimant’s husband was a sergeant in the police service. He had shot and injured the
claimant and thereafter took his own life, using a State-issued firearm that had been allocated
to him as a member of the police service, for use even when not on duty. The claimant
alleged that the shooting and suicide by the deceased were negligently and wrongfully
caused by other policemen who worked with the deceased, in that they had failed to
dispossess him of his official firearm, despite having become aware, over a protracted period
of time, that he was unfit to possess a firearm. She alleged that the other policemen should
have foreseen the deceased’s harmful conduct and had a legal duty to protect her and her
minor children.
 The Supreme Court of Appeal upheld the claim and held that the police, in failing to
remove the deceased’s official firearm from him in the circumstances, was negligent and had
wrongfully caused the dependants’ loss of support.

Spouses and both minor and major children can claim for loss of support to which they had a
right.51 The duty of support is shared by both parents, and so, depending on their means,52 a child
can claim damages in respect of a mother’s death even if the father is still alive 53 provided that the
child has suffered loss.54 Spouses have a duty to support each other 55 by either earning income or
providing support services. In Union Government (Minister of Railways & Harbours) v Warneke 56
the Court held that a husband who had suffered patrimonial harm through losing the assistance and
services of his wife, could claim damages.
Children also have a duty to support their indigent parents.57 According to Oosthuizen v Stanley
58
a child’s duty to support a parent arises ‘if both parents are indigent and are unable to support
themselves and if the child is able to provide support’. In some cases plaintiffs failed because they
did not succeed in proving that they were indigent. In Anthony v Cape Town Municipality 59 the
Court recognised that a child may have a duty to support his or her parents, but held that an 11-
year-old son who had assisted his mother in hawking fish did not have a legal duty to assist his
mother, because she was not indigent.60
Indigence is a question of fact, which depends on the circumstances of each case.61 It is not
enough to prove that a person is poor, or lives on very little.62 There has to be an extreme need, or
want for the basic necessities of life.63 What constitutes the basic necessities of life will in turn
depend on the individual parent’s station in life.64 In Van Vuuren v Sam 65 the Appellate Division
gave some examples of what it regarded as basic necessities: food, clothing, housing and
medication and care in time of sickness.66 In considering the plaintiff’s financial position, courts
may also take into account the financial needs of his or her immediate family members.67
In Fosi v RAF,68 where a mother instituted a claim for loss of support following the death of
her son who had provided her with financial support, the Court sought to emphasise the particular
role that indigenous customary law plays in this context: 69
African law obligates a child who is financially able to do so to provide maintenance to
his/her needy parents. When an African (black) provides support and education to
his/her son/daughter, he/she is not only under a duty to do so on the strength of the
South African legal system, but custom also obliges such a parent. In fact, in African
tradition to bring up a child is to make for oneself an investment in that when the child
becomes a grown-up and is able to participate in the labour market, that child will
never simply forget about where he came from. That child, without being told to do so,
will make a determination (taking into account the amount he/she earns, her travelling
to and from work, food to sustain himself and personal clothing, etc) of how much he
must send home to the parents on a monthly basis. This duty is inborn and the African
child does not have to be told by anybody to honour that obligation. In fact, that is the
trend in almost all black families in rural areas including the so-called urban black
communities. In each family there would invariably be one or two sons or daughters
who is/are employed … The duty of a child to support a needy and deserving parent is
well known in indigenous/ customary law. It is observed by such children. There is
always an expectation on the part of a parent that his child will honour this duty.

In African law it is most certainly an actionable wrong on the part of the child who is
financially able not to provide support to his needy and deserving parents. Quite apart
from it being an actionable wrong, failure to maintain one’s parents by a child who is
financially able to do so is, in black traditional law, contrary to public policy (contra
bonos mores). The parent can successfully proceed civilly against such a child in
traditional courts. It is also a morally reprehensible act to fail to maintain one’s own
 parents who are in need of such maintenance. If the parents were to decide not to
lodge a complaint before the tribal court, but opt somehow to alert members of the
immediate family about this predicament, such a child would be ostracised and be
looked down upon as a person who has no ubuntu. The latter scenario is rather rare
because as stated above every African child is born with this duty consciousness
never to forget his/her roots. It is unacceptable to African traditional law that the death
of a child who is employed and who is conscious of his duty to support and sustain
his parent, should not entitle the parent who has lost such support as a result of the
untimely death of such a child consequent upon any wrongful act on the part of
anybody including an accident caused by a negligently driven motor vehicle (as in the
instant matter) to claim that support.

Taking into account that the plaintiff and the deceased in Fosi were African (black) people, the
Court held that the customary law applied, and that there was therefore a legal duty on the part of
the child to maintain and support his mother.70 As a result, the Court held the defendant liable for
the plaintiff’s loss of support. A child’s duty to support a needy parent, as recognised in Fosi,
extends also to other cultures that share African culture’s societal norms regarding the elderly.71 In
Seleka v Road Accident Fund 72 the Court followed the reasoning in Fosi and held that under
Tswana customary law children are obliged to support their parents when they are able to earn a
living – a duty that nowadays fell on both sons and daughters.73
In Osman v Road Accident Fund,74 the Court continued this line of development:
There can be no doubt that in certain cultures such as Muslim or Hindu cultures,
amongst others, there is a similar duty upon children to support their parents as that
which Dlodlo J so eloquently related in Fosi’s matter. In these communities the family
is not restricted to the nuclear family, but rather to the extended family … In these
societies there are hardly any old-age homes or places where elderly people can retire.
This is not because these communities cannot afford to build such institutions but
rather because the societal mores scorn upon children who do not take care of their
aged parents.

Grandchildren may also have a duty to support their grandparents, but not if there are children alive
who can do so.75 Likewise, grandparents may have a duty to support their grandchildren.76

23.8.4 Damages
Dependants are entitled to be placed in the position they would have been in had the breadwinner
not died.77 Compensation is for patrimonial harm only – the dependants cannot claim a solatium for
loss of companionship or grief. They have to establish that they have suffered patrimonial harm,
once they have taken both losses and benefits (such as accelerated inheritance) into account. For
example, a claim by a husband based on the death of his wife will fail if the deceased earned less
than the benefits she had derived from her and her husband’s pooled income.78 Also, where the
deceased’s estate generates sufficient income to support the dependants in full, no loss is suffered.79
The Court formulated the general principles as follows in Victor NO v Constantia Insurance Co
Ltd: 80
A dependant’s claim is limited to the actual financial loss he has suffered as a result of
the death of the person upon whom he was dependent and the measure of his
damages is the difference between the dependant’s position as a result of the loss of
support he has suffered and the position he could reasonably have been expected to
be in had the deceased not died … From such difference there falls to be deducted any
financial benefit accruing to the dependant in consequence of the breadwinner’s
death.
Precise calculation of an award for the loss of future support is not possible, but courts recognise
that a calculation on an annuity basis is an appropriate guide, based on assumptions indicated as
reasonable by the available evidence. The accuracy of the calculation depends on the soundness of
the assumptions, and these may vary from the strongly probable to the speculative.81 The purpose of
the award is to provide the dependant with a lump sum that will provide a periodic income for the
period of expected dependency, at the end of which there will be no capital left.
Loss of support that derived from income unlawfully earned by the deceased may not be
recoverable. This rule applies when the income was derived from transactions that are void and
unenforceable.82 Included in this category are criminal activities, whether morally colourless or not,
and any other activity that is contrary to good morals or public policy.83 However, evidence that
indicates a reasonable possibility that the earning capacity could in the future have been employed
lawfully, is a sufficient basis for awarding damages to dependants, even if past earnings were
illegal.84 Arguably, a more flexible approach would be appropriate for dependants’ claims for loss
of support. In such cases, an award should depend on what the deceased could legally have earned,
even if the deceased probably would have continued to earn income illegally. This is because the
focus should be on the dependants’ loss of a right to support. However, courts have on occasion
refused to award any damages to dependants in such cases.85 Alternatively, courts should at least
take into account the possibility of lawful future income; a source which was illegal in the past may
become legal in future.

23.9 Claims for loss of support based on injury to the

support provider
Courts have not been consistent in their approach to claims for loss of support based on injury to a
support provider (breadwinner). In certain cases, courts have recognised an Aquilian action for the
dependants of a married woman for patrimonial loss suffered as a result of her injury:
• A man, married in community of property, was awarded damages for ‘loss of his wife’s
services in running the boarding-house’, in Abbott v Bergman.86
• A man, married out of community of property, was awarded damages where his wife was
legally liable to contribute to the common household and was rendered unable to do so through
injury, in Plotkin v Western Assurance Co Ltd.87
• In Erdmann v Santam Insurance Co Ltd 88 the Court expressed the view that a husband should,
in principle, be allowed to claim for the loss of his wife’s services to the household, but
decided that:
there would appear to be more merit in allowing such a claim to be brought by the
spouse who suffered the physical injuries.

However, courts have also taken the opposite view.

• In De Vaal v Messing 89 the Court refused the claim of a wife and children for loss of support
suffered as a result of an injury to the husband. The Court stated that the injured breadwinner
himself must claim for loss of future income, and damages awarded to him can then be used to
support the dependants. The answer to the dependant’s claim is simply that they suffered no
harm.

PAUSE FOR Claiming for harm suffered

REFLECTION Should there not be a claim for whatever harm is suffered? This would
mean that the dependants should be allowed to claim if they lost support
 as a result of their support provider’s injury, provided that the wrongdoer
not be held liable to pay double compensation.

1 Union Government (Minister of Railways & Harbours) v Warneke 1911 AD 657; Abbott v Bergman 1922 AD 53.
2 Guardian National Insurance Co Ltd v Van Gool NO 1992 (4) SA 61 (A).
3 Union Government (Minister of Railways & Harbours) v Warneke 1911 AD 657; Abbott v Bergman 1922 AD 53;
Guardian National Insurance Co Ltd v Van Gool NO 1992 (4) SA 61 (A).
4 De Groot Inleidinge tot de Hollandsche Rechts-geleerdheid 2 ed III (1965) 34 (3).
5 Union Government v Ocean Accident & Guarantee Corporation Ltd 1956 (1) SA 577 (A); Pike v Minister of Defence
1996 (3) SA 127 (CkS).
6 Union Government v Ocean Accident & Guarantee Corporation Ltd 1956 (1) SA 577 (A).
7 Pike v Minister of Defence 1996 (3) SA 127 (CkS).
8 See, generally, Hutchison ‘Relational economic loss (or interference with contractual relations): the last hurdle’ in
Scott and Visser (Eds) Developing Delict: Essays in honour of Robert Feenstra (2001) at 133ff.
9 1956 (1) SA 577 (A) at 585B–D.
10 At 585–586.
11 Lockhat’s Estate v North British and Mercantile Insurance Co Ltd 1959 (3) SA 295 (A) at 304.
12 Pretorius v McCallum 2002 (2) SA 423 (C).
13 Commercial Union Assurance Co of SA Ltd v Mirkin 1989 (2) SA 584 (C); Hendricks v President Insurance Co Ltd
1993 (3) SA 158 (C).
14 Lockhat’s Estate v North British and Mercantile Insurance Co Ltd 1959 (3) SA 295 (A) at 304.
15 Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA).
16 2015 (5) SA 532 (GP) para 10.
17 See Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA) para 12.
18 Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA) para 14. See also Amod v
Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4 All SA 421
(SCA); 1999 (4) SA 1319 (SCA) para 7.
19 2009 (2) SA 94 (SCA).
20 In Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA) para 12:
[O]nly a dependant to whom the deceased, whilst alive, owed a legally enforceable duty to maintain
and support may sue in such an action. Put differently, the dependant must have a right which is
worthy of the law’s protection to claim such support.
21 1908 TS 575 at 583–584.
22 Jameson’s Minors v Central South African Railways 1908 TS 575.
23 1980 (2) SA 814 (A) at 837H–838B.
24 [1999] 4 All SA 421 (SCA); 1999 (4) SA 1319 (SCA).
25 Paras 19–21.
26 Para 26.
27 2003 (11) BCLR 1220 (SCA); 2004 (1) SA 359 (SCA).
28 Paras 11–26.
29 Para 37.
30 1999 (3) SA 421 (SCA) at 427–431.
31 [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA).
32 Para 29.
33 Para 39.
34 Para 39.
35 Para 40. See also Verheem v RAF 2012 (2) SA 409 (GNP).
36 2016 (1) SA 214 (GJ).
37 Paras 10–13.
38 Para 15.
39 2015 (1) SA 609 (GJ).
40 Paras 29–31.
41 2015 (5) SA 532 (GP).
42 Paras 9 and 13.
43 See section 31 of the Black Laws Amendment Act 76 of 1963 and section 2 of the Recognition of Customary
Marriages Act 120 of 1998.
44 Chitima v Road Accident Fund [2012] 2 All SA 632 (WCC) para 14.
45 [2012] 2 All SA 632 (WCC).
46 Para 32.
47 See also Amod v Multilateral Motor Vehicle Accidents Fund (Commission for Gender Equality Intervening) [1999] 4
All SA 421 (SCA); 1999 (4) SA 1319 (SCA); Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA
377 (SCA).
48 2009 (2) SA 94 (SCA) para 6.
49 Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA).
50 [2010] 1 All SA 139 (SCA).
51 Bursey v Bursey 1999 (3) SA 33 (SCA).
52 Jodaiken v Jodaiken 1978 (1) SA 784 (W) at 788–789; Fourie v Santam Insurance Ltd 1996 (1) SA 63 (T) at 65.
53 Senior NO v National Employers General Insurance Co Ltd 1989 (2) SA 136 (W); Ismail v General Accident
Insurance Co SA Ltd 1989 (2) SA 468 (D).
54 Santam Insurance Co Ltd v Fourie 1997 (1) SA 611 (A) at 615–616.
55 Witham v Minister of Home Affairs 1989 (1) SA 116 (ZH).
56 1911 AD 657.
57 See Pike v Minister of Defence 1996 (3) SA 127 (CkS) at 132; Kotwane v Unie Nasionaal Suid-Britse
Versekeringsmaatskappy Bpk 1982 (4) SA 458 (O).
58 1938 AD 322 at 327–328; Fosi v RAF 2008 (3) SA 560 (C) para 3.
59 1967 (4) SA 445 (A).
60 See also Petersen v South British Insurance Co Ltd 1964 (2) SA 236 (C).
61 Oosthuizen v Stanley 1938 AD 322 at 327–328; Smith v Mutual & Federal Insurance Co Ltd 1998 (4) SA 626 (C) at
629.
62 Smith v Mutual & Federal Insurance Co Ltd 1998 (4) SA 626 (C) at 632.
63 Oosthuizen v Stanley 1938 AD 322 at 327–328; Smith v Mutual & Federal Insurance Co Ltd 1998 (4) SA 626 (C) at
632.
64 Fosi v RAF 2008 (3) SA 560 (C) para 13; Burger v Die Padongelukkefonds case No 2223/1999 (unreported).
65 1972 (2) SA 663 (A) at 642. See also Fosi v Road Accident Fund 2008 (3) SA 560 (C) para 13.
66 See also Oosthuizen v Stanley 1938 AD 322 at 328.
67 Singh v Santam Insurance Co 1974 (4) SA 196 (D).
68 2008 (3) SA 560 (C).
69 Paras 16–17.
70 Paras 24–25.
71 Osman v Road Accident Fund 2015 (6) SA 74 (GP).
72 2016 (4) SA 445 (GP).
73 Paras 14–15 and 18–20.
74 2015 (6) SA 74 (GP) paras 20–21.
75 Barnes v Union and SWA Insurance Co Ltd 1977 (3) SA 502 (E).
76 Petersen v Maintenance Officer, Simon’s Town Maintenance Court 2004 (2) SA 56 (C).
77 Legal Insurance Co Ltd v Botes 1963 (1) SA 608 (A) at 614; Groenewald v Snyders 1966 (3) SA 237 (A) at 246;
Constantia Versekeringsmaatskappy Bpk v Victor NO 1986 (1) SA 601 (A) at 611; Lambrakis v Santam Ltd 2002 (3)
SA 710 (SCA) para 12; Paixão v Road Accident Fund [2012] 4 All SA 262 (SCA); 2012 (6) SA 377 (SCA) para 12.
78 Santam Insurance Co Ltd v Fourie 1997 (1) SA 611 (A).
79 Lambrakis v Santam Ltd 2002 (3) SA 710 (SCA) para 19; and see Mqolomba v RAF [2002] 4 All SA 214 (Tk), where
the benefits of a trust fund were taken into account in assessing whether harm had been suffered.
80 1985 (1) SA 118 (C) at 120C–D.
81 Lambrakis v Santam Ltd 2002 (3) SA 710 (SCA) para 14.
82 Dhlamini v Protea Assurance Co Ltd 1974 (4) SA 906 (A) at 915.
83 Dhlamini v Protea Assurance Co Ltd 1974 (4) SA 906 (A) at 915; Mba v Southern Insurance Association Ltd 1981 (1)
SA 122 (TkS) at 124.
84 See Lebona v President Versekeringsmaatskappy Bpk 1991 (3) SA 395 (W) at 405; Dhlamini v Multilaterale
Motorvoertuigongelukkefonds 1992 (1) SA 802 (T) at 806, where the evidence indicated what the deceased would
have earned in lawful employment.
85 Booysen v Shield Insurance Co Ltd 1980 (3) SA 1211 (SE); Santam Insurance Ltd v Ferguson 1985 (4) SA 843 (A);
Mankebe NO v AA Mutual Insurance Association Ltd 1986 (2) SA 196 (D).
86 1922 AD 53 at 56.
87 1955 (2) SA 385 (W) at 394–395.
88 1985 (3) SA 402 (C) at 406 and 408–409.
89 1938 TPD 34.
 Chapter 24

Pain and suffering

A person who suffers bodily injury can recover not only the patrimonial loss that results from the
injury, but also compensation for pain and suffering. The action for pain and suffering developed in
Roman-Dutch law, probably under the influence of customary Germanic law, as an additional
remedy to recover a sum of money as solace (solatium) for injury. The action for pain and suffering
in common law was neither actively nor passively transmissible, which means that it did not pass to
the estate of the claimant after the claimant’s death and was not available against the estate of the
defendant after the defendant’s death, unless the action had progressed to the point of being
formally ready to proceed to trial (the time of litis contestatio).
In the modern South African law of delict, the action for pain and suffering is still actively
non-transmissible (it does not pass to the estate of the claimant after death), unless litis contestatio
has taken place. In modern law, this refers to the procedural stage when pleadings are closed in
terms of the rules of court and the action is ready to proceed to trial.1 The reason for active non-
transmissibility is the highly personal nature of the action, which serves to provide solace to the
victim, and not to the victim’s estate or heirs. However, | in modern law, the action is passively
transmissible, which means that the action remains available against the estate of the defendant
after the defendant’s death.2 Once litis contestatio has taken place, the action can in any event be
heard by a court regardless of the death of any of the parties, because by that stage one can regard
the parties as having agreed that the action is ready to be adjudicated by the Court.
The purpose of an award for pain and suffering is to provide solace for physical pain or
discomfort experienced as a result of injury, as well as for psychological harm in the form of shock
caused by the injury, or distress due to disfigurement, loss of amenities (for example, loss of the
ability to participate in sport), or a shortened life expectancy.3 Harm in the form of pain and
suffering does not have an economic value,4 but courts nevertheless award an appropriate amount
of money, the purpose of which is to provide solace to the victim, and not to serve as punishment
for the wrongdoer. The purpose of criminal law to punish wrongdoers, and this area of law has
certain protections built into it for the benefit of accused persons (the presumption of innocence, the
right against self-incrimination, and a higher standard of proof). It would be inequitable (not to
mention unconstitutional) to punish a person without affording him or her these protections.
The Constitutional Court has confirmed, in Fose v Minister of Safety and Security 5 and in
Dikoko v Mokhatla,6 that the general aim of an award of damages in delict is to vindicate rights and
to console, and not to punish the wrongdoer. This also applies to the action for pain and suffering.
In Collins v Administrator, Cape 7 the Court held that an award of damages for pain and suffering
to an unconscious victim, even if it was a nominal award, is inappropriate, because an unconscious
victim cannot experience consolation. Such an award would mean the unjustified importing of a
penal element into the modern law of delict.8

1 Jankowiak v Parity Insurance Co Ltd 1963 (2) SA 286 (W).

2 Solomon v De Waal 1972 (1) SA 575 (A). In this case, an award for pain and suffering was made against the deceased
estate of the owner of a horse, which had injured the plaintiff. The award was made under the actio de pauperie.
3 Hoffa NO v SA Mutual Fire & General Insurance Co Ltd 1965 (2) SA 944 (C) at 954–955:
The damages awarded to him are in a certain sense analogous to the solatium which is awarded under
the actio injuriarum to someone as a salve to his wounded feelings.
4 Mutual & Federal Insurance Co Ltd v Swanepoel 1988 (2) SA 1 (A) at 10–11; Sandler v Wholesale Coal Suppliers
Ltd 1941 AD 194 at 199.
5 1997 (3) SA 786 (CC) para 63.
6 2006 (6) SA 235 (CC).
7 1995 (4) SA 73 (C) at 93–94.
8 The Court rejected the abstract approach adopted in Gerke NO v Parity Insurance Co Ltd 1966 (3) SA 484 (W), where
an award for ‘loss of happiness’ was made, even though the plaintiff could not realise his loss.
 Chapter 25

Emotional shock

25.1 What constitutes psychological harm or emotional shock?

25.2 Why is it a problem area?

25.3 How did the law develop?

25.4 When will psychological harm justify an award of damages?

25.5 What can be the causes of psychological harm?

25.6 What is the appropriate action?

|
25.7 What is the appropriate criterion for liability?

25.1 What constitutes psychological harm or emotional

shock?
An injury to the brain or nervous system that results from nervous or emotional shock is also a form
of physical or bodily injury, because the nervous system is as much a part of the body as bones and
muscles. In Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1 the Court had to
decide the nature of emotional shock because the plaintiff’s claim was based on legislation that
provided for an award of damages for ‘bodily injury’ caused by, or resulting from driving a motor
vehicle. The Road Accident Fund Act 56 of 1996 2 provides for an award of damages for bodily
injury, but, as of 1 August 2008 the Act specifically excluded claims in terms of the Act for
‘emotional shock sustained by that person when that person witnessed or observed or was informed
of the bodily injury or the death of another person as a result of the driving of a motor vehicle’. 3
Plaintiffs must now bring a claim for damages in delict under the common law.4 In common law,
liability for damages based on psychiatric injury caused by nervous or emotional shock requires
special consideration.

25.2 Why is it a problem area?

On account of various policy considerations, courts approach claims based on nervous or emotional
shock with caution. The effects of shock may be much more widespread than the effects of
ordinary physical injury, and to impose liability for causing shock may therefore place too heavy a
burden upon an individual defendant.5 Courts have also noted that mental injuries cannot readily be
measured in terms of money,6 and that a person may simulate mental injuries, which creates the
possibility of an excess of illegitimate and fanciful claims.7 The concept of injury caused by
nervous or emotional shock is also a wide one. It potentially includes temporary emotions such as
grief or fear, for which courts have consistently refused to award damages.8 It is now settled law
that courts can award damages for nervous or emotional shock that causes a recognised form of
psychiatric injury.9 The criteria for imposing this kind of liability are set out in the following
sections.

25.3 How did the law develop?

At first courts imposed a number of limiting requirements for liability, for example, courts required
that physical injury must have caused the psychological harm.10 Another limiting requirement was
that the person who suffered the psychological harm must have been in personal danger of ordinary
physical injury, as in Mulder v South British Insurance Co Ltd.11 In this case, the Court found that a
child had been negligently killed when he was run over by a bus but there was no duty towards the
child’s mother who suffered emotional shock as a result of witnessing the incident. The Appellate
Division in Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 12 discarded this
restrictive requirement, and established that in our law the criterion of liability for psychological
harm caused by shock is the reasonable foreseeability of such harm. Formerly, courts awarded
damages only to persons who personally experienced the trauma-producing event, but now courts
also impose liability in ‘hearsay cases’, where the trauma is the result of a report of one or other
tragic event.13 Courts used to recognise liability only for psychological harm that resulted from fear
for a person’s own bodily integrity, or that of a relative. In Road Accident Fund v Sauls 14 the Court
said that the issue is one of reasonableness and awarded damages to a fiancée who had a live-in
relationship with a person injured in an accident that she had witnessed. The closer the relationship,
the more likely it will be that the psychological harm will be regarded as foreseeable.15 The
principle that persons must take their victims as they find them applies, and if injury through shock
was foreseeable, then that person is liable for all the loss, even if the loss was aggravated by a pre-
existing condition such as high blood pressure (hypertension).16

25.4 When will psychological harm justify an award of

damages?
For an award of damages, the plaintiff must prove that the shock caused a physical reaction such as
stroke leading to death;17 miscarriage;18 high blood pressure, trembling and collapse;19 or a
detectable and recognised psychiatric injury or lesion that is not passing or trivial,20 such as an
anxiety neurosis,21 acute depression,22 mixed anxiety depressive disorder,23 post-traumatic stress
disorder,24 impaired sleep,25 or emotional trauma.26 In Barnard v Santam Bpk 27 Van Heerden DCJ
said that the term ‘senuskok’ (nervous shock) is not only an obsolete term without any specific
psychiatric meaning, but it may also be misleading. The only relevant question should be whether
the plaintiff sustained a recognisable psychological lesion (‘psigiese letsel’).28 Supporting
psychiatric evidence should, as a rule, prove that such a lesion exists.29 This follows the well-known
statement by Lord Denning MR in Hinz v Berry: 30
In English law no damages are rewarded for grief and sorrow caused by a person’s death.
No damages are to be given for the worry about the children, or for the financial strain or
stress, or the difficulties of adjusting to a new life. Damages are however recoverable for
nervous shock, or, to put it in medical terms, for any recognisable psychiatric illness caused
by the breach of duty by the defendant.

PAUSE FOR Bullying is not a new phenomenon, but society has become increasingly
REFLECTION aware of its harmful side effects. Also, the focus is no longer only on
physical bullying. In contemporary society, individuals are spending more
time engaging in online social interaction, and one of the unpleasant side
effects is the increase in incidents of online bullying, or ‘cyberbullying’.
Victims of both physical bullying and cyberbullying report a range of
forms of psychological harm, such as depression and low self-esteem.
Some victims of cyberbullying have even committed or attempted to
commit suicide. Do you think a victim of physical or cyberbullying should
be able to claim damages for the emotional harm he or she has suffered?

25.5 What can be the causes of psychological harm?

There is no closed list of causes, and courts can impose liability for any conduct that intentionally
or negligently causes psychological harm. Categories that courts have recognised include fear for
one’s own safety; 31 and fear for the safety of a relative or loved one, such as a brother,32 a wife and
child,33 a fiancée,34 or a son found lying in the street after an accident.35 Shock and psychological
harm can be caused by witnessing a gruesome or disturbing accident, for example, a mother seeing
the wheel of a bus drive over her little son’s head; 36 a passenger witnessing the death of a co-
passenger in a bus accident; 37 a wife and mother seeing her husband and children being grievously
injured in a car accident;38 a young woman seeing her fiancé being hit by a car and lying in the
road, seemingly fatally injured; 39 or a pregnant woman hearing a collision between a motorcycle
and a car and seeing the victim’s blood on the road.40 Successful plaintiffs are mostly persons who
personally experienced the trauma-producing event, but courts have also recognised claims in
‘hearsay cases’, where the trauma is the result of a report of one or other tragic event, for example,
where the victim learns afterwards of the injury or death of a loved one, as in the case of a wife
learning of the death of her husband,41 or a mother being told about the death of her son.42 Plaintiffs
can also recover damages for trauma induced by a disturbing experience unrelated to injury or
death, as in the case of a mother learning that the child she has raised for two years is not her own,
but another person’s child, swapped with hers in the maternity ward.43 Shock and psychological
harm can result from both a fear for the safety of loved ones and an anticipation of damage to
property.44

25.6 What is the appropriate action?

• Where the claim is for psychological harm only, without other physical injury, the plaintiff
must prove a detectable and recognised psychiatric injury or lesion (‘psigiese letsel’) that is not
passing or trivial.45 Where the psychological harm is accompanied by other physical injury,
courts simply assess the extent of the pain, suffering or other psychological harm, and make an
appropriate award. They do not require that the pain and suffering must constitute a
recognisable psychological lesion.
• The action for pain and suffering is also available where the harm was caused intentionally.
For example, in Boswell v Minister of Police 46 the plaintiff succeeded with a claim for general
damages through the action for pain and suffering on the ground of intentional infliction of
emotional shock. A policeman had deliberately provided false, distressing information to the
plaintiff, by telling her that her nephew had been shot. The plaintiff collapsed and lost
consciousness as a result of emotional shock. Furthermore, she felt weak, had a headache and
suffered from high blood pressure. She felt shaky for a month after the policeman inflicted the
shock. A physician testified that the shock would have ‘a substantial effect on her health’. 47
The Court held that the action for pain and suffering was appropriate in instances where a
defendant acted intentionally.48
• An alternative remedy for psychological harm caused by intentionally infringing a personality
right is the actio iniuriarum. Intentional infliction of emotional shock is an iniuria actionable
under the actio iniuriarum. In Waring & Gillow Ltd v Sherborne,49 after rejecting an action
based on negligently causing emotional shock, the Court stated:
It would be different, under certain circumstances, in an actio iniuriarum based upon a
wilful attack upon or violation of the feelings of another. In such a case it might be
possible to award compensation for the outrage of the feelings or the insult to the
honour.
• In Els E v Bruce, Els J v Bruce 50 the Court dealt directly with emotional shock that was
intentionally caused and held that the actio iniuriarum was the appropriate remedy where the
defendant adversely affected the victim’s health by addressing her in an insulting and
threatening manner. The defendant in the Boswell case, referred to above, clearly acted
intentionally. The actio iniuriarum would have been an appropriate action for the recovery of
non-patrimonial damages for psychological harm, but the plaintiff’s particulars of claim were
not framed to include damages intentionally caused. The Court awarded damages on the basis
of harm that was caused negligently, under the action for pain and suffering. However, the
principles of the actio iniuriarum are broad enough to accommodate cases of intentional
infliction of emotional shock.
• Where the psychological harm is accompanied by patrimonial loss, plaintiffs can recover
damages for the patrimonial loss using the Aquilian action. In Bester v Commercial Union
Versekeringsmaatskappy van SA Bpk 51 the victim, a minor, suffered only psychological harm
as a result of the negligent infliction of emotional shock, but no patrimonial loss.
His father succeeded in recovering compensation on his behalf with the action for pain and
suffering. The father himself suffered patrimonial loss, which he recovered using an Aquilian
action.
• The dependants of a breadwinner who died as a result of the infliction of severe shock can also
institute an action for loss of support.52

25.7 What is the appropriate criterion for liability?

In Bester v Commercial Union Versekeringsmaatskappy van Suid-Afrika Bpk 53 the Appellate
Division held that liability for psychological harm must be determined on the basis of reasonable
foreseeability, the question being whether the general nature of the harm to the plaintiff(the
psychological harm) and the general manner of its occurrence are foreseeable. Courts in subsequent
cases have continued to endorse the foreseeability test for establishing liability. However, the
Supreme Court of Appeal has pointed out that the concept of reasonable foreseeability is not
logically well suited for taking into account policy considerations, such as the range of persons who
should be protected against psychological harm and the possibility of indeterminate liability.54
In some cases, courts have considered policy issues that relate to limiting liability for
psychological harm under the heading of legal causation.55 According to the modern approach to
legal causation, the overriding question is whether there was a sufficiently close relationship
between the wrongdoer’s negligent act and the psychological harm, for courts to impute such harm
to the wrongdoer, taking into account policy considerations based on reasonableness, fairness and
justice. The foreseeability of harm is now a subsidiary factor, although an important one.
Another possible approach is to take the relevant policy considerations into account when
assessing wrongfulness, which allows scope for considering the pertinent issues, such as the range
of persons who should be protected against psychological harm and the possibility of indeterminate
liability. In terms of this approach, one should not regard the factual causing of psychological harm
in itself as indicating wrongfulness. Causing psychological harm is only wrongful if it is considered
to be unreasonable with regard to factors such as the seriousness of the harm, the plaintiff’s
apprehension of personal danger, intent on the part of the defendant, and the relationship between
the primary and secondary victims.

PAUSE FOR Causing psychological harm

• Is the criterion of reasonable foreseeability an appropriate test for liability for
REFLECTION
causing psychological harm? Reasonable foreseeability is one part of the
negligence test, and depends on factors such as the risk of harm occurring and
the extent of the possible harm. Why is this concept not well suited for taking
into account policy factors, such as the range of persons who should be
protected against psychological harm and the possibility of indeterminate
liability?
• Why is wrongfulness better suited for taking into account such policy factors?
What other factors may be relevant in determining the wrongfulness of causing
psychological harm? (Refer to the chapter on wrongfulness.)
• Is the concept of legal causation well suited for taking into account policy
factors, such as the range of persons who should be protected against
psychological harm and the possibility of indeterminate liability? Can
reasonable foreseeability be relevant to either negligence or legal causation
(Refer to the chapter on legal causation.)
• In Barnard v Santam Bpk56 Van Heerden AHR remarked57 that from a practical
point of view, it does not make any difference whether foreseeability is
assessed for the purposes of negligence or legal causation. Do you agree?
• Consider also the following statement:58

There is no need to adhere to rigid categories for

determining liability, especially when it makes no difference
which element of liability is used. Courts should have
flexible criteria at their disposal when attempting to resolve
difficult issues and provided that the basic principles do
not conflict, there appears to be no harm in choosing which
of the two elements, wrongfulness or causation, is the more
suitable tool for dealing with a particular issue.
1 1973 (1) SA 769 (A).
2 As amended by the Road Accident Fund Amendment Act 19 of 2005.
3 Section 19(g) of the Act provides that the Fund or one of its agents is not obliged to compensate any person for any
loss or damage:
suffered as a result of an emotional shock sustained by that person when that person witnessed or
observed or was informed of the bodily injury or the death of another person as a result of the driving
of a motor vehicle.
Paragraph (g) was added to the Act by section 8(b) of the Road Accident Fund Amendment Act 19 of 2005. The
amendment came into effect on 1 August 2008.
4 Section 21(1) provides that:
no claim for compensation in respect of loss or damage resulting from bodily injury to or the death of
any person caused by or arising from the driving of a motor vehicle shall lie against the owner or
driver of a motor vehicle, or against the employer of the driver.
However, section 21(2)(b) provides that section 21(1) does not apply:
to an action for compensation in respect of loss or damage resulting from emotional shock sustained
by a person, other than a third party, when that person witnessed or observed or was informed of the
bodily injury or the death of another person as a result of the driving of a motor vehicle.
5 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1972 (3) SA 68 (D) at 74D–E.
6 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1972 (3) SA 68 (D) at 73D–G.
7 Union Government (Minister of Railways & Harbours) v Warneke 1911 AD 657.
8 Barnard v Santam Bpk 1999 (1) SA 202 (SCA) at 217. In Muzik v Canzone Del Mare 1980 (3) SA 470 (C), the
plaintiff experienced discomfort after eating poisonous food in a restaurant, but failed to prove any mental or physical
harm; and in Lutzkie v South African Railways and Harbours 1974 (4) SA 396 (W), the plaintiff sustained no
‘permanent consequences’ as a result of shock suffered when witnessing the death of a boy in a bus accident. Hing v
Road Accident Fund 2014 (3) SA 350 (WCC) held that our law has not evolved to the extent that any claim for grief
or sorrow could be recognised. Such damages can also not be claimed under the head of loss of support: MacDonald v
Road Accident Fund [2012] 4 All SA 15 (SCA).
9 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A).
10 Hauman v Malmesbury Divisional Council 1916 CPD 216 at 220; Bester v Commercial Union
Versekeringsmaatskappy van SA Bpk 1972 (3) SA 68 (D) at 73D–F.
11 1957 (2) SA 444 (W).
12 1973 (1) SA 769 (A).
13 Barnard v Santam Bpk 1999 (1) SA 202 (SCA).
14 2002 (2) SA 55 (SCA).
15 Barnard v Santam Bpk 1999 (1) SA 202 (SCA) at 215.
16 Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C) at 342–343.
17 Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C).
18 Waring & Gillow Ltd v Sherborne 1904 TS 340; Hay or Bourhill v Young [1942] 2 All ER 396.
19 Boswell v Minister of Police 1978 (3) SA 268 (E).
20 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A) at 779; Barnard v Santam Bpk
1999 (1) SA 202 (SCA).
21 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A) at 779 and 782.
22 Majiet v Santam Ltd [1997] 4 All SA 555 (C).
23 Clinton-Parker v Administrator, Tvl; Dawkins v Administrator, Tvl 1996 (2) SA 37 (W).
24 Road Accident Fund v Sauls 2002 (2) SA 55 (SCA); Minister of Justice and Constitutional Development v X [2014] 4
All SA 586 (SCA); 2015 (1) SA 25 (SCA); Naidoo v Minister of Police [2015] 4 All SA 609 (SCA) at 152 para 35.
25 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A); Road Accident Fund v Sauls
2002 (2) SA 55 (SCA).
26 Barnard v Santam Bpk 1999 (1) SA 202 (SCA).
27 1999 (1) SA 202 (SCA).
28 The definition in Swartbooi v RAF 2013 (1) SA 30 (WCC); [2012] 3 All SA 670 (WCC) para 17 (‘Emotional shock is
defined as shock suffered by a person without necessarily personally sustaining bodily injury.’) is not entirely
accurate.
29 Barnard v Santam Bpk 1999 (1) SA 202 (SCA) at 216–F.
30 1970 (2) QB 40 at 42–43; [1970] 1 All ER 1074 (CAC).
31 Hauman v Malmesbury Divisional Council 1916 CPD 216.
32 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A).
33 Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C).
34 Road Accident Fund v Sauls 2002 (2) SA 55 (SCA).
35 Majiet v Santam Ltd [1997] 4 All SA 555 (C).
36 Mulder v South British Insurance Co Ltd 1957 (2) SA 444 (W).
37 Lutzkie v South African Railways and Harbours 1974 (4) SA 396 (W).
38 Hinz v Berry [1970] 1 All ER 1074 (CAC).
39 Road Accident Fund v Sauls 2002 (2) SA 55 (SCA).
40 Hay or Bourhill v Young [1942] 2 All ER 396.
41 Waring & Gillow Ltd v Sherborne 1904 TS 340.
42 Barnard v Santam Bpk 1999 (1) SA 202 (SCA).
43 Clinton-Parker v Administrator, Tvl; Dawkins v Administrator, Tvl 1996 (2) SA 37 (W).
44 In Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C), the deceased was subjected within minutes to a
sequence of stress-causing events, being himself assaulted, seeing his wife being hit by a car and witnessing his own
car with occupants in it, including his child, being struck by another car.
45 Bester v Commercial Union Versekeringsmaatskappy van SA Bpk 1973 (1) SA 769 (A) at 779; Barnard v Santam Bpk
1999 (1) SA 202 (SCA).
46 1978 (3) SA 268 (E) at 275A–G.
47 At 272B.
48 At 273A–B.
49 1904 TS 340 at 348.
50 1922 EDL 295 at 298–299.
51 1973 (1) SA 769 (A).
52 Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C) at 343G–344A. One cannot include a claim for grief or
sorrow in an action for loss of support: MacDonald v Road Accident Fund [2012] 4 All SA 15 (SCA).
53 1973 (1) SA 769 (A).
54 Barnard v Santam Bpk 1999 (1) SA 202 (SCA) at 212–213.
55 Barnard v Santam Bpk 1999 (1) SA 202 (SCA) at 209–210; Road Accident Fund v Sauls 2002 (2) SA 55 (SCA) para
8; see also Masiba v Constantia Insurance Co Ltd 1982 (4) SA 333 (C) at 342.
56 1999 (1) SA 202 (SCA) at 215.
57 At 210.
58 Midgley ‘The role of foreseeability in psychiatric injury causes’ (1992) 55(3) THRHR 441 at 445.
 Chapter 26

Infringements of bodily integrity

26.1 Introduction

26.2 Infringements of one’s corpus

26.2.1 Assault
26.2.2 Deprivation of personal freedom
26.2.3 Seduction

|
26.1 Introduction
The general remedy for the infringement of personality rights is the actio iniuriarum. Its main aim
is to protect plaintiffs against wrongful and intentional infringements of these rights and, where
they are violated, to provide for the recovery of damages. There are three basic elements for an
action under the actio iniuriarum: (a) an infringement of a personality right in a (b) wrongful and
(c) intentional way.1
In Roman law, and then later in Roman-Dutch law, the actio iniuriarum protected
infringements of corpus (bodily integrity), dignitas (dignity) and fama (reputation). Modern law
has added the rights to privacy and identity to this list. The common-law concept of dignitas is a
very broad one, encompassing dignity in its ordinary dictionary meaning, privacy and identity. The
right to identity is a particular manner in which the right to privacy can be infringed, and is
therefore also an infringement of the right to dignity in a broad sense. Any factual violation of
corpus, dignitas or fama constitutes the harm element that forms the basis of the action.
Although each of these rights is recognised as a single, clearly defined interest of personality in
our law, overlaps do occur. Violations of one of these rights could also involve impairments of any
of the others. For example, an assault could simultaneously involve both an infringement of bodily
integrity and an impairment of dignity.
It should also be noted that the same source or event can simultaneously give rise to several
claims in delict under the various actions. For example, John assaults Vusi in the workplace, in
front of Vusi’s colleagues. As a result of the assault, Vusi cannot work for a week and receives
medical treatment at a hospital. Assuming Vusi is on a ‘zero-hours’ contract and is paid only for the
work he does and has no medical-aid entitlement, Vusi would potentially have three claims arising
from this incident. He could claim special damages for patrimonial loss arising from his loss of
earnings and medical expenses (under the Aquilian action), general damages for pain and suffering
experienced by him during and after the assault (under the action for pain and suffering), as well as
general damages for the impairment of his dignity (under the actio iniuriarum), because the assault
took place in front of his colleagues and was a demeaning experience. Vusi would not bring three
separate claims in delict; he would bring a single claim before the courts by means of a ‘rolled-up’
action.2
Similarly, an attack on a person’s reputation often impairs that person’s dignity as well. This
notion was confirmed in Khumalo v Holomisa,3 where O’Regan J pointed out:
In our new constitutional order, no sharp line can be drawn between these injuries to
personality rights … The value of human dignity in our Constitution therefore values both
the personal sense of self-worth as well as the public’s estimation of the worth or value of an
individual.

TERMINOLOGY Iniuria
Voet 47.10.14 refers to an iniuria as an act committed in contempt
of another’s personality. In fact, the term has three possible
meanings:
• Anything contrary to justice and equity – that is, wrongfulness
(dictionary meaning). This covers all delictual situations, irrespective of
whether the harm is patrimonial or non-patrimonial.
• The wrongful and intentional impairment of a person’s personality rights
of bodily integrity (corpus), dignity (dignitas) or reputation (fama). This
covers only those situations that fall within the actio iniuriarum.
• Impairment of dignity in the form of an insult where contumelia
(humiliation) is present. This covers only those situations that fall within
the narrow meaning of the dignity concept.

26.2 Infringements of one’s corpus

The right to bodily integrity protects a person’s physical and mental integrity – that is, it supports
the notion of the mind being part of the physical body.5 Ways in which a person can violate that
right often involve an assault 6 or other manifestations of an assault, such as rape,7 or infecting a
partner with the human immunodeficiency virus.8 The violation does not have to be a violent one –
a plaintiff could also found an action upon an act of seduction 9 (although possibly outmoded and
unconstitutional), a surgical operation,10 taking a blood sample,11 or depriving a detainee of exercise
and basic intellectual stimulation and psychological comforts.12

26.2.1 Assault

PAUSE FOR Assault

REFLECTION When a person claims for damages for failed surgical operations, the
cause of action is often framed as a violation of bodily integrity in the form
of an assault. Is it conceptually correct to do this? See Broude v
McIntosh.13
Interfering with the right to bodily integrity is not necessarily actionable in law. For example,
although accidentally pushing a person in a crowded lift or a crowded stadium is a factual
infringement of a person’s bodily integrity, conduct of this nature is ordinarily not wrongful (on the
basis of the principle de minimis non curat lex) and the intention element would also not be
satisfied (the person factually interfering with the other person’s right to bodily integrity would not
have directed his or her will towards bringing about the wrongful consequence if the interference
was merely accidental, and he or she would in any event in such circumstances not have intended to
act wrongfully).

Bennett v Minister of Police14

A policeman ordered Bennett, for no reason, to get into a police van. When Bennett refused
to do so, and resisted attempts to put him into the van by clinging to the wire mesh
surrounding the van’s passenger compartment, the policeman used a wooden police baton to
beat Bennett’s hands loose. He broke two fingers on Bennett’s right hand. Bennett did not
suffer any patrimonial loss in the form of medical expenses, or any loss of earnings, because
he was laid off from work for two months on full pay. He was nonetheless awarded general
damages for the pain and discomfort that resulted from the incapacitation of his hand.
Bennett also alleged an iniuria in that he had also suffered an infringement of his dignity in
the form of humiliation. The Court found ‘that the assault on Bennett was violent,
unprovoked, unnecessary and excessive’,15 and even though he had been resisting arrest, it
constituted an unlawful and intentional impairment of Bennett’s personality. The Court found
that the difficulty with this claim was that Bennett had not alleged that he had in fact been
humiliated: ‘There is a very large subjective element in any injuria, and if plaintiff feels
aggrieved in his dignity he must say so.’16 The Court thereafter considered whether
McKerron’s statement17 that ‘assaults of all kinds’ could be injurious and concluded:18

The fact is that not all assaults necessarily involve contumelia. It depends upon
the circumstances. A policeman who unlawfully shoots a person does not
normally impair that person’s dignity; a robber who stabs his victim does not
normally insult the victim by so doing. But an assault by a policeman with a
baton is probably on a different footing. I would think that this does involve a
measure of contumelia objectively regarded.

As a result, Bennett received general damages for physical injuries amounting to R600 and R50 for
the ‘objectively-regarded contumelia’.
The judgment raises a number of issues:
• An infringement of the personality right to bodily integrity may give rise to separate claims
under the Aquilian action (for patrimonial loss), the Germanic remedy (for pain and suffering),
and the actio iniuriarum (for infringement of dignity). See also April v Minister of Safety and
Security,19 where the Court recognised that claims for assault involve issues that relate to
bodily injury and pain and suffering, as well as those that involve contumelia. Even though
these issues stem from the same source or event, the claims remain distinct. The Court noted
that the award for contumelia was different from general damages ordinarily awarded in cases
of bodily injury. Nonetheless, this does not mean that they involve the invasion of a different
personality right, as the Court suggested.20
• Did the Court award damages under the actio iniuriarum for the mere infringement of bodily
integrity? It appears not: R50 was awarded for the infringement of dignity, and R600 for pain
and suffering.
• Could there be an interrelationship between the general damages under the Germanic remedy
and any claim for infringement of bodily integrity? In other words, could the Germanic
remedy provide the necessary solace and so make a claim under the actio iniuriarum in respect
of bodily integrity redundant?
• Dignity appears to have both a subjective and an objective component. In this instance, only
the latter was alleged and proven.
• The Court did not hold that all use of force during an arrest would constitute an actionable
assault. It found that in this instance, there was unlawful conduct because the force was
unnecessary and excessive, and so:
the policeman using the force steps out from under the umbrella of protection which the
law holds over him as long as he is using only the permissible degree of force.21
• The Court also raised the issue of proof:
The allegation of assault is an allegation of an unlawful inroad upon Bennett’s right to
the integrity of his personality and the animus is sufficiently alleged by the allegation of
the unlawful assault.22
• Later the Court noted that:
proof of the unlawful aggression raises the presumption that the aggressor was animated
by animus injuriandi which presumption he must rebut if he wishes to escape liability.23
• Courts have not been consistent in dealing with presumptions under the actio iniuriarum. We
suggest elsewhere why this is so and how some consistency could be achieved.24

26.2.2 Deprivation of personal freedom

A person’s right to bodily integrity is also violated where a person’s personal freedom is restricted
or a person is deprived of liberty – for example, in an arrest 25 or an incarceration.26 These instances
are commonly referred to as either wrongful arrest and detention, or malicious arrest and detention.
The rationale for liability is set out in Thandani v Minister of Law and Order: 27
… sight must not be lost of the fact that the liberty of the individual is one of the
fundamental rights of a man in a free society which should be jealously guarded at all times
and there is a duty on our courts to preserve this right against infringement.

The essence of wrongful arrest and detention lies in (a) intentionally (b) depriving a person of his
or her liberty or physical freedom (c) without lawful justification. In other words, there must be a
wrongful and intentional interference with the person’s bodily freedom. This is no different from
the usual elements of the actio iniuriarum – namely, intention, harm in the form of a factual
infringement of a right, and wrongfulness. However, because of the significant value placed on
individual liberty in a democratic legal culture, these elements have developed certain peculiarities.
For wrongful arrest and wrongful detention, the plaintiff must allege and prove that the
defendant, or an agent acting on the defendant’s behalf, committed the act that led to the plaintiff’s
physical freedom being restricted (the factual interference with the right to bodily integrity by
depriving the person of his or her personal freedom). Arrest is frequently followed by detention,
and if the initial arrest is unlawful, the subsequent detention will also be wrongful.
• An arrest constitutes an interference with the liberty of the individual, and once the plaintiff
has established, on a balance of probabilities, that his or her bodily freedom has been factually
infringed, wrongfulness is presumed and the onus shifts to the defendant to justify why the
deprivation of liberty took place.28 For example, section 40(1)(b) of the Criminal Procedure
Act 51 of 1977 (dealing with arrests without a warrant) obliges an arrestor, among other
requirements, to prove that he or she suspected that the arrestee committed a Schedule 1
offence and that the suspicion rested on reasonable grounds.
• Where the deprivation of liberty carries with it the imputation of criminal conduct of which
there was no reasonable suspicion, courts consider the injury to be very serious.29
• If the arresting officer had an improper motive, then one cannot say that the suspicion rested
on reasonable grounds. Motive is therefore an indicator of the reasonableness of the
defendant’s conduct, and is a factor that could be relevant in determining wrongfulness.
(Unlike instances of malicious deprivation of liberty, motive is not a necessary component in
claims for wrongful deprivation. It is merely a potential indicator of wrongfulness.) 30
• Once the police have arrested a person, they have an ongoing duty to reconsider the lawfulness
of the arrest should they subsequently receive sufficient information to conclude that the
arrested person is innocent.31
• Where police fail to inform a prosecutor and the Court that there is no evidence justifying the
arrest and detention of a person pending a criminal trial, the detention will be unlawful and the
person’s right to freedom and security in terms of section 12(1)(a) of the Constitution is
potentially infringed. The orders of a magistrate that such a person be held in custody pending
the outcome of the trial do not render that person’s detention lawful.32
• An investigating officer has a public-law duty not to violate an accused’s right to freedom,
either by not opposing his or her application for bail, or by placing all relevant and readily
available facts before the magistrate. A failure to perform this public duty could be considered
wrongful in delict.33
• The principle that the State bears the onus of proving that a deprivation of liberty is lawful also
applies to matters where the detention involves the application of immigration law, and
therefore falls outside the realm of criminal law and procedure.34 On appeal, the Constitutional
Court held that:
the deprivation of personal liberty is prima facie unlawful, calling for justification to avoid
liability for damages.35

Whether intention is an element of this iniuria is controversial. As it is a claim under the actio
iniuriarum, logic dictates that it should be an element. In Minister of Justice v Hofmeyr 36 the
Appellate Division confirmed that depriving liberty involves ‘the wrongful and intentional
infringement of an interest of personality’ and continued:
In this limited class of delicts dolus remains an ingredient of the cause of action, but in a
somewhat attenuated form, in the sense that it is no longer necessary for the plaintiff to
establish consciousness on the part of the wrongdoer of the wrongful character of his act.
Included in this limited class are cases involving false imprisonment and the wrongful
attachment of goods.

So, in these cases it is not the full animus iniuriandi that is required, but an attenuated version that
involves only the intention to arrest, and does not require that a defendant should also have known
that the conduct was wrongful. The effect of this passage is that a defendant cannot attempt to
exclude fault by pleading any defence that is aimed at negating consciousness of wrongfulness,
such as mistake. For example, if a police officer arrests a person in circumstances where he or she
cannot objectively justify the arrest (because he or she did not have a reasonable basis for believing
that the person arrested was committing an offence, about to commit an offence or had committed
an offence, as required by the Criminal Procedure Act), the officer cannot escape delictual liability
by satisfying the Court that he or she genuinely believed he or she was justified in effecting the
arrest. The officer’s direction of will towards arresting the plaintiff (his or her taking action to
deprive the plaintiff of his or her bodily freedom) will on its own satisfy the intention requirement
and delictual liability will ensue.

COUNTER Wrongful deprivation of liberty

 POINT Neethling, Potgieter and Visser37 contend that in wrongful deprivation of
liberty cases, courts have simply ignored the intention requirement, and
so this has become a form of liability without fault (that is, liability is
strict).38 Midgley39 suggests that it is wrong to equate the effect of an
attenuated form of intention with that of strict liability and to say that in
reality there is no difference.
At the heart of this difference of opinion lies the interpretation of a
passage in Smit v Meyerton Outfitters,40 which in Minister of Justice v
Hofmeyr 41 was found to be ‘a correct statement of our modern law’. It
reads:

In the case of actio iniuriarum the fault element involves

two considerations. The first is that the defendant acted
intentionally and the second is that the defendant knew that
the act was wrongful. In the case of wrongful arrest, even
though it developed out of the actio iniuriarum, the second
consideration is not a requirement for liability.

There is another category of cases – malicious arrest and detention – that requires the additional
element of malice. Malice, or improper motive or reason for doing something, is not intention, and
nor is it any other specific form of fault. The fault element in this category is no different from
other unlawful arrest cases, and nor, for that matter, are the other elements of harm and
wrongfulness. The peculiarity lies in the fact that society will not consider the category of
deprivation of liberty to be wrongful unless the defendant had an improper motive. For reasons of
policy, malice therefore becomes a necessary additional requirement for a wrongfulness finding in
such cases.
What, then, is the distinctive characteristic of this category of cases? Unlike wrongful
deprivation of liberty, malicious deprivation of liberty takes place under the guise of a valid judicial
process. Crucially, in this type of case, the defendant is alleged to have used the legal machinery of
the State to achieve an improper result. There may be a lawful intervening act between the
defendant’s conduct and the plaintiff’s deprivation of liberty – for example, the defendant arrested
the plaintiff, but the plaintiff’s detention came about because a magistrate ordered the plaintiff to be
incarcerated. The magistrate’s decision is lawful, so the only way in which the plaintiff can claim
damages is to show that the defendant had abused the process and had acted maliciously. The
feature that distinguishes this category of cases from ordinary wrongful arrest and detention cases is
that, although the defendant’s conduct is linked to the ultimate result, the immediate or proximate
cause for the detention was not the defendant’s conduct but someone else’s lawful conduct.42
As with wrongful arrest and detention cases, liability arises only if the plaintiff can prove all
the elements of the delict. The controversy over whether liability is strict or whether an attenuated
form of intention is required is not relevant in these cases. This is because if malice involves the
improper reason for a person’s action, then that person can hardly say that he or she did not direct
his or her will for that purpose. Accordingly, in this category of cases intention normally involves
both direction of the will and consciousness of wrongfulness.
In Relyant Trading (Pty) Ltd v Shongwe 43 the Court explained the distinction between a
wrongful arrest and a malicious deprivation of liberty. A wrongful arrest is one where the defendant
is the person who unjustifiably interfered with the plaintiff’s personal liberty. In contrast, a
malicious arrest is one where the defendant is the person who set a lawful process in motion in
terms of which the plaintiff was arrested and prosecuted.44 A claim for malicious prosecution
requires that an arrest or prosecution be instigated with animus iniuriandi and in circumstances
where there is no reasonable or probable cause for it.45

26.2.3 Seduction
Seduction occurs where a man induces a virgin (who is not his wife) to have consensual sexual
intercourse with him. In so doing, the man commits an iniuria, the consequence of which, in earlier
law, was either marriage or payment of a solatium. Only the latter option prevails today, but a claim
under the lex Aquilia would also hold for any patrimonial harm suffered, such as when a child is
born, lying-in expenses, maintenance for the mother before, during and after the confinement, and
maintenance for the child (and, if it dies, its funeral expenses).
The harm element lies in the actual deflowering of a virgin woman (the physical change to the
woman’s body as a result of the act of sexual intercourse). Wrongfulness arises from the fact that
the man behaved in a seductive manner to overcome the woman’s resistance and to induce her
consent. Courts consider this behaviour to be against public policy, and this is also the reason why
the defendant cannot use the woman’s consent as a defence to negate wrongfulness. The form of
intention, as with wrongful deprivation of liberty cases, is attenuated, which means that a mistake,
such as not knowing that the woman was a virgin, would not exclude intent.46

PAUSE FOR Seduction

• The action is available to women only. Is this appropriate in light of the equality
REFLECTION
provisions in our Constitution?47
• Although female virginity remains a protected interest in our law today, should
the action for seduction be formally abolished in a modern society where the
regard for virginity is much reduced?
• Does the action for seduction objectify women by placing a monetary value on
their virginal status, thereby unreasonably and unjustifiably infringing their
rights to equality and human dignity?
• Does the action for seduction, in its apparent reluctance to recognise that a
virgin woman can validly consent to sexual intercourse with a man in certain
circumstances, undermine women’s agency and promote a patriarchal and
sexist notion of passive female sexuality?

1 See DE v RH 2015 (5) SA 83 (CC) at fn 5, where the Constitutional Court gives a brief summary of the purpose and
scope of the actio iniuriarum.
2 For an example of a rolled-up action, see Blignaut v Protea Coin Group 2015 JDR 0962 (ECP), where the
plaintiffbrought claims for iniuria/insult (dignitas), defamation (fama) and assault (corpus) by means of a single
action, although each wrong was particularised in the pleadings. The plaintiffwas unsuccessful in his claim, which
pertained to an alleged assault, demeaning conduct and defamation which occurred during an altercation in a petrol-
station forecourt between the plaintiff(a private citizen) and an employee of an armed cash-in-transit guard.
3 2002 (5) SA 401 (CC) para 27; see also Le Roux v Dey (Freedom of Expression Institute and Restorative Justice
Centre as Amicus Curiae) 2011 (6) BCLR 577 (CC); 2011 (3) SA 274 (CC) at 274.
4 Voet Commentarius ad Pandectas 47.10.1 (1829).
5 Minister of Justice v Hofmeyr 1993 (3) SA 131 (A) at 145–146.
6 Bennett v Minister of Police 1980 (3) SA 24 (C).
7 N v T 1994 (1) SA 862 (C).
8 Venter v Nel 1997 (4) SA 1014 (D).
9 Bull v Taylor 1965 (4) SA 29 (A); M NO v M 1991 (4) SA 587 (D).
10 Broude v McIntosh 1998 (3) SA 60 (SCA).
11 Nell v Nell 1990 (3) SA 889 (T).
12 Minister of Justice v Hofmeyr 1993 (3) SA 131 (A) at 145–146.
13 1998 (3) SA 60 (SCA).
14 1980 (3) SA 24 (C).
15 Para 35A.
16 Para 37.
17 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 53.
18 Para 37.
19 [2008] 3 All SA 270 (SE).
20 Para 18.
21 Bennett v Minister of Police 1980 (3) SA 24 (C) at 35E–F.
22 At 34G.
23 At 35G–H.
24 See Chapter 8 section 8.3.8.
25 Tsose v Minister of Justice 1951 (3) SA 10 (A); Bennett v Minister of Police 1980 (3) SA 24 (C).
26 Minister of Justice v Hofmeyr 1993 (3) SA 131 (A) at 145–146.
27 1991 (1) SA 702 (E) at 707B.
28 Minister of Law and Order v Hurley 1986 (3) SA 568 (A) at 589E–F; Zealand v Minister of Justice and Constitutional
Development 2008 (4) SA 458 (CC); 2008 (2) SACR 1 (CC) paras 24 and 25; Naidoo v Minister of Police [2015] 4
All SA 609 (SCA) para 39.
29 May v Union Government 1954 (3) SA 120 (N) at 130. Even accusing a person of shoplifting could result in a
deprivation of liberty, as the suspect is stopped, questioned and may be expected to subject themselves to a search
(Pieterse v Clicks Group Ltd 2015 (5) SA 317 (GJ)).
30 In Areff v Minister van Polisie 1977 (2) SA 900 (A), a businessman was arrested by policemen who, without
reasonable grounds, had assumed that tearing up a summons not addressed to him amounted to a Schedule 1 offence.
31 Minister of Police v Du Plessis 2014 (1) SACR 217 (SCA) para 18 onwards.
32 Minister of Safety and Security v Tyokwana 2015 (1) SACR 597 (SCA) paras 42–44.
33 Woji v Minister of Police 2015 (1) SACR 409 (SCA) paras 28–29.
34 Rahim v Minister of Home Affairs 2015 (4) SA 433 (SCA) para 24.
35 Minister of Home Affairs v Rahim 2016 (3) SA 218 (CC) para 27.
36 1993 (3) SA 131 (A) at 154 H–I.
37 Neethling and Potgieter (2015) at 350; Neethling, Potgieter and Visser Neethling’s Law of Personality 2 ed (2005) at
119–120.
38 See Relyant Trading (Pty) Ltd v Shongwe and others [2007] 1 All SA 375 (SCA), where the Court remarked at para 4
that liability for wrongfulness in respect of unlawful arrest is strict, and neither fault nor awareness of wrongfulness is
required.
39 Midgley ‘Fault under the actio iniuriarum: Custer’s last stand?’ in Boezaart and de Kock (Eds) Vita perit, labor non
moritur: Liber Memorialis Visser (2008) at 187.
40 1971 (1) SA 137 (T) at 139 (our translation).
41 1993 (3) SA 131 (A) at 157.
42 The key requirements for malicious arrest and prosecution were restated by the Supreme Court of Appeal in
Magwabeni v Liomba (198/13) [2015] ZASCA 117 (11 September 2015) para 9. See also Minister of Justice and
Constitutional Development v Moleko [2008] 3 All SA 47 (SCA) para 8 and Rudolph v Minister of Safety and Security
2009 (5) SA 94 (SCA) para 16.
43 [2007] 1 All SA 375 (SCA).
44 Para 4.
45 Para 5.
46 Neethling and Potgieter (2015) at 347–348 suggest that this could be another example of liability without fault.
47 See Bennett, Mills and Munnick ‘The Anomalies of Seduction: A Statutory Crime or an Obsolete, Unconstitutional
Delict?’ (2009) 25(2) SA Journal on Human Rights at 330–352.
 Chapter 27

Infringements of dignity

27.1 Introduction

27.2 Insult

27.3 Conclusion

27.1 Introduction |
Dignity is an umbrella concept that embraces both constitutional and common-law notions of
dignity. Our Constitution recognises dignity as a core value and also as a fundamental right. In S v
Makwanyane 1 O’Regan J noted:
Recognising a right to dignity is an acknowledgment of the intrinsic worth of human beings:
human beings are entitled to be treated as worthy of respect and concern. This right
therefore is the foundation of many of the other rights that are specifically entrenched in …
(the Bill of Rights).

Our common law protects similar values, and these serve as the basis for a remedy under the actio
iniuriarum. However, dignity in constitutional law and dignity in common law have different
content. The constitutional notion of dignity includes reputation (the public aspect of the right to
dignity), but not privacy, which the Bill of Rights protects separately in section 14. The common-
law notion includes privacy, but excludes reputation. Nonetheless, the common law, although
different in many respects, is compatible with the rights and values expressed in the Constitution.
Harms DP confirmed this point in Le Roux v Dey 2 and noted the interplay between the
constitutional and common-law concepts: 3
The term ‘dignity’ covers a number of concepts in section 10 of the Constitution, but in the
present context we are concerned with the plaintiff’s sense of self-worth. Melius de Villiers
spoke of the inborn right to the tranquil enjoyment of one’s peace of mind; and the valued
 and serene condition in one’s social or individual life which is violated when one is subjected
to offensive and degrading treatment, or exposed to ill-will, ridicule, disesteem or contempt.

The common-law concept of dignity (dignitas) is a complex one that has both a broad and a narrow
meaning. In its broad sense, dignitas includes a person’s right to his or her feelings (the right not to
be subjected to an iniuria or insult), as well as the rights of privacy and identity. Courts could in
future add further rights to the concept. The narrow meaning restricts the right to feelings and, in
particular, the impairment of dignity in the form of an insult.

27.2 Insult
Insult, or the violation of a person’s feelings, is what we have referred to as iniuria in the narrowest
of the dignity concept’s three meanings. An invasion of dignity in the form of an iniuria occurs
when a person’s subjective feelings of self-respect or self-esteem – that is, a person’s pride and
moral value (self-worth) – are violated.4 To constitute a delict, there must be (a) a factual violation
of the plaintiff’s feelings that is both (b) wrongful and (c) intentional.
Factual violation occurs when a person feels humiliated – in other words, where contumelia is
present. It does not matter what caused the humiliation. Usually it is belittling or insulting words,5
but insulting behaviour can be any form of conduct.6 The focus here is on the conduct’s impact on
the person, so we are primarily concerned about people’s own opinions of themselves, viewed
subjectively, and not with the opinion that others might hold. From the subjective nature of this
enquiry, we can draw the following conclusions:
• A person who does not feel insulted has not suffered harm, even if the defendant’s conduct is
objectively insulting.
• Since hurt feelings are an essential prerequisite for insult, artificial persons necessarily cannot
suffer this type of harm.
• Others do not have to be aware of the insulting behaviour or its effect. So, unlike with
defamation, publication of the words to a third person is not necessary to constitute an
impairment of dignity (although this requirement does not preclude publication).

PAUSE FOR The subjective nature of dignity

REFLECTION Despite what we say about its subjective nature, it appears that dignity
might not be an entirely subjective concept. The matter was raised in
Universiteit van Pretoria v Tommie Meyer Films (Edms) Bpk,7 and in
Bennett v Minister of Police,8 where the Court awarded R50 in damages
in respect of ‘a measure of contumelia objectively regarded’ that was
inherent in that assault. Van der Walt and Midgley9 say that dignity:
‘consists of two aspects: one closely associated with human feelings –
one’s sense of “self-worth” and “self-respect”; and another associated
with someone’s outward dignity, status, and esteem – in the sense that
someone is considered to be a dignified person.10 Natural persons have
dignity in both senses, but the dignity of juristic persons is limited to
status and esteem only.’

The next step is to establish wrongfulness. In essence, not only must the feelings have been violated
subjectively, but society should consider the invasion of the interest to be unreasonable to such an
extent that the defendant should be held liable for assuaging the wounded feelings. Persons are
expected to show some resilience and tolerance towards offensive behaviour, and courts will not
impose liability unless society’s sense of justice (boni mores) would point to the situation being
offensive, degrading and not trivial.11
Delange v Costa12
Costa wrote a letter to Delange, an advocate of the High Court and a senior official in the
olive industry, in which he accused Delange of being motivated by self-interest and claimed
that his involvement in a particular transaction was not in the industry’s best interests. The
remark offended Delange, who was described in the judgment as ‘a man of not
inconsiderable self-esteem and heightened sensitivity’.
The Court first looked at the relevant principles:13

I now turn to consider the law which is applicable to the facts of the present
matter. Melius de Villiers The Roman and Roman-Dutch Law of Injuries at 27
notes three essential requisites to establish an action for injuria. They are:

‘An intention on the part of the offender to produce the effect of his act;
An overt act which the person doing it is not legally competent to do; and which
at the same time is
An aggression upon the right of another, by which aggression the other is
aggrieved and which constitutes an impairment of the person, dignity or
reputation of the other.’

Logically, in an action for injuria one should commence by enquiring into the
existence of the second of these requisites, viz whether there has been a
wrongful overt act. (It is more common, and probably juristically more correct, to
speak of a ‘wrongful’ rather than an ‘unlawful’ act.) A wrongful act, in relation to
a verbal or written communication, would be one of an offensive or insulting
nature. Once the wrongfulness of such act has been determined animus
injuriandi will be presumed. … It would be open to the defendant to rebut such
presumption by establishing one of the recognised grounds of justification. If
the defendant fails to do so the plaintiff, in order to succeed, would have to
establish the further requirement that he suffered an impairment of his dignity.
This involves a consideration of whether the plaintiff’s subjective feelings have
been violated, for the very essence of an injuria is that the aggrieved person’s
dignity must actually have been impaired. It is not sufficient to show that the
wrongful act was such that it would have impaired the dignity of a person of
ordinary sensitivities. Once all three requisites have been established the
aggrieved person would be entitled to succeed in an action for damages, subject
to the principle de minimis non curat lex.

Later in the judgment, the Court looked at the element of wrongfulness:14

In determining whether or not the act complained of is wrongful the Court

applies the criterion of reasonableness – the ‘algemene redelikheidsmaatstaf’… .
This is an objective test. It requires the conduct complained of to be tested
against the prevailing norms of society (ie the current values and thinking of the
community) in order to determine whether such conduct can be classified as
wrongful. To address the words to another which might wound his self-esteem
but which are not, objectively determined, insulting (and therefore wrongful)
cannot give rise to an action for injuria … . For words to be injurious they must
infringe one of the ‘absolute rights of personality’. … There is no such thing as
an absolute right not to be criticised. A person must be prepared to tolerate
legitimate criticism, ie criticism which is fair and honest. Put differently, an act
done in the exercise of a right is not a wrongful act, and can therefore not
constitute an injuria. Honest criticism is such an act. … Whether in given
 circumstances criticism may be regarded as legitimate must depend upon, inter
alia, the relationship of the parties involved and the nature of the affairs they
engage in. Businessmen who engage in competition (like politicians who take
part in public life) expose themselves to, and must expect, a greater degree of
criticism than the average private individual.

The judgment raises a number of interesting issues:

• Relying on Melius de Villiers,15 the Court set out16 the three essential requirements for
establishing an action for iniuria: intention, a wrongful overt act, and an aggression upon
another’s right in the form of an impairment of the other person’s dignity and reputation. Note
that here the term ‘iniuria’ was used in its broadest sense, because the Court was setting out
the requirements in respect of all types of rights normally associated with the actio iniuriarum
and the requirements were not restricted to instances of insult. However, after setting out the
requirements, the Court proceeded to focus on aspects of iniuria in the narrow sense of
offensive and insulting behaviour.
• The Court suggested that, logically, we should start the enquiry with the second requirement,
the wrongful conduct. We hold a different view, and suggest that the third requirement, the
impairment of the personality right, should be the logical starting point, for without the harm
element there can be no delict.
• Both Melius de Villiers and the Court speak of a wrongful ‘act’. We suggest that, unlike in
criminal law, it is incorrect to focus on the conduct element when assessing wrongfulness.
This is because courts decide wrongfulness based on all the circumstances. It is the causal
sequence resulting in the harm caused that is either wrongful or not. So, we would disagree
with the Court’s approach in suggesting that after it has been found that the elements of
wrongfulness and intention exist that the plaintiff has to prove that his dignity was impaired.
• The Court also indicated that once conduct is considered to be wrongful, a rebuttable
presumption of animus injuriandi arises. While this statement is in line with previous
authority, it is not compatible with what happens when infringements of other personality
rights occur. In defamation, for example, the infringement of a person’s reputation leads to
two rebuttable presumptions, one of wrongfulness and one of animus. We suggest,
therefore, that the plaintiff should instead prove that a personality right (dignity) has been
infringed/factually disturbed, in which event presumptions of wrongfulness and intention will
arise. The defendant will then bear the onus of rebutting those presumptions.
• The Court also said that the presumption of animus may be rebutted by one of the
recognised ‘grounds of justification’. However, to avoid confusion, we should restrict the term
‘grounds of justification’ to defences excluding wrongfulness – that is, those that justify a
person’s conduct. Defences aimed at negating intention, such as mistake or jest, do not
justify a person’s conduct and so do not fall into this category.
• The judgment highlights the fact that the harm element – the impairment of dignity – involves
a subjective inquiry, and that an intentional infringement of dignity on its own cannot
constitute an iniuria. All three requisite elements need to be present. For example, in Walker
v Van Wezel17 the Court stated that:
Although the effect of the words used upon the person to whom they are addressed is
not irrelevant, the fact that he feels himself aggrieved is only one element in the
injuria; the communication itself must be of an insulting or offensive nature … .
• Therefore, not only must the plaintiff feel subjectively insulted, but the behaviour, regarded
objectively, must also be of an insulting nature. We assess this by considering wrongfulness,
for which the test is the criterion of reasonableness. This enquiry involves objectively
assessing the situation according to the norms of society.
• The Court added a rider to its statement that liability arises when the three prerequisites are
present: ‘subject to the principle de minimis non curat lex’. We suggest that, while the idea is
correct, the Court should not have presented this principle as a rider. This principle is an
expression of society’s norms (defendants should not be liable for compensation in respect
 of trivial matters) and should, therefore, properly be assessed as part of the wrongfulness
enquiry.

The Court found that, despite Delange subjectively feeling that he had been insulted, this did
not constitute an iniuria, for the comments were more similar to honest criticism that did not
exceed legitimate bounds. The Court also took into account the fact that both Delange and
Costa were businessmen who were expected to display mutual robustness in their dealings
with one another. Accordingly, the wrongfulness prerequisite for liability had not been
established.

Cele v Avusa Media Limited18

The plaintiff, a high-ranking politician who had made certain statements calling on the police
to ‘shoot to kill’ when dealing with criminals, claimed that the Sowetan newspaper had
insulted him by publishing a digitally altered image of him which depicted him as a sheriff in
the ‘Wild West’. The Court held that the image would have been understood by reasonable
readers of the Sowetan to mean that the plaintiff was taking a tough stance on crime and that,
like a sheriff in the Wild West, he wanted criminals to be harshly dealt with by the police and
brought to justice, either dead or alive.19 The Court found that this image was satirical (a
caricature or parody) and was protected by the right to freedom of expression.20 In addition,
the Court held that the image was not realistic and would not have been regarded by a
reasonable person as a genuine depiction of the plaintiff.21 The Court held that a reasonable
person in the position of the plaintiff (in other words, a reasonable seasoned politician) would
not have felt insulted and humiliated by the image.22 Accordingly, the interference with the
plaintiff’s right to dignity was not wrongful, and the action for insult failed.23

Note that posting offensive and injurious statements to a Facebook page constitutes insult for the
purposes of the actio iniuriarum.24
The final requirement is intention, which in this instance retains its ordinary meaning –
namely, that a person must have directed his or her will towards achieving the wrongful
consequence (to offend or insult the other person), and that he or she knew that what he or she was
doing was wrong, in that it would not carry society’s approval. Once the plaintiff has proved the
infringement of dignity (factual disturbance of the personality right), a rebuttable presumption of
intention arises to assist the plaintiff. The onus is then on the defendant to raise and successfully
establish a defence excluding intention (for example, mistake, jest or provocation) in order to
escape liability.

PAUSE FOR Breach of promise to marry

REFLECTION In Van Jaarsveld v Bridges 25 Harms DP stated that:

the time has arrived to recognise that the historic approach

to engagements is outdated and does not recognise the
mores of our time, and that public policy considerations
require that our courts must reassess the law relating to
breach of promise.26

The Court, accordingly, found that breach of promise did not in itself give
rise to a claim, and that an action lies only if the elements of the actio
iniuriarum are met.
Van Jaarsveld had been engaged to Ms Bridges. A month prior to the
scheduled marriage, Van Jaarsveld sent Bridges a text message
terminating the engagement. Bridges had three previous marriages, and
Van Jaarsveld’s choice of her as his potential wife did not find favour with
his mother, who thought Bridges had ulterior motives in wanting to marry
 her son. Van Jaarsveld thus stated in his message that, after giving the
matter serious thought, it seemed to him that the marriage should not
take place. In apologising for the inconvenience and hurt his actions were
likely to cause, he stated it was more appropriate not to go through with
the marriage than to proceed with it and later be forced to divorce.
The Court noted that an engagement, being a contract, could be
cancelled without financial consequences if there is just cause for such
cancellation.27 Just cause is usually defined as any event or condition or
actions of the other party which would jeopardise a long and happy
marriage, and which would induce any right-minded member of society to
rescind the engagement. Unwillingness to marry is clear evidence of the
irretrievable breakdown of the engagement, and it would be illogical to
attach more serious consequences to an engagement than to a marriage.
The Court preferred not to commercialise the engagement relationship
and rejected the concept that parties, when promising to marry,
contemplate that a breach of their engagement would have financial
consequences as if they had in fact married.28 Accordingly, contractual
damages were not awarded.
To found a claim for delictual damages, Bridges had to show that she
had suffered an iniuria – in other words, that the engagement had been
brought to an end in an intentionally insulting (contumelious) manner 29 in
circumstances society would consider wrongful, regardless of whether
the contractual repudiation had been justified.30 The manner in which the
engagement was brought to an end was central to the issue of liability,
and the fact that the feelings of the ‘innocent’ party were hurt or that he or
she felt slighted or jilted was not enough.
The Court found that the manner in which Van Jaarsveld notified
Bridges of his intention not to go through with the marriage was not
contumelious or insulting, and the claim for delictual damages was also
denied. The injury or contumelia was considered to be de minimis.
• Is a delictual claim for mere breach of promise still justifiable in modern society,
or was the Court correct in saying that such claims can be entertained only if
the principles of the actio iniuriarum have been met?
• The Court found that whether or not the breach of contract was wrongful and
without just cause did not affect the delictual claim, and that wrongfulness in
delict constituted a separate enquiry.31 Was the Court correct in stating that
there could still be a delictual claim even where there had been a just cause for
repudiating the engagement?
• The Court said that words:
which might wound the self-esteem of the addressee but which are not,
objectively determined, insulting (and therefore wrongful) cannot give
rise to an action for injuria.
• The Court went on to say:
Importantly, the character of the act cannot alter because it is
subjectively perceived to be injurious by the person affected thereby.32
• Do you agree that Van Jaarsveld’s actions were not insulting, even though
Bridges felt insulted? Should a person whose feelings have been hurt not have
those feelings assuaged in some way?
• Is the principle of de minimis non curat lex applicable in delictual cases? What
function does the principle serve?

After Wiese v Moolman 33 confirmed the validity of an innocent spouse’s action for delictual
damages against an adulterous third party, the question arose as to whether the action should
continue to exist in our law. In RH v DE 34 both the Supreme Court of Appeal and thereafter the
Constitutional Court held that the action for adultery should be abolished, albeit for different
reasons.
RH v DE35
The plaintiff husband claimed damages from a third-party defendant arising out of adultery
the third party had committed with the plaintiff’s wife. Damages were claimed for both
contumelia (humiliation) and loss of consortium (comfort and society between spouses). The
Court held that the claim for loss of consortium could not succeed. Since the plaintiff and his
wife had separated prior to the first instance of adultery, the adultery had not caused the loss
of consortium and the defendant could not be held responsible for it.36 On the other hand,
potentially, the claim in respect of contumelia could succeed – provided the action for adultery
was itself still valid.37
The Court observed that an action for damages for adultery was no longer available in
most Western countries,38 which would suggest that the action is seen as outdated and
archaic.39 The arguments supporting its retention were based on the need to protect the
institution of marriage, as well as the wronged spouse’s need to be compensated for the
harm he or she had suffered.40 As to the first issue, the Court held that the institution of
marriage was important and should be protected,41 but that the adultery action had little
deterrent effect and therefore did not protect the institution of marriage in any meaningful
way. Marriages are held together by the moral commitment of the parties to them, not by the
threat of litigation.42 In addition, adultery is not necessarily the reason for the breakdown of a
marital relationship. In many cases, the adultery is merely a symptom of a marriage in crisis.
Therefore, it cannot be said that making a third party pay damages for having engaged in
adultery would in any way protect an already dysfunctional relationship.43
As regards the question whether the action for adultery provided necessary solace or
compensation for the aggrieved spouse, the Court held that in today’s society a reasonable
person whose spouse had committed adultery would not feel humiliated or insulted by such
conduct. Accordingly, there was no need for the action.44 The Court noted that actions for
adultery often cause great pain and embarrassment for everyone involved. They are also
very expensive for both parties, yet the damages awarded by courts in these cases are
usually minimal.45
The above arguments militated strongly against the continued recognition of the action for
adultery. The Court felt that:

the time has come for our law to recognise, in harmony with most other legal
systems, that in the light of changing mores, these reasons advanced for the
continued existence of the action have now also lost their persuasive force.46

It concluded that social values have shifted to the degree that the legal convictions of the
community no longer support a finding that adultery is wrongful conduct for the purposes of
the law and held that: ‘(i)n the light of the changing mores of our society, the delictual action
based on adultery of the innocent spouse has become outdated and can no longer be
sustained’, and ‘the time for its abolition has come’.47

DE v RH48
The Constitutional Court endorsed the Supreme Court of Appeal’s finding, but grounded its
decision on fundamental rights and constitutional considerations. The Court held that the act
of adultery was no longer wrongful and therefore should not give rise to a cause of action
under the actio iniuriarum for contumelia and loss of consortium. The action for adultery
brought by the innocent spouse had a deleterious impact on the constitutional rights to
privacy, freedom and security of the person and freedom of association of both the adulterous
spouse and the defendant third party.
The key issue for the Court was: ‘whether nowadays the act of adultery meets the element
of wrongfulness in order for delictual liability to attach’ and this ‘pivotal question concerns
wrongfulness’.49
 The common law, including the law of delict, had to be developed in accordance with the
Constitution,50 and this entailed developing the common law in accordance with extant public
policy,51 infused with constitutional norms.52 Madlanga J, for the majority, explained the rights-
based reasons for abolishing the action for adultery as follows:53

The right of a non-adulterous spouse that is implicated by the act of adultery is

the right to dignity. Not surprisingly, that is the right that the applicant asserts.
Undoubtedly, adultery has the potential to infringe the non-adulterous spouse’s
right to dignity … . Adultery … entails a significant intrusion of a third party into
a person’s most intimate relationship without their consent. That intrusion is not
made any less severe by present day attitudes towards adultery.

Nevertheless, this potential infringement of dignity must be weighed against the

infringement of the fundamental rights of the adulterous spouse and the third
party to privacy, freedom of association and freedom and security of the person.
These rights demand protection from state intervention in the intimate choices
of, and relationships between, people. This must be viewed in light of current
trends and attitudes towards adultery both nationally and internationally. These
attitudes also demonstrate a repugnance towards state interference in the
intimate personal affairs of individuals.

I am led to the conclusion that the act of adultery by a third party lacks
wrongfulness for purposes of a delictual claim of contumelia and loss of
consortium; it is not reasonable to attach delictual liability to it. That is what
public policy dictates. At this day and age it just seems mistaken to assess
marital fidelity in terms of money.

Mogoeng CJ and Cameron J wrote a separate concurring judgment in which they stressed
that the law created a regulatory framework for concluding a marriage and set out the
obligations of the parties to it, but that the essence of marriage was the moral commitment of
the parties to it to sustain it. Accordingly, the law could not prop up an ailing marital
relationship with a delictual damages claim: 54

I am in agreement with these views. The law does and can only create a
regulatory framework for the conclusion of marriage and the enforcement of
obligations that flow from it. It can also help ensure that barriers to family life are
removed. The rest is in the hands of the parties to the marriage. Barring
exceptions, they decide freely to get married and it is within their ability to
protect their marriage from disintegrating.

… Like the Supreme Court of Appeal, I also believe that parties’ loss of moral
commitment to sustain marriage may lead to its failure. For abuse of one by the
other and other factors that could lead to the breakdown of marriage are, in my
view, likely to creep in when that commitment ceases to exist.

The law cannot shore up or sustain an otherwise ailing marriage. It continues to

be the primary responsibility of the parties to maintain their marriage. For this
reason, the continued existence of a claim for damages for adultery by the
‘innocent spouse’ adds nothing to the lifeblood of a solid and peaceful marriage.

PAUSE FOR Criminal defamation

REFLECTION Were the courts correct in abolishing the action in its entirety? Are there
not circumstances in which public policy might regard extra-marital
relationships as wrongful? Could there not be circumstances where the
extra-marital relationship unacceptably interferes with the dignity and
 self-esteem of the innocent spouse? It may well be that the innocent
spouse might have a claim in such circumstances, not by way of an
action for adultery, but for an infringement of dignitas.55

27.3 Conclusion
The right to dignitas (in the narrow sense) means the right not to be insulted. If a person through
words or conduct wrongfully and intentionally interferes with another person’s right to dignity, the
insulted person is entitled to damages in delict. In line with general principles, the plaintiffis
required to prove that his or her right has been factually disturbed by proving on a balance of
probabilities that he or she subjectively felt insulted or demeaned by the conduct. The onus then
shifts to the defendant to justify the factual disturbance, either by proving on a balance of
probabilities that the conduct in question was not objectively insulting, or that it was for some other
reason justifiable in the circumstances (the wrongfulness enquiry). Animus iniuriandi can be
rebutted by raising one of the defences excluding intention, such as mistake, jest or provocation. As
in all claims brought under the actio iniuriarum, a court must consider the matter against the
prescriptive normative framework of the Constitution, by weighing the constitutional rights to
freedom of expression and dignity in deciding whether or not to allow the claim.

1 1995 (3) SA 391 (CC) para 328.

2 2010 (4) SA 210 (SCA).
3 Para 20.
4 Cele v Avusa Media Ltd [2013] 2 All SA 412 (GSJ) para 41; Greeff v Protection 4U h/a Protect International 2012
(6) SA 392 (GNP) at 408–409.
5 Brenner v Botha 1956 (3) SA 257 (T); Delange v Costa 1989 (2) SA 857 (A). In the former case, the plaintiff was
called ‘a bloody bitch’ and told ‘your face makes me want to be sick’.
6 Boswell v Union Club of South Africa (Durban) 1985 (2) SA 162 (D). In this case, the insulting conduct complained
of was the unlawful expulsion of the plaintiffs from a private club.
7 1979 (1) SA 441 (A) at 453–457.
8 1980 (3) SA 24 (C).
9 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 101.
10 See also Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6)
BCLR 577 (CC); 2011 (3) SA 274 (CC).
11 Cele v Avusa Media Ltd [2013] 2 All SA 412 (GSJ) para 41; Greeff v Protection 4U h/a Protect International 2012
(6) SA 392 (GNP) at 408–409.
12 1989 (2) SA 857 (A).
13 At 860I–861F.
14 At 862E–I.
15 De Villiers The Roman and Roman-Dutch Law of Injuries: A Translation of Book 47, Title 10 of Voet’s Commentary
on the Pandects with Annotations by Melius de Villiers (1899) at 27.
16 At 860.
17 1940 WLD 66 at 71.
18 [2013] 2 All SA 412 (GSJ).
19 Para 28.
20 Para 47.
21 Para 45.
22 Para 46.
23 Para 49.
24 Dutch Reformed Church Vergesig v Sooknunan 2012 (6) SA 201 (GSJ) paras 86–99.
25 2010 (4) SA 558 (SCA).
26 Para 3.
27 Para 5.
28 Para 8.
29 Para 4.
30 Para 4.
31 Para 19.
32 Para 19.
33 2009 (3) SA 122 (T).
34 2014 (6) SA 436 (SCA); 2015 (5) SA 83 (CC).
35 2014 (6) SA 436 (SCA). For a critical discussion of this case, see Carnelley ‘Vonnisbespreking: Die doodskoot vir of
slegs die verwonding van die eis teen die derdeparty-egbreker? RH v DE 2014 6 SA 436 (HHA)’ (2015) 12(2) LitNet
Akademies at 333–346, in which the author criticises the Court for having failed to engage in the section 39(2) enquiry
mandated by the Constitution, and for its superficial comparative legal approach. For an explicitly feminist critique of
the judgment, see Bonthuys ‘RH v DE: A Feminist Minority Judgment on Adultery’ (2015) 31(2) SAJHR at 379–400,
where the author crafts a notional feminist minority judgment on the issues. In PV v AM 2015 (3) SA 376 (ECP), the
Court held that oral sex is sufficient to ground a claim for loss of consortium and iniuria under the actio iniuriarum
(paras 31–35). However, this decision preceded RH v DE.
36 Paras 13 and 15.
37 Para 15.
38 Paras 21 and 27.
39 Para 21.
40 Para 28.
41 Para 33.
42 Para 34.
43 Para 34.
44 Para 35.
45 Para 39.
46 Para 38.
47 Para 40.
48 2015 (5) SA 83 (CC).
49 Para 11.
50 As mandated by section 39(2).
51 Para 16.
52 Para 17.
53 Paras 60–63 (footnotes omitted).
54 Paras 69–71 (footnotes omitted).
55 See Bonthuys (fn 35), who argues in her feminist ‘minority judgment’ that the action for adultery should be
maintained in a limited category of cases (para 51).
 Chapter 28

Infringements of privacy

28.1 Introduction

28.2 The common-law right to privacy

28.3 Juristic persons and the right to privacy

28.4 The constitutional right to privacy

28.5 Conclusion
|

28.1 Introduction
Privacy, like dignity, involves both constitutional and common-law concepts. Privacy is a self-
standing right in our Constitution,1 but at common law the right to dignity (in the wide sense)
includes the right to privacy and is protected under the actio iniuriarum.2 So, although the right to
privacy exists as an independent personality right, courts perceive such claims as infringements of
dignity. Nonetheless, courts recognise that the concepts of dignity and privacy are different,
because proof of contumelia is not a prerequisite for establishing a privacy claim.3
In line with general principles, for liability to arise there must be (a) a factual violation of the
right to privacy that was (b) wrongful and (c) intentional. The right to privacy can be infringed in
two main ways: (i) an intrusion on the private realm (for example, by reading a private and
confidential letter addressed to another person, or by listening in on a person’s private telephone
conversation) and (ii) a disclosure of private facts to the public (for example, publishing private
facts about a person’s life in the media, both traditional and online).
Courts assess wrongfulness in privacy cases by examining a wide range of policy
considerations, including constitutional norms. If publishing a private fact was in the public interest
(for example, the private conduct of a politician that might have an impact on how he or she
performs his or her public functions), a court may well regard the harm caused to that person’s right
to privacy as reasonable in the circumstances. Consent is a wrongfulness defence often used to
justify the publication of private facts about a person. However, the courts have stressed that a
limited and specific consent given by a person to make public certain private facts about himself or
herself cannot be used as a blanket justification for the publication of those facts in any manner or
form.4 In particular, if a person consents to the publication of private facts about him or her to a
limited range of persons, such consent cannot then be used as the basis for the mass publication of
those facts.5
Unlike in the law of defamation, courts have not extended the fault requirement in privacy
cases involving media defendants to include negligence. Accordingly, a media defendant in a
privacy case need only rebut the presumption of intention in order to escape liability. However,
there are signs that this position could change in the future.6

28.2 The common-law right to privacy

Simply put, the right to privacy is the right to be left alone.7 It includes freedom from intrusion and
interferences, as well as from unauthorised disclosures of information about a person’s private life.8
Therefore, each person has the right to decide what he or she would like to keep private, and when
and under what conditions his or her personal life can be made public. The right to privacy refers to
the most personal aspects of a person’s existence, often referred to as a person’s inner sanctum or
the truly personal realm (such as a person’s family life, sexual orientation and home environment),
which are excluded from others. The scope of personal or private space is diminished when we
consider the social environment in which people act – for example, when they are in their offices,
in their cars or on mobile telephones.9 Although privacy is jealously protected in the truly personal
realm, as a person moves into communal relationships and activities, such as business and social
interactions, the scope of personal space shrinks accordingly.10 In other words, the extent to which a
person can reasonably expect his or her privacy to be protected depends on the context and
circumstances.
The factual violation of a person’s right to privacy occurs when one’s personal space is
disturbed. Privacy is violated only if outsiders become acquainted with a person’s personal life or
personal affairs. This may occur either by (i) a direct intrusion, or (ii) the unauthorised disclosure of
a person’s private facts or personal affairs to another.11 Private facts include:12
• Contents of private correspondence
• Debts
• Physical deformities and health
• Lifestyle
• Childhood and background
• Family life
• Past activities (embarrassing facts)
• Confidential information
• Information stored in data banks.
Intrusions into a person’s private life or affairs have included a raid on a brothel,13 watching
someone undress,14 setting up bugging and listening devices,15 entering a person’s room or private
residence,16 improperly interrogating a detainee,17 listening to private conversations through
unlawful tape recordings,18 and taking an unauthorised blood test.19
S v A20
In this case a private detective had installed a bugging device in the plaintiff’s apartment. He
was instructed to do so by an estranged wife pending divorce proceedings. She suspected
that her husband was having an affair, and the purpose of the device was to listen in on his
 private conversations. The private detective was found guilty of crimen iniuria, and the Court
held that his conduct amounted to a serious violation of the complainant’s dignity.
• Although this was a criminal case, the issues were similar to those found in delict.
• The Court held that dignity was violated in this case. However, subsequently our courts have
clarified concepts and today they would see it as an invasion of privacy.

Pretorius v Minister of Correctional Services21

In this case, the applicant applied for an interdict to prohibit prison authorities from
broadcasting certain radio programmes into his prison cell. The Court granted the application
on the ground that the broadcast infringed the applicant’s right to acoustic privacy which,
according to the Court:

normally includes the right not to have one’s personal space invaded by any
broadcast to which the individual has not consented to be exposed.22

Developments in technology and electronic communication have threatened and compromised the
protection of privacy in certain circumstances, and legislation in line with constitutional provisions
now regulates this sphere quite closely.23 For example, service providers monitor email
correspondence on a regular basis and in a variety of ways. If a person read or disclosed content,
this would be an invasion of privacy, unless the service provider could show that the monitoring
took place for a legitimate purpose.24 In S v Naidoo 25 false and misleading information had been
furnished to a judge to obtain an order (in terms of the Interception and Monitoring Prohibition Act
127 of 1992) to tap a telephone. Since the judge granted the order based on false information given
to him, the order was unlawful and the monitoring was accordingly declared an unlawful violation
of the accuser’s right to privacy. Employers may also monitor employees’ electronic
communication provided that it is reasonably connected to business activity.

PAUSE FOR Privacy threats in the online world

REFLECTION Privacy has become a scarce commodity in the virtual world of the
internet. Once a person starts interacting online, much of his or her
personal information becomes available to numerous other online users.
Many websites use ‘cookies’ to track an individual’s online interaction –
which websites a person visits and the links he or she clicks on. A visit to
an online retailer and a mere click on an item may result in a person
finding for some time thereafter that the same item is displayed in adverts
on many other sites the person visits. While this might seem benign, it
has implications for one’s privacy rights. Each website might track only a
small portion of a person’s online visiting habits or online purchases, but
as sites share information, the aggregation of data leads to a threat of
data mining (where small pieces of one’s private information collected
from different online sources are aggregated in one site, called a data
bank). The aggregated information may contain sensitive information
ranging from personal details, such as age or weight, to shopping
behaviours, credit histories, banking details, criminal records and medical
information. Data-bank owners use a person’s information for their own
purposes, or sell the information to other users. The law of delict is not
ideally placed to deal with such threats to persons’ privacy, which is why
the Protection of Personal Information Act 4 of 2013 (‘POPI’) was
enacted to help protect each person’s online information. Is there still
room for delict to play a role?
 PAUSE FOR Protection of Personal Information Act 4 of 2013
REFLECTION POPI aims to ensure that data consisting of or containing personal
information is collected, processed, stored and shared in a manner that
protects the right to privacy of South African citizens. The Act applies to
the processing of personal information,26 and provides that this must take
place lawfully.27
POPI provides a data subject (the person to whom personal
information relates) with certain key rights, including the right to institute
civil proceedings regarding an alleged interference with his, her or its
personal information.28 POPI envisages that it is possible to bring a claim
for damages in delict arising out of a wrongful and culpable interference
with one’s personal information. In this regard, a data subject or, at the
request of a data subject, the Regulator established in terms of the Act,
may institute a civil action for damages in a court having jurisdiction
against a responsible party for an interference with the protection of
personal information of a data subject,29 whether or not there is intention
or negligence on the part of the responsible party.30 POPI provides for
strict liability in delictual actions brought in terms of POPI in respect of
interferences with the protection of personal information of a data subject.
A court may award damages for both patrimonial and non-patrimonial
loss, as well as ‘aggravated damages’ suffered by a data subject in such
circumstances.31

Disclosures of private information or personal facts occur when one person publicises known
private facts about another person who is unaware that they are being publicised. Common
examples include unauthorised use of photographs,32 publishing a love story in a magazine,33 a
doctor telling colleagues that a patient has Aids,34 and mass publication of the HIV status of
persons.35 Posting personal details, such as a person’s private email address and telephone number,
to a Facebook page also constitutes an unlawful disclosure of private facts.36
Most intrusions involve individuals becoming aware of private information or personal facts
about others. Such information may or may not be published to third parties. If the individuals do
publish it to others in an unauthorised manner, it would give rise to an additional action for
invasion of privacy based on wrongful publication of private facts. Irrespective of whether the
violation occurs as an intrusion or a disclosure, the invasion of privacy must also have been
wrongful. General principles apply, and the court must be satisfied that the invasion occurred in a
manner that justifies the imposition of liability in delict. In accordance with general principles,
courts use the general criterion of reasonableness (the boni mores or legal convictions of the
community) to determine whether they should recognise a claim. Factors that courts consider
include whether the sense of justice in the community would dictate that confidentiality should be
protected in the circumstances (for example, a doctor-patient relationship37 or boardroom
deliberations),38 whether a public value or constitutional right such as freedom of expression is
involved, or whether the information disclosed is of public concern. Society might, for example,
recognise a claim where a person’s extra-marital sexual behaviour is exposed, but not if that person
is a celebrity who touts himself or herself as a person of sound moral values. So, even though the
information disclosed is often true, the issue is whether the disclosure was made in the public
interest.
Mhlongo v Bailey39
A magazine published an article entitled ‘Dolly and Her Men’ about a celebrity. The article
contained photographs of what was termed ‘the young men in her young life’. Two
 photographs were of Mhlongo, one of him as a younger man and a more recent one.
Mhlongo sued for invasion of privacy. The Court held:40

The mere unauthorised publication of photographs does not necessarily in itself

entitle the aggrieved party to damages, and the question whether such
publications constitute an aggression upon a person’s dignitas will depend
upon the circumstances of each case, the nature of the photograph, the
personality of the plaintiff, his station in life, his previous habits with reference
to publicity and the like.

The traditional defences associated with the actio iniuriarum would also be relevant in determining
wrongfulness, as was the case in Jansen van Vuuren v Kruger,41 where a doctor who had disclosed
the HIV status of his patient raised the defence of privileged occasion, albeit unsuccessfully.
National Media Ltd v Jooste42
Jooste, the lover of a well-known rugby player, was interviewed about their relationship and
the child subsequently born of it. She agreed to the publication of two articles in You and 0
magazines, on payment, subject to her approval of the final version of the article and
photographs, and on an agreed publication date. The publishers presented her with a draft
article before publication, and she requested that certain amendments be made to it. She
also requested that the publication be delayed, although the publishers were keen to publish
as soon as possible. When the publisher failed to abide by certain contractual conditions,
Jooste withdrew her consent to publish the articles. The publishers nonetheless went ahead
and published them. About a week later, Jooste received and accepted payment. She then
instituted a claim for damages, alleging that the publication of the article had invaded her
privacy.

The judgment raised the following issues:

• The private facts must be worthy of protection:
The general sense of justice of the community as perceived by the Court … does not, in
a case such as this, require the protection of facts whose disclosure will not ‘cause
mental distress and injury to anyone possessed of ordinary feelings and intelligence,
situated in like circumstances as the complainant’.43
• The law will not protect information that an individual does not want to keep private.
However, an individual cannot solely determine what constitutes private information:44
The boundary of a right or its infringement remains an objective question. As a
general proposition, the general sense of justice does not require the protection of a
fact that the interested party has no wish to keep private.
• The individual must subjectively want the information to be kept private and society’s general
sense of justice must also consider that the information should be protected.
• What is the scope of the right to privacy? The Court said:45
A right to privacy encompasses the competence to determine the destiny of private
facts … . The individual concerned is entitled to dictate the ambit of disclosure, for
example to a circle of friends, a professional adviser or the public … . He may
prescribe the purpose and method of the disclosure … . Similarly … a person is
entitled to decide when and under what conditions private facts may be made public.
A contrary view will place undue constraints upon the individual’s so-called ‘absolute
rights of personality’ … .
• The appellants alleged that Jooste had consented to the publication in terms of an
agreement between them and so, in the circumstances, she could hardly claim an unlawful
invasion of her privacy right. According to the Court:
 [i]t is axiomatic that the defence of consent can only succeed if the prima facie
wrongful act falls within the limits of the consent … .46
• On the facts, the publication did not comply with the terms of consent, and even though
Jooste had accepted payment from the publishers, the publication had violated the terms of
consent.

Not only does an infringement of privacy have to be wrongful, it also has to be intentional. The
general principles in respect of animus iniuriandi, or intention to injure, also apply to privacy cases.
This means that intention is a subjective concept that involves the direction of the defendant’s will
towards infringing the plaintiff’s privacy, and the defendant’s knowledge that such infringement is
wrong in the circumstances. The defences that negate fault also apply; in particular, the defence of
mistake. Jest and provocation could also be used to rebut the presumption of intention.
NM v Smith (Freedom of Expression Institute as Amicus Curiae)47
The applicants were three HIV-positive women who lived in informal settlements near
Pretoria. The respondents were a journalist, Charlene Smith, a Member of Parliament,
Patricia de Lille, and a publisher, New Africa Books. The applicants had participated in clinical
trials for a combination of HIV drugs run by the University of Pretoria’s medical faculty. Along
with other participants in the trials, the applicants had raised concerns about illness and
fatalities among those involved. They also raised these concerns within the support group for
people living with HIV/Aids they had been attending. The priest who ran the support group
approached De Lille and asked her to investigate their complaints. She subsequently met
with some of the participants in the trials, including the applicants. The university
commissioned an external enquiry into the conduct of the trials. The enquiry issued a report
that exonerated the medical faculty and the doctor in charge of the trial. The report contained
the applicants’ names and indicated their HIV-positive status. This report was sent to a
number of interested parties, including De Lille. But the copy sent to De Lille lacked certain
annexures to the full version.
Smith was later commissioned by New Africa Books to write De Lille’s biography. The
book contained an account of the events leading up to the publication of the report, and it
revealed the names and HIV-positive status of the applicants as included in the report, which
De Lille had sent to Smith to assist her in her research. However, when the book was
published, the applicants sought an urgent interdict to prevent further circulation of these
details, arguing that they had not consented to their mass publication. Their consent had
extended only to the report itself, which was intended for limited circulation. The limited terms
of their consent appeared only in the annexures to the report, which had not been sent to De
Lille. The report itself merely stated that the applicants had agreed to their names being
published, and there was no indication the report was confidential.
The applicants brought an action for the violation of their rights to privacy, dignity and
psychological integrity resulting from the publication of their names and HIV-positive status
without their consent. When the matter came before the Constitutional Court, the majority
upheld the objective test for private information adopted in National Media Ltd v Jooste.48
Private facts are:

those matters the disclosure of which will cause mental distress and injury to
anyone possessed of ordinary feelings and intelligence in the same
circumstances and in respect of which there is a will to keep them private.

The respondents contended that the information was already in the public domain and there
was no intention to keep the facts private. However, the majority found that the applicants
had never consented to the wide circulation of the information, and that there were strong
privacy interests in maintaining confidentiality of sensitive medical information.49 It was in the
public interest that the autonomy of such information was recognised and respected.50 (In a
separate judgment, O’Regan J pointed out that simply because the information had reached
 some people did not necessarily place that information in the public domain.)51 Accordingly,
the applicant’s right to privacy had been infringed.
Turning to the right to dignity, the majority confirmed that, while there is nothing shameful
about living with HIV or Aids, it is nonetheless:

an affront to the infected person’s dignity for another person to disclose details
about that other person’s HIV status or any other private medical information
without his or her consent.52

In this instance, the rights to dignity and psychological integrity had also been infringed.53
There are two points to note:
• The applicants sought to vindicate their constitutional rights through the medium of the actio
iniuriarum, yet:
[w]hile the claim falls to be dealt with under the actio iniuriarum the precepts of the
Constitution must inform the application of the common law.54
• This judgment confirms the modern trend to separate the common-law rights to privacy and
dignity.

Although under the actio iniuriarum fault has always taken the form of intention, there are signs
that in the future negligence might also suffice in privacy cases. In NM v Smith 55 the majority of the
Constitutional Court confirmed the traditional approach to fault and declined to extend liability
under the actio iniuriarum to negligent conduct. However, two judges delivering minority
judgments were open to the idea.

NM v Smith (Freedom of Expression Institute as Amicus Curiae):56Minority judgment

In a minority judgment, O‘Regan J considered whether the Court should extend the common-
law requirements for liability under the actio iniuriarum to include negligent infringements of
privacy. O’Regan J and Langa CJ also thought that there were sound reasons for
distinguishing between ordinary defendants and those whose business it is to circulate
information, and both held that the author and publisher were media defendants.57
Although O’Regan J thought that the principles that apply to defamation do not necessarily
apply to privacy cases, she concluded that the common-law requirements for fault in privacy
cases should be similarly modified:

… it is appropriate to require the media when publishing private facts without

consent to establish either that the publication is reasonable in the
circumstances, in which case they will rebut wrongfulness, or that they have not
acted negligently in the circumstances in which instance they will need to rebut
the requirement of intention.58

Based on the facts, O’Regan J found that the author and publisher had not been negligent.
Langa CJ disagreed, holding that a reasonable journalist or publisher would have foreseen
the possibility that there had not been consent to disclose the information and would have
guarded against the possibility of publishing the material without consent.59 Accordingly, he
considered them liable in damages.
The result of the minority judgments is that the usual fault principles apply, except in
respect of media defendants, who may also be exposed to liability if their conduct is
negligent. In such cases, the fault element is intention. In line with general principles, once a
plaintiff proves that his or her privacy was violated, courts then presume that the media
defendant intentionally committed this violation. Other defendants (that is, those who are not
media defendants) can rebut this by showing that there was no intention, but media
defendants have to show not only that they did not intend the violation, but also that they
were not negligent. Only then is the presumption of intention rebutted.
 Our law on this point is in a state of flux. One interpretation is that fault can now take the
form of either intention or negligence, because the practical effect of judgments like this one
is that media defendants have to rebut both a presumption of intention and a presumption of
negligence. Another view is that there is still only one presumption – of intention – and the
negligence in question does not refer to the negligent conduct, but to the defendant’s state of
mind (that the media defendant had negligently believed that publishing the private
information was not wrongful in the circumstances, yet a reasonable media defendant would
have realised that such publication was or might be wrongful in the circumstances).
Accordingly, the media defendant cannot claim to have lacked intention, because it failed to
do what a reasonable media defendant would have done in the circumstances.

28.3 Juristic persons and the right to privacy

Historically, the actio iniuriarum was available to protect the personality rights of natural persons
only, based on the traditional acceptance that artificial or juristic persons cannot have rights that are
closely associated with being human. By implication, it followed that the actio iniuriarum was not
available to juristic persons, because they have ‘no feelings to outrage or offend’. 60 However, in
Universiteit van Pretoria v Tommie Meyer Films (Edms) Bpk,61 the Court gave some recognition to
a university having a right to privacy and a right to identity, because on appeal the Court accepted
(without deciding) that the university had a right to privacy, dignitas in the sense of esteem, and
fama that could be infringed in certain circumstances.62 This was confirmed in Financial Mail (Pty)
Ltd v Sage Holdings Ltd,63 where the Appellate Division indicated that:
as a matter of general policy, the courts have, in the sphere of personality rights, tended to
equate the respective positions of natural and artificial (or legal) persons where it is possible
and appropriate for this to be done.64

In privacy and identity cases, which do not involve wounded feelings, legal persons should be
entitled to sue.65
With the advent of a constitutional era, the Financial Mail position was confirmed. This is in
line with section 8 of the Constitution and in Investigating Directorate: Serious Economic Offences
v Hyundai Motor Distributors (Pty) Ltd: In re Hyundai Motor Distributors (Pty) Ltd v Smit NO 66
the Constitutional Court confirmed that juristic persons have a right to privacy:
The right to privacy is applicable, where appropriate, to a juristic person … . Juristic
persons are not the bearers of human dignity. Their privacy rights, therefore, can never be
as intense as those of human beings. However, this does not mean that juristic persons are
not protected by the right to privacy. Exclusion of juristic persons would lead to the
possibility of grave violations of privacy in our society, with serious implications for the
conduct of affairs … .

28.4 The constitutional right to privacy

Section 14 of the Constitution protects persons’ right to privacy by prohibiting unlawful searches
and seizures of their persons, homes, property or possessions, as well as interference with the
privacy of their communications. As drafted, the right appears to be narrower than the common-law
right, but eventually Constitutional Court judgments may indicate that the right has a broader
content. However, section 14’s primary purpose is not to provide a basis for compensation, as cases
based on this right have focused primarily on the validity of laws. For example, in Mistry v Interim
National Medical and Dental Council of South Africa,67 the Constitutional Court declared section
28(1) of the Medicines and Related Substances Control Act 101 of 1965, which authorised
unwarranted entry into private homes and rifling through intimate possessions, invalid. (It has since
been repealed.) More recently, in Estate Agency Affairs Board v Auction Alliance (Pty) Ltd,68 the
Constitutional Court held that certain provisions of the Estate Agency Affairs Act 112 of 1976 and
the Financial Intelligence Centre Act 38 of 2001, which allowed Estate Agency Affairs Board
officials to enter and search premises without a warrant (non-routine ‘targeted searches’), were
unreasonably broad and unreasonably violated the right to privacy of the persons whose premises
were searched.69
Bernstein v Bester 70 provides the starting point for understanding the constitutional right to
privacy. In this case, the applicants challenged the constitutionality of enquiries under section 417
and section 418 of the Companies Act 61 of 1973, which allowed the summoning and examination
of persons regarding a company’s affairs. One aspect of their challenge was that such enquiries
invaded witnesses’ privacy because the sections forced them to:
• Go to a place where they did not want to be
• Produce private books and documents that they wanted to keep confidential
• Reveal confidential information that they wanted to keep private
• Give incriminating evidence, both orally and by producing documents
• Participate in the hearings without being heard on the decision that subjected them to the
mechanism.71

Ackermann J took the view that no right can be considered absolute:72

Privacy is acknowledged in the truly personal realm, but as a person moves into communal
relations and activities such as business and social interaction, the scope of personal space
shrinks accordingly.

It is only in the ‘truly personal realm’ that we can expect courts to consider extensive privacy
protection as reasonable, compared with a privacy expectation in the context of ‘communal
relations and activities’. Even though the applicants’ subjective expectation of privacy was clearly
invaded when they were forced to disclose or reveal information that they clearly wished to keep
private, it was not a reasonable expectation for the following reasons:73
The establishment of a company as a vehicle for conducting business on the basis of limited
liability is not a private matter. It draws on a legal framework endorsed by the community
and operates through the mobilisation of funds belonging to members of that community.
Any person engaging in these activities should expect that the benefits inherent in this
creature of statute, will have concomitant responsibilities. These include amongst others, the
statutory obligations of proper disclosure and accountability of shareholders. It is clear that
any information pertaining to participation in such a public sphere cannot rightly be held to
be inherent in the person, and it cannot consequently be said that in relation to such
information a reasonable expectation of privacy exists. Nor would such an expectation be
recognised by society as objectively reasonable.

From this, we can see that there is some similarity between the constitutional right to privacy and
its common-law counterpart.
The constitutional right to privacy can be broadly divided into cases that involve personal
autonomy, informational privacy and intrusions. Personal autonomy cases are those in which
individuals can make personal decisions about such interests as their family relationships and
private life. For example, in Case v Minister of Safety & Security; Curtis v Minister of Safety &
Security,74 the applicants were charged with contravening section 2(1) of the Indecent or Obscene
Photographic Matter Act 37 of 1967 (now repealed). They had been found with a variety of video
cassettes that contained sexually explicit material. The Constitutional Court found that section 2(1)
of the Act sought to impose a ban on possessing erotic material kept within the privacy of a
person’s home. The private realm of a person’s home is a protected zone of privacy, and therefore
section 2(1) clashed with the constitutional provision. Accordingly, it was invalid. Similarly, with
respect to a person’s sexual orientation, and the practice of sodomy in particular, the Constitutional
Court noted in National Coalition for Gay and Lesbian Equality v Minister of Justice: 75
Privacy recognises that we all have a right to a sphere of private intimacy and autonomy
which allows us to establish and nurture human relationships without interference from the
outside community. The way in which we give expression to our sexuality is at the core of
this area of private intimacy. If, in expressing our sexuality, we act consensually and without
harming one another, invasion of that precinct will be a breach of our privacy.

In Teddy Bear Clinic for Abused Children v Minister of Justice and Constitutional Development 76
the Constitutional Court held that sections 15 and 16 of the Criminal Law (Sexual Offences and
Related Matters) Amendment Act 32 of 2007, which criminalised consensual sexual conduct
between children from the ages of 12 to 16, violated their constitutional rights to dignity and
privacy.77
The protection of the privacy of information further limits the ability of persons to gain,
publish, disclose or use information about others without their consent.78 For example, in C v
Minister of Correctional Services,79 prison officials obtained a prisoner’s blood for HIV testing
without his consent, which the Court declared unlawful, as the blood was drawn and the test
performed without the prisoner’s informed consent.
The Bernstein notion of privacy proved its usefulness in privacy of information cases that
focus specifically on the extent to which we should restrict collecting, using and disclosing private
information. In Mistry v Interim National Medical and Dental Council of South Africa 80 a member
of the public gave information that concerned a possible violation of the law to the Medical
Council. The Medical Council conveyed this information to the person statutorily responsible for
carrying out regulatory inspections for protecting public health. Inspection powers in terms of the
Medicines and Related Substances Control Act 101 of 1965 were phrased widely enough to permit
inspectors:
to enter any home where aspirins, ointments or analgesics happen to be … and, once there,
… (to) inspect not only medicine cabinets or bedside drawers, but also files which might
contain a person’s last will and testament, private letters or business papers.81

By authorising intrusion into the ‘inner sanctum’, the Act permitted violation of privacy 82 and, as
noted above, the Constitutional Court declared the legislation invalid. In doing so, the Court set out
the following factors to consider when assessing the informational aspect of the right to privacy: 83
• Whether the information was obtained in an intrusive manner
• Whether it was about intimate aspects of the applicant’s personal life
• Whether it involved data provided by the applicant for one purpose, which was then used for
another
• Whether it was circulated to the press, general public, or persons from whom the applicant
could reasonably expect that such private information would be withheld.

The Constitution also protects the right of individuals to decide who should enter their homes, and
further protects individuals from unauthorised intrusions into their homes by officers of the State
and other uninvited persons.84

PAUSE FOR Social media

REFLECTION Social media usage often poses challenges to privacy rights. The ease
with which information posted online can be accessed and shared
without requiring the permission of the original person who created the
post means that an individual’s privacy can easily be infringed. For
example, an online user shares a person’s posts and pictures with
persons other than those for whom the post was originally intended
without the consent of the person who posted the information. Privacy
may also be infringed where other users post pictures they took
themselves which present another person in an unflattering light.
The unauthorised sharing of images or posts on social media could
constitute either a disclosure of private facts or an intrusion on the private
realm. As seen above, disclosures or intrusions are not prima facie
wrongful, and will only be so if the legal convictions of society would
consider the posting of such information or images to be wrongful.
A court determines the question of wrongfulness by examining the
legal convictions of the community. This raises questions as to who forms
part of the community and what the legal convictions of the community
are. It could be argued that the real-world community and the online
community are two very distinct and different entities. Online
communities are often more diverse and geographically scattered.
However, when a court decides whether the harm in a particular instance
was caused wrongfully, it will judge the matter in accordance with the
sense of justice in the jurisdiction in which the conduct occurred.
The conduct of individuals in the real world often differs from their
actions in the online world. People often reveal more private information
and act more freely (and even recklessly) in the online environment than
they do in a real-world environment. Could this mean that the legal
convictions of people in the real world are different from those in the
online world? Should the conduct of users of social media not simply be
judged in accordance with the ordinary principles of delict? (It may be
that people act recklessly in divulging personal information online simply
because they are not aware that much of what they post online can be
viewed and shared by others. People often assume that messages
posted to social media are private and confidential, when in fact they are
not. Moreover, the ease with which messages can be posted to or shared
on social media means users often act impulsively, not thinking through
their actions carefully.)
The judgment in Mhlongo v Bailey 85 indicates that in deciding whether
a breach of privacy has occurred a court will look at the individual’s
previous habits. This could mean that a court is more likely to hold that a
wrongful breach of privacy has occurred in respect of an individual who
posts very few photos or personal information online, in contrast to a
person whose postings are prolific.
While one may sue for an infringement of privacy which has occurred
online, in reality such action is expensive, time-consuming, and often fails
to provide appropriate and effective relief. Other practical steps, such as
simply asking the website owner or social media network to remove the
offensive posts, are usually more effective. Many service providers for
websites and social networking sites have policies which regulate how
offensive pictures can be removed. In addition, an aggrieved person can
ask the online user who breached that person’s privacy to remove the
offending post. If all such measures fail and the aggrieved person
successfully litigates to compel the person who posted the offensive
material to remove it,86 the court will take into account the offender’s
 refusal to remove the material when deciding on an appropriate remedy,
including the extent of a damages award and costs.

28.5 Conclusion
The right to privacy is the right to be left alone and to keep personal and confidential facts about
oneself within the private domain. The right to privacy may also be understood as relating to
personal autonomy – the right of each person to decide for himself or herself what personal facts
should be made public. Personal autonomy includes highly personal decisions about one’s body
and reproductive health, such as the decision whether or not to receive medical treatment or
terminate a pregnancy.
While privacy is a stand-alone fundamental right enshrined in the Constitution, it is also a
common-law personality right that has long been protected by the actio iniuriarum. Accordingly,
damages are recoverable in delict when there has been a wrongful and intentional interference with
the right to privacy, either by an intrusion on the private realm or the public disclosure of private
facts. The plaintiff must prove that the right has been interfered with and should the court be
satisfied that the right has been disturbed, the onus will be on the defendant either to justify this
infringement, by proving that the interference was in the circumstances not wrongful, or to show
that it did not occur intentionally. Lack of intention to injure can be established by proving that
there was no direction of the will to infringe the right to privacy, or that in the circumstances the
defendant was not conscious of the wrongfulness of his or her conduct.
The fault requirement for invasion of privacy actions remains, for the time being, intention –
for both media and non-media defendants. However, the law on this point appears to be in a state of
flux. It is possible that in future the fault requirement for the invasion of privacy will be aligned
with that of the defamation action, and that negligence will play a role in claims involving media
defendants.

1 Constitution of the Republic of South Africa, 1996, section 14.

2 See, for example, Jansen van Vuuren v Kruger 1993 (4) SA 842 (A) at 849 and Bernstein v Bester 1996 (2) SA 751
(CC) paras 68–73.
3 Financial Mail (Pty) Ltd v Sage Holdings Ltd 1993 (2) SA 451 (A).
4 National Media Ltd v Jooste 1996 (3) SA 262 (A).
5 NM v Smith (Freedom of Expression Institute as Amicus Curiae) 2007 (5) SA 250 (CC).
6 See the discussion of the minority opinions in NM v Smith (Freedom of Expression Institute as Amicus Curiae) 2007
(5) SA 250 (CC) below.
7 Warren and Brandeis ‘The Right to Privacy’ (1890–1891) 4(5) Harvard Law Review at 193.
8 Case v Minister of Safety & Security; Curtis v Minister of Safety & Security 1996 (3) SA 617 (CC) para 91.
9 Investigating Directorate: Serious Economic Offences v Hyundai Motor Distributors (Pty) Ltd: In re Hyundai Motor
Distributors (Pty) Ltd v Smit NO 2001 (1) SA 545 (CC) at 557.
10 Bernstein v Bester 1996 (2) SA 751 (CC).
11 Greeff v Protection 4U h/a Protect International 2012 (6) SA 392 (GNP) at 406–407.
12 McQuoid-Mason The Law of Privacy in South Africa (1978) at 170–199.
13 De Fourd v Town Council of Cape Town (1898) 15 SC 399.
14 R v Holliday 1927 CPD 395.
15 S v A 1971 (2) SA 293 (T).
16 S v I 1976 (1) SA 781 (RA).
17 Gosschalk v Roussouw 1966 (2) SA 476 (C); Minister of Justice v Hofmeyr 1993 (3) SA 131 (A).
18 Financial Mail (Pty) Ltd v Sage Holdings Ltd 1993 (2) SA 451 (A).
19 D v K 1997 (2) BCLR 209 (N).
20 1971 (2) SA 293 (T).
21 2004 (2) SA 658 (T).
22 Para 39.
23 Electronic Communications and Transactions Act 25 of 2002; Regulation of Interception of Communications and
Provision of Communication-Related Information Act 70 of 2002.
24 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 102.
25 1998 (1) BCLR 46 (D).
26 Section 3.
27 Section 4.
28 Section 5.
29 Interferences with the protection of the personal information of a data subject set out in section 73.
30 Section 99(1).
31 Section 99(3). See Chapter 33 for a discussion regarding POPI and the principle of vicarious liability.
32 Mhlongo v Bailey 1958 (1) SA 370 (W).
33 National Media Ltd v Jooste 1996 (3) SA 262 (A).
34 Jansen van Vuuren v Kruger 1993 (4) SA 842 (A).
35 NM v Smith (Freedom of Expression Institute as Amicus Curiae) 2007 (5) SA 250 (CC).
36 Dutch Reformed Church Vergesig v Sooknunan 2012 (6) SA 201 (GSJ) para 78.
37 Jansen van Vuuren v Kruger 1993 (4) SA 842 (A).
38 Financial Mail (Pty) Ltd v Sage Holdings Ltd 1993 (2) SA 451 (A).
39 1958 (1) SA 370 (W).
40 At 371–372.
41 1993 (4) SA 842 (A).
42 1996 (3) SA 262 (A).
43 At 270H–J.
44 At 271A–D.
45 At 271G–272B.
46 At 272A.
47 2007 (5) SA 250 (CC).
48 1996 (3) SA 262 (A).
49 Para 41.
50 Para 42.
51 Para 143.
52 Para 48.
53 Para 54.
54 Para 28.
55 2007 (5) SA 250 (CC).
56 2007 (5) SA 250 (CC).
57 Langa CJ at paras 98–99 and O’Regan J at para 182.
58 Para 179. Previously National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) and Khumalo v Holomisa 2002 (5) SA
401 (CC) followed a similar approach to defamation cases.
59 Para 111.
60 Die Spoorbond v South African Railways; Van Heerden v South African Railways 1946 AD 999 at 1011.
61 1979 (1) SA 441 (A).
62 At 455, 456 and 458.
63 1993 (2) SA 451 (A).
64 At 461F.
65 See the discussion of Media 24 Ltd v SA Taxi Securitisation (Pty) Ltd (AVUSA Media Ltd as Amici Curiae) 2011 (5)
SA 329 (SCA) in Chapter 30, where the Court addressed the question whether, and on what basis, juristic persons can
sue for defamation.
66 2001 (1) SA 545 (CC) at 557D–G.
67 1998 (4) SA 1127 (CC).
68 2014 (3) SA 106 (CC).
69 Paras 40–43. See, also, Gaertner v Minister of Finance 2014 (1) SA 442 (CC), where the Constitutional Court
similarly held that a statutory provision that allowed officials to conduct warrantless non-routine targeted searches of
any premises unjustifiably infringed the right to privacy (paras 68–70 and 74).
70 1996 (2) SA 751 (CC).
71 Bernstein v Bester 1996 (2) SA 751 (CC) para 136.
72 Para 67.
73 Para 85.
74 1996 (3) SA 617 (CC).
75 1998 (12) BCLR 1517 (CC) para 32.
76 2014 (2) SA 168 (CC).
77 Paras 52–58 and 59–64.
78 McQuoid-Mason ‘Privacy’ in Chaskalson, Kentridge, Klaaren, Marcus, Spitz and Woolman (Eds) Constitutional Law
of South Africa, 2 ed Part II The Bill of Rights Vol 3 Chapter 38 (1996) at 18–11.
79 1996 (4) SA 292 (T).
80 1998 (4) SA 1127 (CC).
81 1998 (4) SA 1127 (CC) para 21.
82 Para 23.
83 Para 51.
84 S v Madiba 1998 (1) BCLR 38 (D).
85 1958 (1) SA 370 (W).
86 As happened, for example, in Heroldt v Wills 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ).
 Chapter 29

Infringements of identity

29.1 Introduction

29.2 The common-law right to identity

29.3 Recent cases concerning the right to identity

29.4 Conclusion

29.1 Introduction
As pointed out in Chapter 27, the right to dignity is a broad concept that includes the right to
privacy. In turn, the right to privacy has for some time been thought to incorporate the right to
identity. More recently, courts have recognised the right to identity as a free-standing right which,
like the right to privacy, falls within the concept of dignity in the broad sense.

29.2 The common-law right to identity

Interferences with identity can take two forms:

• Placing a person in a false light (in other words, presenting a person in a manner which creates
a false sense of who that person is) and/or
• Misappropriation of a person’s identity (for example, using a person’s name and/or likeness
without justification, usually for commercial advantage or gain).

Sometimes an infringement of identity can simultaneously involve both placing a person in a false
light and misappropriation of the person’s identity (it can be difficult to draw a clear line between
these concepts). In addition, an interference with identity can sometimes simultaneously constitute
an insult in the narrow sense and an infringement of the right to privacy.
 PAUSE FOR False light
REFLECTION A person is placed in a false light when someone makes false statements
about him or her which are neither defamatory nor insulting, so as to
create a false picture of the person in the public eye.
In Kidson v SA Associated Newspapers Ltd 1 a newspaper published
a false story and photographs under the headline ‘97 Lonely Nurses
Want Boyfriends’. The nurses had previously consented to their
photographs being taken for fundraising purposes to build a recreation
hall near the nursing training centre. Some of the nurses were married,
and others were engaged. A married nurse sued for defamation and
iniuria. The Court found that the nurse had been insulted, and awarded
damages for an infringement of dignity. Clearly, although embarrassing,
the story and photographs were not defamatory in the absence of any
secondary meaning (implication) or innuendo.
McQuoid-Mason points out that this is a false-light case and,
accordingly, an invasion of privacy.2 (Others view it as an infringement of
identity.)

Appropriation
Appropriation occurs where a person’s image or likeness is used, without
his or her consent, for profit – for example, using a photograph for an
advertisement without the permission of the person pictured in the
photograph.
In O’Keeffe v Argus Printing and Publishing Co Ltd,3 the plaintiff, a
popular radio presenter at the time, had agreed a newspaper could
publish her photograph as part of a news article. However, the
newspaper used the photograph in an advertisement for a gun
manufacturer (the photograph was of the plaintiff holding a crossbow at
an archery range). The plaintiff claimed this violated her personality
interests and, in particular, constituted a violation of her dignity. The
Court found in her favour, holding that to use a person’s photograph and
name without his or her consent for advertising purposes may reasonably
constitute offensive conduct on the part of the user.4
This is an obvious case of appropriation, but, as with the Kidson case,
the principles illustrated in the judgment are no longer current. The Court
framed its discussion in terms of the concept of dignitas and did not
expressly refer to the subsidiary rights to identity and privacy.

The jurisprudence that concerns personality rights has in recent years become more sophisticated
and precise, and the prevailing view is that a person’s identity is a distinct personality right,
separate from rights in respect of dignity and privacy. As explained in Chapter 4, a person’s right to
identity includes those aspects of, for example, one’s name or image that distinguish one person
from the next.5 Instances of false light and appropriation involve primarily infringements of
identity, and therefore it is incorrect to equate them with infringements of privacy. In neither of
these instances is the law attempting to protect the private sphere; by its very nature, a person’s
identity extends beyond the private domain.
The principles that apply in identity cases follow the same pattern as in privacy cases. The
right to identity is an independent personality right, but claims fall under the broad umbrella of
infringements of dignitas.6 The concepts are different, however, because proof of contumelia is not
a prerequisite for establishing a claim for violating identity.7 So, in line with general principles, for
liability to arise, there must be (a) a factual violation of a person’s identity that was (b) wrongful
and (c) intentional.
A factual violation of a person’s identity occurs only when aspects associated with a person’s
name or likeness are used outside their sphere or scope. Using an employee’s name or image in a
company newsletter would normally not constitute using that person’s identity outside its proper
sphere or scope, and no harm would therefore have occurred in such an instance. Given the nature
of the right, in that it is not associated with personal feelings, both natural and juristic persons are
capable of having an identity and courts can assess its violation objectively.
Once the plaintiff has established the factual violation of the right, inferences of wrongfulness
and animus iniuriandi arise, which the defendant must rebut in order to avoid liability. A court
would accordingly have to determine whether the violation has occurred in a legally acceptable or
legally unacceptable way (for example, with or without consent), an enquiry that involves applying
the usual policy considerations.8 After this, courts must enquire into the fault element, which, in
line with principle, must take the form of intention. However, the issues regarding negligent
liability that currently arise in defamation and privacy cases will in due course be extended into this
sphere.

29.3 Recent cases concerning the right to identity

Violating identity may, but will not necessarily, coincide with violating other personality rights,
such as privacy or dignity. In theory, an insulting infringement of identity would also constitute an
additional infringement of dignity, and a violation of a private aspect of a person’s identity (for
example, unwanted publicity) would amount to an additional infringement of one’s privacy.
However, in practice it is likely that only one action for infringement of identity will be brought,
with the existence of contumelious conduct or invasion of privacy serving as factors that are
relevant when assessing damages.

Grütter v Lombard9
Grütter and Lombard had been partners in a firm of attorneys called ‘Grütter and Lombard’.
When their partnership ended, Grütter began practising with another attorney as ‘Grütter and
Grobbelaar’, while Lombard continued to practise under the old partnership name. Grütter
asked Lombard not to continue using his name, but Lombard refused to do so. Grütter then
sought a court order, which the Supreme Court of Appeal granted. In doing so, the Court
made the following observations:
• ‘… features of personal identity are indeed capable (and deserving) of legal protection’.10
• There is a variety of personal rights that ‘are included in the concept of dignitas in the context of the actio
injuriarum’.11
• ‘The interest that a person has in preserving his or her identity against unauthorised exploitation seems to me to be
qualitatively indistinguishable and equally encompassed by that protectable “variety of personal rights”.’12
• ‘… not every intrusion upon those protectable rights of personality will necessarily constitute an injuria. Whether a
particular act constitutes a wrongful (or unlawful) violation, and thus an injuria, must necessarily be determined by
considerations of legal policy as in the case of any civil wrong. For, an individual who chooses to live in a
community cannot expect always to be shrouded in anonymity. One can envisage various circumstances in which
considerations of public policy will justify conduct that impinges upon features of a person’s identity.’13
• ‘I can see no such considerations that justify the unauthorised use by the respondents of Grütter’s name for their own
commercial advantage. What is conveyed to the outside world by the use of Grütter’s name is that he is in some way
professionally associated with the respondents, or at least that he is willing to have himself portrayed as being
associated with them, which … is a misrepresentation of the true state of affairs for which there can be no
justification.’14

Kumalo v Cycle Lab (Pty) Ltd15

A picture was taken of a television personality in a cycling store without her consent, and the
image was then used in an advertisement for the store. The Court broadly applied the
 reasoning in the Grütter case and held that the celebrity’s right to identity had been unlawfully
and intentionally infringed. The Court further pointed out that the use of the plaintiff’s image in
the advertisement without her consent also constituted an interference with her rights to
privacy and dignity.16 In particular, there had been an unlawful misappropriation of identity, as
her image had been used by the store for its own commercial gain without justification.17 The
store had directed its will towards using the image and also must have known, or at least
reasonably foreseen, that what it was doing was wrong. Therefore, it had acted with animus
iniuriandi.18

Cele v Avusa Media Limited19

At the time the cause of action arose the plaintiff was a minister in the KwaZulu-Natal
provincial government. He claimed that the Sowetan newspaper had defamed him by
publishing two articles about him and a digitally altered image of him which depicted him as a
sheriff in the ‘Wild West’. A picture of his head had been superimposed on an image of a real
person, but it was clear that the image was not a true depiction of the plaintiff. In the
alternative, he claimed that the digitally altered image had violated his right to dignity. The
plaintiff was affronted by the image in part because it allegedly constituted an interference
with his identity, in that he had been presented in a false light by the ‘mixing’ of images of him
and another person. The image had also been used without his consent.

The Court held that the fact that the image had been used without the plaintiff’s consent did
not give rise to an iniuria. Given that the image was of a politician (a public figure) and used in
relation to an article dealing with an important public issue, it could not be said that the use of
the digitally altered image without his consent was wrongful in the circumstances:

I am of the view that the mere unauthorised publication of plaintiff’s image or

likeness in the circumstances under consideration, i.e. where a politician’s image
has been used to create a caricature for purposes of commenting on his or her
public statements on issues that are in the public interest, and not for
commercial purposes, would not be considered wrongful. Having regard to the
importance of striking the correct balance between the right to freedom of
expression and the right to dignity, public policy dictates that in these
circumstances the defendant should not be held liable for publishing the
photographic image (albeit altered) of the plaintiff.20

PAUSE FOR The right to identity

REFLECTION Despite what appeared to be a clear directive in Grütter v Lombard,
courts are still struggling to separate the dignity, privacy and identity
concepts. In Wells v Atoll Media (Pty) Ltd 21 a surfing magazine published
a photograph of a 12-year-old girl without her knowledge, authority or
consent. Although the photograph depicted the girl from behind, many
people apparently recognised the girl in the photo. The photograph
appeared in a section of the magazine entitled ‘dishing up the photo
feast’. The magazine’s cover carried the statement ‘100% pure filth
photos inside’, and the published photograph, which bore the caption ‘all-
natural Eastern Cape honey’, was over-stamped with the word ‘filth’. The
magazine also placed an advertisement containing the photograph on the
national television programme SuperSport. The publication of the image
exposed the girl to a barrage of lewd and crude suggestions via text
messages sent to her phone, drawing unwelcome attention of a sexual
nature that affected her private life. The girl’s claim was for defamation
and invasion of privacy.
Regarding the claim for invasion of privacy and infringement of
dignity, the Court stated:22
 [I]n the context of this case therefore the appropriation of a person’s
likeness or image for the commercial benefit or advantage of another
may well call for legal intervention in order to protect the individual
concerned. That may not apply to the kinds of photographs or television
images of crowd scenes which contain images of individuals therein.
However, when the photograph has been employed, as in this case, for
the benefit of a magazine sold to make profit, it constitutes an
unjustifiable invasion of the personal rights of the individual, including
the person’s dignity and privacy.
• In this passage, the Court seems to focus on the wrongful appropriation as the
crux of the violation of privacy, as opposed to the requirement of insult. No
mention is made of violation of identity, even though the case was concerned
with the unjustified and unauthorised use of an individual’s image for financial
gain or economic exploitation (appropriation of identity) and a
misrepresentation of the individual (placing a person in a false light) – similar to
a misrepresentation that an attorney is a partner in a firm when in fact this is
not the case. Which of the rights is more appropriate in such instances: dignity,
privacy or identity? Does the unauthorised use of a person’s image necessarily
violate a person’s physical and intimate space, or wound a person’s feelings?
Could it be that the judgment is correct, given the special facts of the case, but
that the principles should not be generally applied?
• The Court also reiterated the comments in Grütter regarding a person’s right to
identity, indicating that the right can be infringed if the attributes of the person
are used in an unauthorised manner which (i) cannot be reconciled with the
true image of the individual (misrepresentation) and (ii) violates the privacy of
the person. Is this restatement in line with general principles and Grütter?
• Although both judges generally accepted that features of personal identity are
deserving of delictual protection, Nugent JA in Grütter23 and Davis J in Wells
differed in their approaches to the violation of the right to identity. For Nugent
JA, it is the right to identity that is primarily violated when a person’s image is
used without permission for advertising purposes. Davis J, on the other hand,
decided that such use is a violation of a person’s right to decide for oneself who
should have access to one’s image and likeness – something that goes to the
root of individual autonomy and privacy. It is therefore arguable that, whereas
Nugent JA asserted identity in such cases as a separate right, Davis J did not.
Is Davis J’s interpretation in line with the Grütter judgment?
• What is the implication of the decision in Wells for media freedom? As with any
action involving the infringement of a subjective right (in this case, the use of a
person’s image), the rights of identity, dignity, and privacy of an individual must
be weighed against the user’s right to freedom of expression and media
freedom. The Court did not engage with this important question.

24
PAUSE FOR To what extent would Wells be relevant in instances of online privacy
REFLECTION infringements?
• What would the legal position be if someone were to use private photographs
or images from another person’s Facebook account to promote or market a
product?
• What would the position be in circumstances where a person’s private
photographs are used outside the commercial sphere – for example, where
someone creates a Facebook account and uses another person’s image as his
or her own to create the impression that the picture is actually that of the
account owner?
Would the ordinary general principles of delict be adequate in such
instances, or do the courts need to adjust the rules specifically for
conduct in the online world?
29.4 Conclusion
The right to identity is a component of the right to dignity in the broad sense. Although at first
thought to be a form of the right to privacy, courts have in recent years come to recognise that the
right to identity exists as a separate and free-standing personality right. When a person’s name or
likeness is used wrongfully and intentionally outside its proper scope, this can give rise to an action
for damages in delict, despite the fact that the conduct in question is not objectively insulting and
does not result in an interference with the right to privacy.
In line with general principles, the plaintiff is first required to prove that he or she has suffered
harm through having been placed in a false light or having had aspects of his or her identity
misappropriated for commercial gain. Once the plaintiff has proved this fact, the defendant must
rebut presumptions of wrongfulness and fault.
Cases emphasise the importance of balancing the right to identity with the right to freedom of
expression. In particular, courts seem less inclined to award damages for the factual infringement of
the right to identity in circumstances where the person bringing the action is a public figure and the
infringement occurred in the course of legitimate comment on issues of public interest.

1 1957 (3) SA 461 (W).

2 McQuoid-Mason ‘Invasion of privacy: common law v constitutional delict – does it make a difference?’ (2000) Acta
Juridica 227 at 231 suggests further that insult was not a necessary requirement of the action and the nurse should
have succeeded even in the absence of contumelia.
3 1954 (3) SA 244 (C); see also Grütter v Lombard 2007 (4) SA 89 (SCA).
4 At 249A.
5 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 373–374.
6 Grütter v Lombard 2007 (4) SA 89 (SCA) para 12.
7 In Grütter v Lombard 2007 (4) SA 89 (SCA) para 13 the Court noted that ‘not every intrusion upon those protectable
rights of personality will necessarily constitute an injuria.’
8 Grütter v Lombard 2007 (4) SA 89 (SCA) para 13.
9 2007 (4) SA 89 (SCA).
10 Para 8.
11 Para 12.
12 Para 12.
13 Para 13.
14 Para 13.
15 2011 JDR 1480 (GSJ).
16 Paras 23 and 24.
17 Para 22.
18 Paras 33–38.
19 [2013] 2 All SA 412 (GSJ).
20 Para 51.
21 [2010] 4 All SA 548 (WCC).
22 Para 49.
23 Paras 8 and 9.
24 [2010] 4 All SA 548 (WCC).
 Chapter 30

Infringements of reputation

30.1 Introduction

30.2 Who can sue for defamation?

30.3 Publication
30.3.1 Who is responsible?
30.3.2 The communication
30.3.3 Understanding

30.4 Defamatory matter

30.4.1 Step 1: Determining |the meaning of the words or conduct
30.4.1.1 The primary or ordinary meaning
30.4.1.2 The secondary or hidden meaning – innuendo
30.4.1.3 Quasi-innuendo
30.4.1.4 The reasonable reader, listener or viewer
30.4.2 Step 2: Determining whether the meaning is defamatory
30.4.2.1 The benchmark

30.5 Reference to plaintiff

30.6 The presumptions

30.7 Wrongfulness

30.8 Animus iniuriandi

30.1 Introduction
The law of defamation is mainly concerned with protecting the fama (the good name or reputation)
of both natural and juristic persons. A well-known definition of reputation is from De Villiers CJ in
O’Keeffe v Argus Printing and Publishing Co Ltd:1
A person’s reputation is … that character for moral or social worth to which he is entitled
amongst his fellow-men.

A person’s reputation refers to the good name the person enjoys in the estimation of others – that is,
what others think of that individual as a person. Reputation is thus distinct from dignity in that there
is an essential difference between what others think of a person (reputation) and what a person
thinks of himself or herself (dignity).
Interestingly, the Constitution of the Republic of South Africa, 1996 does not specifically
protect the right to reputation by name in the same way that it protects the rights to dignity and
privacy. However, this does not mean that there is no constitutional right to a person’s reputation.
The Constitution protects reputation via the right to dignity, and courts have indicated that the right
to dignity includes the right to reputation.2 While there is no doubt that the value of respecting
another person’s dignity underpins the right to reputation, the idea that the right to reputation is
included in the right to dignity creates a theoretical anomaly: at common law, dignitas and fama are
distinct concepts, yet constitutional jurisprudence views them as one. Perhaps the best way to deal
with the anomaly is to mirror the thought processes we find in the common-law treatment of
privacy and identity cases, which is that, although a separate right, the right to reputation falls
under the umbrella category of dignity.
Therefore, the law of defamation seeks to protect a person’s right to an unimpaired reputation
or good name against any unjust attack. In doing so, the right to reputation is often pitted against
the right to freedom of speech and expression.3 Society values these rights equally highly, so courts
often have to balance them and protect reputation in a manner that does not obstruct freedom of
communication. So, for example, the opinion of the Constitutional Court is:
It has long been recognised in democratic societies that the law of defamation lies at the
intersection of freedom of speech and the protection of reputation or good name.4

This aspect is particularly important for the press and other news media.5
Defamation is the (i) wrongful and (ii) intentional (iii) publication of (iv) defamatory material
that (v) refers to the plaintiff. Therefore, the essence is that the material must reach someone other
than the person to whom it refers, and that the nature of the communication must be defamatory.
The nature of the material must be likely to damage the person’s reputation by having the
propensity or tendency to lower the person’s good name and standing in the opinion of others.6 For
liability to arise, there must be (a) a factual violation of the right to reputation that was (b) wrongful
and (c) intentional. A person’s reputation is factually disturbed when (i) defamatory material (ii)
that refers to him or her (iii) is published. The plaintiff bears the onus of establishing that factual
violation of the right to reputation has occurred – that is, elements (i)–(iii). In doing so, two
rebuttable presumptions arise: one of wrongfulness and one of intention (animus iniuriandi). The
onus is then on the defendant to rebut those preliminary (prima facie) conclusions by establishing a
defence on a balance of probabilities.7 Failure to do so leads to liability.

PAUSE FOR Criminal defamation

REFLECTION The delictual (civil) action for defamation should not be confused with
criminal defamation. As pointed out in the chapter on harm,8 the law of
delict exists to compensate the plaintiff for harm caused wrongfully and
culpably, proved by the plaintiff on a balance of probabilities. It is not
 intended to punish the wrongdoer. However, criminal defamation is the
wrongful, intentional publication of defamatory matter about another
person. Each element must be proved beyond a reasonable doubt.
Accordingly, the requirements for a successful prosecution for criminal
defamation are more onerous than those for a successful claim in delict
for damages for defamation. The North Gauteng High Court has recently
held that criminal defamation does not unreasonably interfere with the
right to freedom of expression protected by section 16 of the Constitution,
and is therefore not unconstitutional.9

Figure 30.1 Problem-solving flow chart

30.2 Who can sue for defamation?

All natural persons, including assisted minors, may sue for defamation. Artificial persons may also
do so, but only in restricted circumstances. Courts are prepared to protect the business reputation of
trading corporations,10 but not any reputation that they might have outside their trade or business
spheres. Non-trading corporations can also sue for defamation in instances where the material
refers to the way in which they conduct their affairs and is calculated to cause financial loss.11
Courts treat political parties12 and trade unions13 similarly to non-trading corporations.14

PAUSE FOR Artificial persons

REFLECTION In all those instances that involve artificial persons, one is concerned with
whether they can sue for general damages under the actio iniuriarum.
Accordingly, they do not have to prove any actual financial loss or
prejudice – only that such loss or prejudice is likely. If these persons can
show actual loss, then the appropriate action would be the lex Aquilia, but
only to the extent that they can prove loss. This distinction is important,
because the fault element of these two actions differs. For liability to arise
for defamation, where no actual loss need be proven, the infringement of
the reputation must have been intentional. For compensation in respect
of actual loss that was incurred, fault can take the form of either intention
or negligence. It is, of course, possible that both actions could exist side
by side, but only where animus iniuriandi is present.
Is the basis upon which courts recognise artificial persons’ fama
correct? The business reputation referred to is no different from goodwill,
a concept that is well recognised and accounted for as an asset in
business balance sheets. As such, the business reputation forms part of
the corporation’s universitas and a person should claim for any reduction
in reputation under the lex Aquilia.
Financial loss or prejudice is not the principle upon which reputation
of natural persons is based, so why should it be the principle upon which
reputation of juristic persons is established? Why should a university, for
example, not have a claim when a person publishes defamatory material
about its employment policies, or the behaviour of its SRC, and it cannot
show any negative effect on its financial status?15 The first case in which
this issue was commented upon involved a trading corporation that was
suing for patrimonial loss, that is, goodwill, not reputation. This obiter
restriction was perpetuated and subsequently expanded to non-trading
corporations. A rule, crafted for a particular type of juristic person (which
is questionable in itself) has been extended beyond its intended scope to
areas in which its application is not appropriate. Is the tail not wagging
the dog?
Media 24 Ltd v SA Taxi Securitisation (Pty) Ltd 16 provides some
conceptual clarity. SA Taxi Securitisation (Pty) Ltd, a juristic person and a
trading entity, had been defamed by an article published by a newspaper
owned by Media 24 Ltd. The Supreme Court of Appeal held by a majority
that a trading corporation, like any other person, was entitled to the
protection of its reputation and to sue for defamation,17 and that such an
entity is entitled to general damages even though it is a juristic person
that is incapable of being wounded in its feelings.18 Further, if the entity
wished to recover special damages for consequential patrimonial loss,
such as loss of profits, it would need to recover these separately by
means of the Aquilian action.19 The Court found that to entitle trading
corporations to sue for general damages for injury to their reputations
would not constitute an unreasonable interference with the constitutional
right to freedom of expression.20
In Gold Reef City Theme Park (Pty) Ltd v Electronic Media Network
Ltd; Akani Egoli (Pty) Ltd v Electronic Media Network Ltd 21 the Court
reached similar conclusions.22 In particular, the Court held that a trading
 corporation can sue for damages for defamation under the actio
iniuriarum:

For over a century our courts have recognised that trading

corporations have a personality right to fama worthy of
protection, and can sue for defamation. While it may not be
identical to that of a human plaintiff, a trading corporation
does have a reputation which may be disparaged … The
plaintiff in a defamation action, including a trading
corporation, does not only seek to recover economic loss,
but also reparation for a wrong inflicted. I do not see how a
trading corporation’s right to sue for defamation under the
actio iniuriarum would be inconsistent with the
Constitution. In my view, the common law requires no
development to bring it into harmony with the spirit,
purport and objects of the Bill of Rights.23

The Court also agreed that a claim for patrimonial damages instituted
against a trading corporation should be brought under the Aquilian action,
so as not to infringe the right to freedom of expression unreasonably:

In respect of loss of profit, the enormity of the awards will

certainly have a chilling effect on the freedom of
expression. In my view, this cannot serve the interests of
democracy, and the disproportionality may well constitute
an unjustifiable limitation to the right of freedom of speech.
By limiting such claims to the lex Aquilia, the plaintiff is not
non-suited, but the onus is shifted so as not to burden the
defendant with the presumptions of unlawfulness and
intention under the actio iniuriarum. There seems to be
ample support for this proposition.24

It is not possible for the government (central government departments and local authorities) to sue
for defamation,25 the reason being that to allow the government to sue to protect its governing
reputation would unduly inhibit freedom of speech and expression in a democracy. However,
courts have limited the scope of this restriction to government entities. This means that cabinet
ministers, politicians, and public officials can sue in respect of infringements of their personal
reputations.26 The same applies to judicial officers.27

PAUSE FOR Defamation and organs of state

REFLECTION In Post & Telecommunications Corporation v Modus Publications (Pvt)
Ltd28 the Court denied an organ of state the right to sue for defamation.
The reason was that the organ of state was more similar to a government
department than to other corporations. The Court also set out criteria for
deciding on which side of the line an organ of state would fall. An organ
of state would not be able to sue for defamation where its activity is
closely associated with government work and with democratic
governance, but when it does have the right to sue, the principles
applicable to corporations would apply. In some instances (for example,
universities or the Gender Commission), the distinction might be easier to
draw, while in others (for example, Transnet, Telkom, Eskom, South
 African Airways and the Reserve Bank), courts would have to determine
the position of each entity individually. In fact, whether some of the
entities mentioned above are indeed organs of state is also moot.29

30.3 Publication
Since reputation involves what others think of someone, a person’s right to reputation is factually
interfered with only when another person communicates defamatory material referring to that
person and makes it known to at least one other person. So publication, which is the required
conduct element in defamation, is vital. If nothing has been published, the plaintiff at best has a
claim for infringement of dignity.
Whether material has been published is a question of fact. Publication usually occurs through
written or spoken words, but even non-verbal conduct such as gestures can constitute conduct for
the purposes of a defamation claim. In Pieterse v Clicks Group Ltd 30 the Court appeared to accept
that someone could be defamed where a public search of a handbag in a shop in front of other
shoppers suggests that the person whose handbag is being searched may have been guilty of
shoplifting.31 However, for publication to occur it is not sufficient merely that a person presents the
information to another. The person who reads or hears the information, or who observes the
conduct, must also understand the meaning of what is being conveyed. So, the element of
publication has two components: the act of making the material known to another (the
communication), and the understanding and appreciation on the part of the recipient of the
material’s meaning and significance.

30.3.1 Who is responsible?

Often a number of people are involved in the publication process. In these instances, subject to one
exception, every person who has contributed to the publication may be sued. For example, the
following could be considered as having published the material:
• In cases that involve written material: the author, editor, newspaper owners, printers,
publishers, distributors and libraries
• In cases involving oral material: the speaker, the broadcaster and, where appropriate, the
presenter or talk-show host
• With internet publications: the host, network provider, service provider, as well as a person
who is ‘tagged’ in an online post and does not remove the ‘tag’. 32

An exception is made in respect of intermediaries – that is, those who participated in the
distribution process but did no more than provide the infrastructure for transporting the material.
Examples include the postal services in the case of letters or in the case of internet publications, the
owners of the land or sea cables along which messages are sent from one place to another.
When a person repeats material that another has published, both the original and the
subsequent communication would amount to publication. So, it is not a defence to say that the
material is already in the public domain. A person who repeats the material will also be liable.33
Whether the first person will be held liable for the subsequent republication of the material by
someone else will depend upon whether any of the following circumstances are present: 34
• Where a person intended the republication of the material
• Where, even though republication was not intended, the repetition by another is a natural and
probable result of a person’s communication
• Where, even though republication was not intended, the other person had a moral duty to
repeat the material to others.

In cases of unintended communications or where the material is published by mistake, the issue is
not whether there has been communication to another person – that has already occurred – but
whether courts should attribute the communication to the defendant. In such instances, courts
would hold that the defendant had published the material, even though the publication was not
intended, where it was reasonably foreseeable that publication would occur – in other words, where
the publication is a natural and probable result of a person’s conduct. In Pretorius v Niehaus 35 the
Court found that a person could not reasonably have expected, in the particular circumstances, that
the plaintiff’s wife would have opened the letter that had been addressed to the plaintiff, and so the
defendant was held not to have published the material.

30.3.2 The communication

Communication of material from one person to another generally takes the form of positive
conduct. The most common forms are oral communication or written communication in books,
newspapers, pamphlets and cartoons.36 However, communication also includes instances of
publication using technologies such as email 37 and internet websites,38 which include, among other
things, postings to newsgroups and chat rooms. Content posted on social media and social
networking sites, such as Facebook, Twitter and Instagram, would also constitute the
communication of information for the purposes of a defamation action. In circumstances where
there was a duty to act positively, a person can also publish by omission, for example, where an
internet service provider fails to remove defamatory material from its servers after being requested
to do so. In this case, the publication element is satisfied.39
Persons often publish directly to a specific person or group of persons, in which case it is
reasonably certain that someone has heard or read the communication. Where this is not obviously
the case, publication may sometimes be presumed. Some of the circumstances in which this
rebuttable presumption arises are:
• Where the material is distributed to the public in any form, usually a newspaper, book, journal
or magazine, but also by postings to websites, newsgroups, internet bulletin boards and social
media, including social networking sites
• Where words are spoken within earshot of others40
• Where the material appears on a postcard or in a telegram.41

The presumption merely helps the plaintiff to prove publication in cases in which it might be
uncertain, and the defendant may still show that no other person was aware of the material – for
example, by showing that no other person had accessed the website or social media/social
networking page to which the material had been posted.

PAUSE FOR Defamation on social-media and social-networking sites

REFLECTION Millions of South Africans use social-media and social-networking sites
every day, using online platforms such as Facebook, Twitter, Snapchat,
Instagram and LinkedIn.42 Globally, Facebook has over 800 million users
and is the most popular social-networking site worldwide.43 It is also the
most popular social-networking site in South Africa, with 9,4 million
users.44 In recent years, a slew of cases has come before the courts in
which it has been alleged that persons using online social-media or
social-networking sites45 have posted defamatory content to such
applications.46 In general, our courts have taken the view that posting
 defamatory matter to the internet or social media constitutes publication
which can give rise to a claim for damages in delict if such posting is
proved to be defamatory, wrongful and intentional. In other words, the
normal rules pertaining to the publication of defamatory matter apply
equally to the internet and social-media and social-networking sites.47
• Ketler Investments CC t/a Ketler Presentations v Internet Service Providers’
Association 48 held that listing the applicant as a ‘spammer’ (a sender of bulk
unsolicited electronic communications) on a ‘Hall of Shame’ on the
respondent’s webpage, in the context of the overall wording of that webpage,
was defamatory of the applicant.49
• Dutch Reformed Church Vergesig v Sooknunan 50 held that a person who is the
owner, creator and user of a Facebook wall is the person capable of regulating
access to and censoring postings to it. As the person who created the
opportunity for postings to be made to the wall, that person is to be regarded as
the publisher of what is posted to it, and will accordingly be held responsible in
law for defamatory postings on the wall.51
• Isparta v Richter52 held that, even though the second defendant was not the
author of the defamatory posts in issue, as he had been ‘tagged’ in them and
had not removed the ‘tag’, he was in part responsible for them.53
• In RM v RB 54 the respondent posted various statements on Facebook which
were defamatory of the applicant. Friends of the respondent commented on the
post and were critical of the applicant. The applicant sought an order that the
respondent remove the defamatory statements from Facebook, and that the
respondent be interdicted from publishing further defamatory statements about
him on Facebook or in any other way. The Court granted the first order, as the
statements in question were clearly defamatory and without justification.55 It
refused the interdicts, however, as these were too drastic and, if granted, would
constitute an unreasonable interference with the respondent’s right to freedom
of expression.56

Heroldt v Wills57
Wills posted defamatory information regarding Heroldt to Facebook. Heroldt sought an
interdict ordering Wills to remove the defamatory posts from the social networking site. Wills
contended Heroldt was not in the circumstances entitled to an interdict ordering her to
remove the information from Facebook, because if Heroldt believed he had been defamed,
he was entitled to sue for damages.
The Court held that, given that it was relatively easy for online information to be removed
from the site or sites to which it had been posted, Heroldt should not be denied a right to the
removal of the offensive matter. The Court also drew a distinction between social media and
news media (whether print or online news media), and held that the courts may legitimately
treat the former differently from the latter with regard to the question of ordering the removal
of online information.58
The Court rejected the view that, where offensive matter has been posted to Facebook,
the proper response of the aggrieved party should be to approach Facebook itself for relief.59
The Court held that by issuing an interdict that Wills was to remove the posting, it would be
providing a remedy to a problem for which there was no other relief to the same effect. An
interdict would resolve the issue without the needless expense, drama, trauma and delay
that were likely to accompany an action for damages in a case such as this one.60
Accordingly, an interdict was issued to remove the posted material.61 The Court further
advised those who post offensive material on social media to remove such matter when
requested to do so, or face legal consequences.62
This case contains a useful description of Facebook and other social-networking sites, as
well as a discussion of some of the legal implications of such social media, particularly
regarding the right to privacy.63 Regarding the impact of social-media and social-networking
sites on the law, the Court held that:
 the pace of the march of technological progress has quickened to the extent that
the social changes that result therefrom require high levels of skill not only from
the courts, which must respond appropriately, but also from the lawyers who
prepare cases such as this for adjudication.64

The law recognises a significant exception to the publication rules where communications are
considered to be privileged. Two such categories for which there is case authority are
communications between spouses,65 and communications between client and attorney.66

PAUSE FOR Privileged communication

REFLECTION Communication between spouses is privileged, but communication by an
outsider to a spouse is not. Two cases that illustrate this point are
Whittington v Bowles 67 and Pretorius v Niehaus.68 Similarly,
communication between attorney and client is protected, but a
communication to another person’s attorney is not.

30.3.3 Understanding
The second element of publication is that the recipient of the communication must understand and
appreciate the material’s meaning and significance. So, for example, there is no publication where a
person makes a defamatory statement to another person in a language that the other person does not
understand, where the person is unaware of the true meaning or significance of the material, or, in
particular instances, where the person is deaf or blind.69 This does not mean that a reader or listener
must grasp the meaning immediately. Publication is satisfied if a person hears the information, does
not immediately grasp its meaning and significance, but does so later. In other words, regardless of
the time between communication and understanding, publication is complete only if and when the
recipient of the information discovers the meaning.70

30.4 Defamatory matter

The second component of factual violation of a person’s fama is that the content of the published
material must be defamatory. This would also satisfy the harm aspect of the violation of the
personality interest. Although they do not always do so explicitly, courts follow a two-step
approach when determining whether the material is defamatory. First they have to establish the
meaning of the material (what the words meant), and then whether the words, as meant, conveyed a
defamatory imputation.71

30.4.1 Step 1: Determining the meaning of the words or conduct

The words or conduct can convey either a primary meaning (the ordinary meaning) or a secondary
meaning (a hidden meaning or innuendo).

30.4.1.1 The primary or ordinary meaning

The primary or ordinary meaning of words or conduct is the natural meaning, both expressed and
implied, that an ordinary or reasonable reader, listener or viewer would give to the words or
conduct, with specific reference to the context and circumstances in which the words have been
published. It is not necessarily the dictionary meaning.72 So, for example, the word ‘cow’ may mean
a female bovine animal in one context, but in another, when spoken with reference to a woman, it
may mean a rude, disagreeable person. In each context, the primary meaning, that which a
reasonable reader would attribute to it, is different.
Be careful not to confuse the implied meaning of a word or phrase with an innuendo (see
below). The implied meaning is a meaning that is inherent in the words themselves. An ordinary
reasonable person of average intelligence would be able to grasp the implied meaning of the words
in question simply by construing them in context. He or she would not need any additional
knowledge of special facts or circumstances in order to understand the defamatory meaning of the
words. When the primary meaning, both express and implied, of words or conduct is defamatory,
the words or conduct are sometimes described by the courts as defamatory per se.

30.4.1.2 The secondary or hidden meaning – innuendo

Words can sometimes assume meanings that are very different from their ordinary meaning. These
additional secondary meanings, or innuendos, usually arise in special circumstances known to both
the communicator and the communicatee.73 In some instances, the plaintiff may be required to
identify those persons whom he alleges knew the special facts. The plaintiff does not need to prove
that those people understood the words to bear the alleged defamatory meaning, but simply that
they knew the facts that might have led them to understand the words in the sense that is alleged to
be defamatory. It is then for the Court to determine whether the words would in fact have been
understood by reasonable people, with the requisite knowledge, to bear the meaning that the
plaintiff alleges.

30.4.1.3 Quasi-innuendo
Our law also recognises what is referred to as a ‘quasi-innuendo’. As the name implies, it is not a
proper innuendo in the sense that there is a second, hidden meaning. In fact, a quasi-innuendo has
nothing to do with the meaning of the material, but more with whether the meaning that the words
bear is ‘more defamatory’ in a particular context than would otherwise have been the case. When a
quasi-innuendo is present, the words have an additional sting to them. For example, a claim that a
religious leader was guilty of sexual misconduct would be more defamatory than a similar claim
made about an ordinary member of the public with no particular claim to moral righteousness.

30.4.1.4 The reasonable reader, listener or viewer

A core feature in establishing the meaning of words is the concept of the reasonable reader, listener
or viewer. In a multicultural, diverse society, where ordinary men and women have different
temperaments and outlooks, we should attempt to understand who this standardsetter is. In
Demmers v Wyllie 74 Muller JA expressed the concept of the ordinary, reasonable reader in the
following way:
From the above it is clear, I think, that the words ‘reasonable person’ or ‘reasonable man’
referred to in the decisions cited is a person who gives a reasonable meaning to the words
used within the context of the document as a whole and excludes a person who is prepared
to give a meaning to those words which cannot reasonably be attributed thereto.

When arriving at the natural and ordinary meaning of words, a reasonable reader is guided by his or
her own general knowledge, outside of any sophistication or specialised training, especially in law
or logic. The reader will avoid the elaborate analytical attention that a lawyer might give to the
meaning of the words or conduct. Therefore, it is crucial for courts to avoid any intellectual
analysis of the contents, or what might be implied by the material’s content. Courts should also not
assume that the ordinary reader has a tendency to analyse in this way, or an ability to recall more
than an outline or overall impression of what he or she has just read.
However, there is some subjective aspect given to the concept of the reasonable reader. Courts
have acknowledged that the reasonable reader of a publication aimed at a particular target market
would have the characteristics of that target market. So, the ordinary reader of the Sunday Times
Extra 75 would be black, and a reader of a financial publication would have ‘a somewhat higher
standard of education and intelligence and a greater interest in and understanding of financial
matters than newspaper readers in general have’. 76 Similarly, the reasonable reader of a
sophisticated magazine dealing with politics, current affairs and the arts would be different from
the reasonable reader of a supermarket tabloid focusing on celebrities and entertainment.
When applying the test, courts must take into account the manner in which an ordinary
reasonable reader, listener or viewer would read and understand the material. Courts generally
accept that readers of newspaper articles do not scrutinise or over-analyse, and they often draw
conclusions after having skimmed, browsed, or glanced quickly over the material.77 Reasonable
listeners and viewers (for example, of the radio or television), should be treated differently from
reasonable readers, because the printed or written words can be re-read, often more than once.78
This principle would obviously not apply in circumstances where a person was listening to a
podcast, as in this instance it would be possible for the listener to hear the material again and satisfy
himself or herself as to its true meaning.
One should also keep in mind the context in which the material was published. In other words,
one must not assess the offending statement or conduct in isolation, but instead in the context of the
entire publication, and together with any favourable parts. The document or context must be viewed
as a whole. This means that a defamed person cannot select a part, or parts, of the material to
support the meanings that he alleges the publication bears, and ignore other parts that qualify or
negate the alleged meaning.79 If contained in a book, one should consider the effect of the entire
book, and if in a newspaper, other articles might also be relevant. (Sometimes an entire page of a
newspaper is filled with a number of stories about the same news event or issue, and if this is the
case all the articles will need to be considered together in order to determine whether or not the
material in question is defamatory.) 80 The circumstances in which the words are used, and the tone
in which they are uttered, are also factors to take into account. Therefore, courts are required to take
a holistic view of the matter in deciding whether or not the matter is defamatory.
The points we have made in relation to assessing the meaning of words and conduct apply
equally to assessing whether the material is defamatory, or whether it is merely meaningless abuse
or robust criticism.81

30.4.2 Step 2: Determining whether the meaning is defamatory

Once courts have determined the primary and/or secondary meaning of the material, they then have
to decide whether that meaning is defamatory. Therefore, it is necessary to determine what type of
material society considers defamatory, what would not be defamatory, and how courts ascertain the
difference.

30.4.2.1 The benchmark

The benchmark criterion that South African courts apply was first set out in the English case of Sim
v Stretch: 82
 Would the words tend to lower the plaintiff in the estimation of right-thinking members of
society generally?

However, as was pointed out in Mohamed v Jassiem,83 immediately prior to these remarks the Court
in Sim v Stretch 84 noted that one has to consider the reaction of the person or class of persons who
would receive the material. As we describe below, the test to apply is a combination of these two
statements.
Following English law, our courts at first tended to ask whether the imputation was capable of
lowering the plaintiff in the estimation of right-thinking members of society generally.85 It is
important to note that the plaintiff does not need to prove that the matter actually had the effect of
lowering the plaintiff’s reputation; the plaintiff need merely show that the matter would be likely to
lower the reputation of the plaintiff in the estimation of reasonable, right-thinking persons.86
However, this test is not appropriate for South Africa’s multicultural and diverse society. What, for
example, is the correct thinking that is referred to, and who constitutes the general society that
determines what is right-thinking? Courts have diluted this test somewhat by interpreting that what
‘right-thinking members of society’ think is determined by what an average person of ordinary
intelligence who subscribes to the norms and values of the Constitution would think.87 Also,
‘society generally’ might in some instances include a section of society. So, where society
generally might not consider a statement to be defamatory, but a well-defined sector of society
might, the material could be viewed from that sectoral perspective.
In Mohamed v Jassiem 88 the Court pointed out that in a heterogeneous society, where the
statement complained of was alleged to be defamatory only in the eyes of a particular segment of
society that constitutes a distinctive group:
[I]t is the reaction of ‘right-thinking’ members of that segment of society which becomes the
yardstick rather than that of ‘right-thinking’ members of society generally.89

The Court found that right-thinking members of society generally would not regard calling a
Muslim person an Ahmadi sympathiser as defamatory, as the term would have no meaning for the
vast majority of South Africans who are not of the Muslim faith, but Muslims would. This sectional
approach applies only where there is a distinctive community group that forms part of the South
African population. Courts will not consider the opinions of a section of the general public whose
narrow views depart from the general norm.90
The expression ‘lowering a person in the estimation of others’ usually means that the statement
or imputation by conduct injures the reputation of another, by exposing the person to hatred,91
contempt 92 or ridicule,93 or by reflecting upon the person’s moral character.94 In so doing, harm to
the person’s reputation is caused because people in the community in general would think less of
the person. In addition to exposing the claimant to hatred, contempt, or ridicule, a publication
would be defamatory if it causes or has the tendency to cause others to shun and avoid the defamed
person. Regarding reflections on moral character, it is important to bear in mind that morality
changes with the times. For example, today, unlike in the past, it might not be defamatory or an
impairment of dignity to say that a woman had a child out of wedlock and lived with the child’s
father for two years without marrying him.95
In summary, defamatory material usually falls into four broad categories: 96
• Imputations against a person’s moral character or lifestyle, 97 for example, commenting
negatively on the character and public life of a politician 98
• Imputations that arouse hatred and ridicule, for example, making contemptuous remarks
regarding a person’s race or racial views99
• Communications that cause shunning or avoiding, such as making derogatory remarks
regarding a person’s physical100 or mental disposition101
• Impairments of professional or business reputation, such as casting aspersions on a person’s
professional competence that is calculated to cause the person to be regarded with contempt by
others.102

In our democracy, robust debate and criticism form a key component of political activity. To this
extent, our courts are far more flexible with respect to debate and exchanges within the political
arena. Politicians are expected to display more robustness than the average private person, and
should therefore exercise caution in hastily bringing their matters in this context to court.103 In
Kgothule v Majonga 104 the Court pointed out that ‘[p]oliticians must realise that high trees catch
more wind’. 105 Our courts are duty-bound at all times to observe openness, transparency and
accountability, but at the same time protect dignity and privacy. If, however, an improper motive or
dishonourable conduct is imputed to a person who is alleged to have defamed another, the
reasonable bounds of freedom of expression have been exceeded.
There is a special category of cases that involve quasi-innuendos. As the name implies, there is
no innuendo in these cases, but the situation is similar to an innuendo. Whereas an innuendo
changes the meaning of the words, giving them a special primary meaning, a quasi-innuendo does
not, but instead highlights the especially defamatory nature of the words. In such instances, a
plaintiff alleges that there are circumstances that render the meaning of the words more hurtful –
that is, ‘more defamatory’. So, a precondition for a quasi-innuendo is that the meaning of the words
has already been established and that it is defamatory. The focus then is on whether the words that
would ordinarily be defamatory become even more defamatory because there is a ‘sting in the tail’.
The following two examples illustrate this concept. It is defamatory to say that someone is
dishonest, but to call a judge dishonest adds a sting to the imputation, making the statement more
defamatory. It is not defamatory to call someone a ‘ladies’ man’, so any sting that might exist in
calling a king or a president a ‘ladies’ man’ does not render the words defamatory.106
Sindani v Van der Merwe107
The action for defamation arose from an article published in Rapport under the headline
‘Sepeng se breier glo as rassis uitgekryt’ (‘Sepeng’s coach said to be reviled as a racist’). The
appellant alleged that the article would be understood by the readers to mean, in essence,
that he was a racist who conducted himself in a reprehensible manner. The question raised
on appeal was: ‘Is it defamatory of a black man to impute to him that he abused a white man
by calling him “white trash”?’ 108
The appellant relied on the ordinary, primary meaning of the word and did not allege any
innuendo or secondary defamatory meaning arising in the light of special circumstances. The
Court followed the accepted two-stage enquiry, by first establishing the natural or ordinary
meaning of the words, and then considering whether that meaning is defamatory. It
explained the test to apply in the first instance as follows:109

The ordinary meaning of the words under consideration does not necessarily
correspond with their dictionary meaning. The test to be applied is an objective
one, namely what meaning the reasonable reader of ordinary intelligence would
attribute to the words read in the context of the article as a whole. In applying
this test it must be accepted that the reasonable reader will not take account
only of what the words expressly say but also what they imply … . It must also
be borne in mind that ‘the ordinary reader has no legal training or other special
discipline’ and that ‘if he read the article at all would be likely to skim through it
casually and not to give it concentrated attention or a second reading. It is no
part of his work to read this article, nor does he have to base any practical
decision on what he reads there … .’ Consequently, a court that has of necessity
subjected a newspaper article under consideration to a close analysis must
 guard against the danger of considering itself to be ‘the ordinary reader’ of that
article … .

The respondents relied on the dictionary meaning of the expression ‘white trash’ and
submitted that the term constituted mere abusive language without any racial connotation to
it. The Court disagreed, holding that when the abusive term ‘trash’ is coupled with the word
‘white’, and used with reference to a white person, it becomes racially charged and that the
ordinary reader would understand the term to be racially derogatory language:110

On a single perfunctory reading of the article the reasonable reader would

understand from it that the appellant was using the racially derogatory language
not as a shield but as a sword, without any apparent justification for doing so.111

The reasonable reader would understand from the article that it was informing its readers
that the appellant had reviled the first respondent as a racist and had addressed him in
racially derogatory language.
Having established the ordinary meaning of the words, the Court then proceeded to the
second stage of the enquiry:112

What the article attributes to the appellant is the gratuitous use of racially derogatory
language and racial vilification. Such conduct is regarded by right minded members of
South African society not only as conduct that is reprehensible but as something which
must, in accordance with constitutional imperatives, be eradicated. It follows that the
imputation of such conduct to another must be defamatory.

Le Roux v Dey; Freedom of Expression Institute and Restorative Justice Centre as

Amici Curiae113
The defamatory material complained of in this case was an image. Three schoolboys created
a computer-generated image of their school principal and deputy principal’s heads
superimposed on the bodies of two gay men photographed naked sitting side by side in a
sexually compromising position. The private parts/genitalia of the two naked men in the
picture were covered by the school crest. The manipulation of the image was amateurish and
obviously not a real depiction of the principal and the deputy principal. The image was
published and circulated within the school and seen by many pupils. The image was also
placed on the school noticeboard for approximately 30 minutes and then removed. Once the
schoolboys had been identified as the persons behind the act, they were disciplined by their
school at an internal hearing. The boys apologised to the principal, who accepted their
apology and took no further action. However, the deputy principal, Dr Dey, did not accept
their apology. Instead, he sued the schoolboys on the basis of two claims: one for
defamation, and the other for violation of his dignity.
The High Court upheld both claims. On appeal, the Supreme Court of Appeal upheld the
defamation claim and found that the additional claim based on an affront to his dignity was ill-
founded and required no further consideration. The schoolboys then appealed to the
Constitutional Court. The boys argued that the image was not defamatory. It was not
representative of reality and a reasonable viewer would not have understood it as conveying
any actual facts about the plaintiff. Furthermore, even if Dr Dey had felt subjectively that his
dignity had been impaired by the picture, this was not objectively justified.
Brand AJ held that the reasonable observer would infer some association between the two
teachers, on the one hand, and the ‘indecent situation’ described in the picture, on the
other:114

In short, the vision created is one of two promiscuous men who allowed
themselves to be photographed in what can only be described as a situation of
sexual immorality, which would be embarrassing and disgraceful to the ordinary
members of society … The manipulation of the picture in this way is obvious
 and crude. No reasonable person could ever think that the bodies on which the
faces are pasted were actually those of the principal and Dr Dey.

This raised the question – as the second leg of the two-stage enquiry – whether the picture,
thus understood, could be regarded as defamatory of Dr Dey. That in turn depended on
whether the message conveyed by the picture would probably undermine the esteem in
which Dr Dey was held by others – in other words, whether the reasonable observer would
regard the picture as likely to undermine the respect and good name enjoyed by Dr Dey. The
Court considered the context of the publication and concluded that the picture was distributed
among the teachers and pupils of the school. In his capacity as deputy principal, Dr Dey ‘was
a symbol of authority and discipline at the school’.115 In this context, the reasonable observer
would understand the image to associate the two teachers with the bodies and their
behaviour:

The whole purpose and effect of the association created by the picture is to
tarnish the image of the two figures representing authority, to reduce that
authority by belittling them and by rendering them the objects of contempt and
disrespect, and to subject these two figures of authority to ridicule in the eyes of
the observers who would predominantly be learners at the school.116

It followed that the average person would regard the picture as defamatory of Dr Dey:

The aim might have been to destroy Dr Dey’s image as a figure of authority but
the net effect was to belittle and humiliate him as a person, to represent him as
unworthy – or at least less worthy – of respect by the learners of the school,
which is a classic example of defamation.117

Brand AJ’s majority judgment has been criticised – in particular, for its heteronormative and
heterosexist bias118 in construing the image as one of:

two promiscuous men who allowed themselves to be photographed in what can

only be described as a situation of sexual immorality.119

The reasoning, it is said, is not in keeping with the Constitutional Court’s previous progressive
jurisprudence on gay rights and the rights of sexual minorities more generally.120
There were two dissenting judgments. The first is the joint judgment by Froneman J and
Cameron J, who found that Dr Dey was not defamed, but that his dignity was actionably
injured. The second dissenting judgment is that of Yacoob J, with whom Skweyiya J
concurred (for different reasons), but both of whom strongly grounded their reasoning on
child-law principles.
Consider the judgment of Froneman J and Cameron J. In their view, even if we were to
accept that the manipulated image crudely sought to create some association between Dr
Dey and the school principal in the situation the image sought to portray, and that it was an
attempt (albeit amateurish) to ridicule and undermine the authority of Dr Dey and the school
principal, this did not mean that the average reasonable person viewing the image in the
school context, where it was published, would regard the picture as defamatory. Their
reasoning was as follows:

Our common law recognises that people have different claims for injuries to
their reputation (fama) and to their own sense of self-worth (dignitas). Both are
affronts to the rights of personality, and although the Bill of Rights does not
always draw sharp lines between the two, the distinction is important to our new
constitutional order. It illuminates the tolerance and respect for other people’s
dignity expected of us by the Constitution in our public and private encounters
with one another. We may be deeply hurt and insulted by the actions of others,
 in calling or portraying us as what we have chosen, freely, not to be, or to keep
private, even though we are not defamed. It may be that the personal insult or
injury may not be considered, in the public eye, as something that harmed our
reputation. But within limits our common law, and the Constitution, still value
and protect our subjective feelings about our dignity. It is this difference
between private and public esteem that explains, in our view, why Dr Dey cannot
succeed in his defamation claim, but must do so in his dignity claim.121

Once it is clear that our law requires probable impairment of the right to
reputation (the public aspect of the constitutional right to dignity) before a
statement or image may be considered defamatory, general pronouncements
that statements or images that arouse hatred, contempt or ridicule are
defamatory are better understood. It then becomes clear that statements or
images that ‘were calculated or had the tendency or propensity to defame’ are
defamatory only if they objectively and as a matter of probability cause
Impairment of a plaintiff’s good name. If they do not, there is no defamation: but
there may be impairment of dignity.122

And this is what happened here: objectively, the prank did not impair Dr Dey’s reputation. On
the contrary:

… the contextually average reasonable school viewer, learner or teacher, knew

better: Dr Dey and the school principal were not promiscuous, they were
respected teachers.123

But he experienced it as a deep affront to his personal dignity. It may be noted that here, in
determining whether Dr Dey’s dignity was unlawfully infringed, as in determining whether the
publication of the image defamed him, an objective test of reasonableness is applied.124

This raises the question: can one find that the reasonable observer would not
have taken the image so seriously as to have thought less of Dr Dey, yet still
have considered Dr Dey’s sense of injury in finding the image insulting
reasonable? The answer is ‘Yes’. It must be emphasised that the two enquiries
are different, and hence that their outcomes may differ. The one reflects
inwardly, the other outwardly. In dignity claims, the injured interest is self-
esteem, or the injured person’s feelings. In defamation, it is public esteem or
reputation. And the objective reasonableness in a dignity claim is assessed in
relation to feelings of individual affront, not in relation to the audience that sees
the image or reads the statement as in a defamation claim. It is in our view plain
that the reasonable observer may conclude that, objectively seen, an affront did
not damage a person’s reputation, while at the same time concluding that,
objectively seen, the injury to that person’s feelings was palpable and
reasonably felt, and hence actionable.

In this sense, Froneman J and Cameron J give a clear exposition of the nuanced relationship
between reputation and dignity within the constitutional rubric.

Cele v Avusa Media Limited125

The plaintiff, a high-ranking politician, claimed the Sowetan newspaper had defamed him by
publishing two articles about him and a digitally altered image of him depicting him as a
sheriff in the ‘Wild West’. In the alternative, he claimed that the digitally altered image had
insulted him and in so doing injured his dignity. In the articles, the newspaper had alleged the
plaintiff had made various statements about the need for the police to ‘shoot to kill’ and ‘aim
for the head’ when dealing with violent criminals. The Court held that the statements had
indeed been made by the plaintiff and were not defamatory of him.126 Cumulatively, the
 articles and the image would have been understood by reasonable readers of the Sowetan to
mean the plaintiff was taking a tough stance on crime and that, like a sheriff in the Wild West,
he wanted criminals to be harshly dealt with by the police and brought to justice, either dead
or alive:

The altered image, in my view, is a depiction of the plaintiff as a law enforcer or

sheriff from the Wild West, and it would have been understood as such by
reasonable readers of Sowetan. The altered photo image taken together with the
caption and the contents of the article would have been understood to mean that
the plaintiff was taking a tough stance on crime and that, like a sheriff from the
Wild West, he wanted criminals to be harshly dealt with by the police and
brought to justice, either dead or alive.127

Such an understanding of the articles and the image would not have damaged the plaintiff’s
reputation. In reaching this conclusion, the Court took into account the fact that the plaintiff
was a seasoned politician who was required to display more robustness with regard to what
was published about him:128

I therefore remain of the view that the meaning of the 6 July article and the
altered image are not defamatory of the plaintiff. A reasonable reader of the
Sowetan, taking into account the context mentioned above, would know that the
plaintiff was a high ranking politician charged with ensuring safety and security
of the public in the province of KwaZulu-Natal; that he had a well-known history
of involvement in matters relating to safety and security; and that the statements
attributed to him, as well as the altered photo image were reported in the context
of a meeting at which one of the primary topics of discussion was violent
criminal activity which was affecting businesses in Umlazi township − where
crime was reportedly on the increase. The reasonable reader would not ‘think
less’ of the plaintiff for taking a tough stance on crime; for calling on the police
to deal harshly with criminals; and for encouraging them to use deadly force
when dealing with criminals. As is apparent from the numerous articles
contained in the trial bundle, there were varied responses to the stance that the
plaintiff took on violent crime and criminals. Some even lauded his stance. I am
accordingly of the view that the plaintiff’s claim, based on the article of 6 July
2007 read together with the altered photo image, is without merit because
neither the content of the article nor the altered photo image published is
defamatory of the plaintiff.129

With regard to the image, the Court held that a reasonable person in the position of the
plaintiff would not have felt insulted and humiliated by it.130

30.5 Reference to plaintiff

To establish that a person’s fama has been violated, a plaintiff must not only show that the
defamatory matter was published, but also that the defamatory matter referred to the plaintiff. This
establishes the causal link between the publication of the defamatory matter and the damage to the
plaintiff’s reputation.
The reference to the plaintiff can be direct or indirect (by implication). Where a person is
mentioned by name, this aspect usually poses no problem, although there might be others with the
same name, in which case a plaintiff would have to follow the same process as instances of indirect
reference. In such instances, a plaintiff must allege and prove the facts that make reference to him
or her personally, firstly, by proving that the words are capable of referring to him or her (this is a
question of law), and, secondly, that the words actually refer to the plaintiff (this is a question of
fact),131 in that an ordinary reasonable reader, listener or viewer would have understood that the
words complained of refer to the plaintiff personally.132
A statement may be defamatory of a wide class of people or a group of people – for example,
politicians, cabinet ministers, lawyers, trade unionists, black people, or white males. However, not
every person who falls within a class or group can sue for defamation. This is because there is no
group or class defamation in South Africa. To sue in such instances, the plaintiff must be able to
show that a reasonable reader, listener or viewer would have connected the plaintiff, as an
individual, to the defamatory statement. If the plaintiff belongs to a small group and as a result is
easily identifiable – for example, if the plaintiff is one of four trustees of a trust or a cabinet
minister in a cabinet of 15 members – there should be a sufficient connection for the plaintiff, being
a trustee or cabinet minister, to sue, even if the plaintiff is not mentioned by name. However, the
words must be capable of referring to each member of the group, and the plaintiff must show that
the words actually refer to him or her as an individual included in that group.133 In Sauls v
Hendrickse 134 a politician commented that office-bearers of a trade union were involved in causing
unrest.
The Court noted:135
This is not a case where reference was made to all the members of a group. The
statement refers simply to ‘office-bearers of NAAWU’. It does not in express
terms refer to all the office-bearers. Nor can such a reference necessarily be
implied. The position may have been different had it spoken of ‘the office-
bearers of NAAWU’, for that might have implied all. Seen in their proper context
the words ‘office-bearers of NAAWU’ only refer to some office-bearers … . Some
in that sense denotes an unspecified yet relatively limited number.

In other words, the question is whether the words refer to the plaintiff, in the sense that we can say
that he or she was personally pointed out. It follows that there must be something that points to a
particular plaintiff or plaintiffs and links him, her or them to the defamatory statement.
A Neumann CC v Beauty Without Cruelty International136
A poster showed a woman dragging a fur coat that was dripping blood with captions that read
‘It takes up to 40 dumb animals to make a fur coat’ and ‘If you don’t want millions of animals
tortured and killed in leg-hold traps, don’t buy a fur coat’. A furrier and seller of fur coats
sought to interdict further publication of the poster, and one of the issues was whether the
poster referred to the applicant. The Court gave a succinct outline of the principles involved:137

It is obvious that applicant is not named directly in the poster. It is also trite that
a plaintiff or applicant in a defamatory action must allege and prove that the
defamatory matter was published of and concerning him. It must refer to or
concern him personally … . It has been stated on many occasions that the test is
an objective one and it is whether the ordinary reasonable reader would have
understood the words complained of, in conjunction in this case with the
picture, to apply to the plaintiff or as in this case to the applicant … . This gives
rise to a two-stage inquiry. Firstly, whether the words (with the picture) are
reasonably capable of referring to the plaintiff or applicant. This is a question of
law and can be decided on exception. Secondly, and if the answer to the first
part is in the affirmative, whether a reasonable person would regard the words
as referring to the plaintiff or applicant. This is a question of fact on which
evidence would be admissible … .

The Court also said:138

 It is now, I think, well recognised that if a group or class is so small or so readily
ascertainable that what is said of the group or class is necessarily said of every
member of it, then a member of that group or class may be able to bring an
action if there is a defamation of the group or class. The reason for this, it would
seem, is that every member of the group or class, it being so small, is identified
in the libel or, as it has been put, is ‘individually aspersed and can sue and that
therefore any one of them can … .’ For this reason the Court has, for instance,
held that an allegation in a newspaper article of alleged defamatory conduct on
the part of a company could reasonably be interpreted as a reflection upon the
board of directors of the company and that, as there were only five directors on
the board, the article was capable of referring to each and every member of the
board. It was therefore held that the article was capable of referring to the
plaintiff who was one of the directors … . The position is very different in the
present case. The poster and its captions represent, in my opinion, a campaign
against a certain type of wearing apparel, i.e. fur coats and garments; it is an
attack on a fashion cult or a clothing fad. Those who have mounted that
campaign have directed their attack against all those … involved in the
production, supplying, manufacture or selling of such garments. This involves a
very large and wholly indeterminate body of persons. It is clearly not an attack
on any individual and it is emphatically not an attack on the applicant. It is, if
anything, an attack on a wide group or class in which it would be impossible to
identify applicant individually. It is a campaign against a cult; it is not an attack
on individuals.

The Court accordingly found that the applicant had not been sufficiently identified.

30.6 The presumptions

Publication of defamatory matter that refers to the plaintiff involves a violation of the plaintiff’s
reputational personality interest (the factual disturbance of the personality right). The plaintiff bears
the onus of proving that such violation has occurred. However, the violation of the right may or
may not have been wrongful, and it may or may not have been inflicted with animus iniuriandi
(intention to injure) and the onus is also on the plaintiff to prove the other two elements of
defamation: wrongfulness and intention. In these instances the law assists the plaintiff by
recognising, on proof of the violation of the plaintiff’s fama, two rebuttable presumptions: a
presumption that the defendant’s conduct was wrongful (or unlawful) and a presumption that the
defendant had acted intentionally.139 The nature of these presumptions was explained in Botha v
Mthiyane:140
The presumption of culpability relates to the defendant’s subjective state of
mind, i.e. a deliberate intention to inflict injury is presumed, whereas the
presumption of unlawfulness relates to objective matters of law and fact.

Therefore, in respect of the next two elements, we should bear in mind that it is up to the defendant
to prove, on a balance of probabilities:
• First, that the conduct was not wrongful, by showing that social policy favours his or her right
to free expression, or by showing that a recognised defence or ground of justification is
present. This is a full, and not merely an evidentiary, onus.141
• Second, that he had not been at fault, either because there was no direction of the will towards
defaming the plaintiff, or because the plaintiff was not conscious of the wrongfulness of his or
her conduct.
Media defendants can rebut the latter aspect of intention only if they can also show that they had
not been negligent. In other words, in circumstances where, objectively speaking, the defamation
was wrongful, a media defendant can escape liability only if it can prove on a balance of
probabilities that it did not act intentionally or negligently.

30.7 Wrongfulness
Once a plaintiff has proved that defamatory material about him or her has been published, courts
presume that the publication was wrongful, unless the defendant can show otherwise.142 The
criterion of reasonableness, also known as the boni mores of modern society or the legal
convictions of the community, is used to determine whether the defendant’s conduct is wrongful in
the circumstances. The accepted rules for determining wrongfulness, discussed earlier, apply. In
line with those rules, the enquiry into wrongfulness is, in effect, an enquiry into the relevant social
policy to determine whether the plaintiff should be allowed a claim. The question is whether the
sense of justice in the community, crystallised in the boni mores, would dictate that the plaintiff
should be successful in an action for defamation. In defamation cases, this usually requires courts
to balance two conflicting rights: the plaintiff’s right to reputation and the defendant’s right to
freedom of expression. Neither of the rights is more important than the other. In some instances, the
situation leans towards protecting the plaintiff’s interests; in others, society favours the defendant’s
interests. So, the facts of each case determine the result of the normative enquiry into wrongfulness
and, because both of these rights are also protected in the Bill of Rights (reputation is included
under the dignity umbrella),143 the Constitution plays an important role in determining social policy
and the outcome of the wrongfulness enquiry.
Over the years courts have concluded that in certain categories of factual situations freedom of
expression should prevail over the plaintiff’s rights, provided that particular criteria have been met.
These categories have been distilled into what we call grounds of justification or defences
excluding wrongfulness. The grounds most commonly associated with defamation actions are truth
and public benefit, fair comment, and privileged occasion, but the list is not a closed one. The
grounds of justification will be discussed in more detail in the next chapter.

30.8 Animus iniuriandi

Animus iniuriandi is the subjective intention to injure. As in the case of wrongfulness, courts
presume intention to be present once the plaintiff has proved publication of defamatory matter that
refers to the plaintiff. However, the defendant may rebut this presumption.
Animus iniuriandi is a subjective concept that comprises two components:
• The direction of the defendant’s will towards injuring the plaintiff’s reputation
• The defendant’s knowledge that to impair the plaintiff’s reputation is in the circumstances
wrongful or unlawful (knowledge of the wrongfulness of one’s conduct).

In Modiri v Minister of Safety and Security 144 the Supreme Court of Appeal held that, once a court
was satisfied the plaintiff had proved there had been publication of defamatory matter referring to
the plaintiff, the presumptions of wrongfulness and fault arise. The court should first ascertain
whether the defendant had acted with animus iniuriandi. If the answer to that enquiry is ‘no’, there
is no question of liability. If the answer is ‘yes’, the court should then enquire into whether or not
the damage to the plaintiff’s reputation had occurred in a wrongful manner. Neethling, Potgieter
and Visser are of the view that this approach is wrong, pointing out that a court should always
enquire into wrongfulness first. If the defendant’s conduct was not wrongful, there can be no
question of the defendant having acted with animus iniuriandi, as the defendant could not possibly
have been conscious of the wrongfulness of the conduct at the time he or she engaged in it.145
The defences mentioned in the fault chapter apply to defamation cases. However, the defences
negating intent that are most often associated with defamation actions are mistake, provocation and
jest.
The principle that animus iniuriandi is the fault requirement for defamation claims applies to
all defendants, except for the mass media, which term includes the press, radio and television, and
anyone whose business it is to publish, including owners, editors, printers, publishers and
distributors. At common law, all defendants were treated equally, which meant that mass media
defendants could escape liability if subjectively no intention was present – for example, because the
media defendant made a mistake. However, courts began to acknowledge that, because of the
greater potential for harm arising from defamation published in the mass media, as well as the ease
with which mass media could escape liability, special rules should apply to such defendants.
In Trimble v Central News Agency 146 the Appellate Division held that newspaper vendors (and
distributors) could escape liability if they could prove that they had not been negligent. This
approach was followed in Hassen v Post Newspapers (Pty) Ltd,147 where the Court introduced an
objective element into the enquiry and held that a mistake could serve as a defence against
intention, thus excluding consciousness of wrongfulness, but only where the mistake was not made
recklessly or negligently. Subsequently, Pakendorf v De Flamingh 148 took the matter a step further
and held that mass media were strictly liable for defamation. The effect of this decision was that a
media defendant could never raise a defence aimed at negating fault. Fault was no longer an
element of defamation perpetrated by the mass media, and if a plaintiff’s reputation were infringed
and the infringement was wrongful, then liability followed. The landmark decision in National
Media Ltd v Bogoshi 149 overturned the strict-liability regime. There is a fair amount of controversy
over which form of liability Bogoshi introduced in its place. Most would agree that our law is now
similar to that articulated in Hassen v Post Newspapers (Pty) Ltd, but there is no agreement as to
the exact effect of that position. Some commentators and courts suggest that Bogoshi introduced
negligence liability, while others believe that liability is still based on intention, but that media can
rebut the presumption of intention and escape liability if they can show that their conduct was not
negligent.

COUNTER Liability of mass media

POINT Views differ on the nature of the liability in mass-media cases. Neethling,
Potgieter and Visser,150 as well as Burchell,151 contend that Bogoshi
introduced a general principle that negligence suffices as a form of fault.
The Supreme Court of Appeal in Mthembi-Mahanyele v Mail & Guardian
152
supports this view:

… the form of fault in defamation actions against the press

is negligence rather than intention to harm.153

Midgley, on the other hand, contends that Bogoshi considered an

attenuated form of intention to be appropriate,154 and that the Court opted
for a reverse-onus approach in terms of which a defendant can attempt to
negate intention.155 One can trace the difference of opinion to different
interpretations of previous cases.
In Hassen v Post Newspapers (Pty) Ltd,156 the Court said:157

A defamation is not actionable if it was published in the

honest, though mistaken, belief in the existence of
circumstances which would have justified or excused its
 publication, but that is so only if the mistake is not
attributable to the recklessness or negligence of the
defendant … .

However, the Court also said:158

I am of the opinion that there was negligence, attributable

to the first defendant, which was causally connected with
the defamatory publication. That, in my view of the law,
founds liability.

In Bogoshi, the Court noted that Suid-Afrikaanse Uitsaaikorporasie v

O’Malley 159 distinguished between liability for defamation (which) ‘cannot
be founded upon negligence’ and the fact that ‘essentially on the ground
of lack of negligence, news distributors may escape liability for
defamation of which they were unaware’160 and commented:161

Some writers … are in favour of negligence being the basis

of liability and the judgment in Hassen v Post Newspapers
(Pty) Ltd … points the same way; but any suggestion that
liability for defamation can be founded on negligence was
rejected in the obiter dicta in O’Malley’s case. On the other
hand, O’Malley did not overrule the principle discussed at
the outset of this judgment that distributors can escape
liability if they are not negligent.

The Court then concluded:162

Defendants’ counsel, rightly in my view, accepted that there

are compelling reasons for holding that the media should
not be treated on the same footing as ordinary members of
the public by permitting them to rely on the absence of
animus injuriandi, and that it would be appropriate to hold
media defendants liable unless they were not negligent in
the circumstances of the case.

In Khumalo v Holomisa 163 the Constitutional Court noted:

In Bogoshi, too, the question of the rebuttal of intention

was considered. One of the aspects of animus injuriandi
(the intention to cause injury) is subjective intent which,
amongst other things, requires the person who made the
defamatory statement to have been ‘conscious of the
wrongful character of his act’. …

Hefer JA then considered whether media defendants

should be permitted to rebut the presumption of intentional
harm by establishing a lack of knowledge of wrongfulness,
even where that lack of knowledge was as a result of the
negligence of the defendant. He concluded that they should
not … .

In NM v Smith 164 the Constitutional Court, in a majority judgment, held

that cases involving the actio iniuriarum require intention.165 O’Regan J,
who preferred to extend the law to include negligence liability in that
case, nonetheless interpreted Bogoshi as follows:166
 However, the Court at the same time held that the press
could not rebut the presumption of intention that arises
upon proof of publication of defamatory material by simply
showing the absence of knowledge of unlawfulness (more
simply understood as subjective mistake). The press would
in addition have to establish the absence of negligence. In
so doing, the Court established that a media defendant
could not avoid liability for defamation unless it could show
that it had not acted negligently. This was a new
development in the actio injuriarum.

In Pieterse v Clicks Group Ltd 167 the Court expressed the view that, in
cases of suspected shoplifting, where the person suspected of
wrongdoing may be stopped, questioned and subjected to a search of his
or her person or bags, the fault requirement should be both intention and
negligence,168 as is the case with the mass media:

There appears to be no practical impediment to align the

requirement for fault in the actio iniuriarum with the Lex
Aquilia more closely in cases where a person purporting to
exercise authority at a shop accuses a customer of
shoplifting, irrespective of whether or not this is followed
by a request to search.169

Irrespective of our view as to what the cases have said in the past, is
there justification for the view that in a modern constitutional state, fault in
defamation cases, and perhaps for all actio iniuriarum cases, should
include negligence?

1 1954 (3) SA 244 (C) at 247–248.

2 Holomisa v Khumalo 2002 (3) SA 38 (T) at 62.
3 For a recent illustration of the classic contest between the right to dignity, on the one hand, and the right to freedom of
expression, on the other, see Mtyhopo v South African Municipal Workers Union National Provident Fund 2015 (11)
BCLR 1393 (CC), where Cameron J held that the Constitutional Court had jurisdiction to hear the matter because it
affected freedom of expression (para 25). The Court held that what the applicant had said about the respondent was
true and fair comment that could justifiably be expressed. Accordingly, the interdict restraining the applicant from
exercising his right to freedom of expression was set aside (para 42).
4 Khumalo v Holomisa 2002 (5) SA 401 (CC) para 26.
5 In Bosasa Operations (Pty) Ltd v Basson 2013 (2) SA 570 (GSJ), an action for defamation, the Court refused to
compel a journalist and the newspaper he worked for to reveal the sources of an article published about corruption in a
tender process, because to have done so would have undermined the right to freedom of expression of the journalist
and the newspaper (para 38).
6 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 91; The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae)
2011 (4) SA 191 (CC); 2011 (8) BCLR 816 (CC) para 19; Demmers v Wyllie 1980 (1) SA 835 (A) at 842A–C.
7 Neethling v Du Preez; Neethling v The Weekly Mail 1994 (1) SA 708 (A); Khumalo v Holomisa 2002 (5) SA 401
(CC).
8 Chapter 4.
9 S v Motsepe 2015 (5) SA 126 (GP) paras 39, 46 and 49–50.
10 G A Fichardt Ltd v The Friend Newspapers Ltd 1916 AD 1 at 5–6 and 9; Dhlomo NO v Natal Newspapers (Pty) Ltd
1989 (1) SA 945 (A) at 952–953.
11 Dhlomo NO v Natal Newspapers (Pty) Ltd 1989 (1) SA 945 (A) at 954.
12 Argus Printing and Publishing Co Ltd v Inkatha Freedom Party 1992 (3) SA 579 (A) at 600.
13 South African National Defence Union v Minister of Defence 2012 (4) SA 382 (GNP) paras 18–19.
14 Argus Printing and Publishing Co Ltd v Inkatha Freedom Party 1992 (3) SA 579 (A) at 600.
15 It might not be able to attract the staff it would like to have, but it would be speculative and tenuous to link that to
financial prejudice.
16 2011 (5) SA 329 (SCA). For a critical discussion of this case, see Descheemaeker ‘Three Keys to Defamation: Media
24 in a Comparative Perspective’ (2013) 130(3) SALJ at 435–448.
17 Paras 30 and 54.
18 Para 17.
19 Paras 8 and 12.
20 Para 51.
21 2011 (3) SA 208 (GSJ).
22 Paras 42–51.
23 Paras 47–48 (footnotes omitted).
24 Para 51.
25 Die Spoorbond v South African Railways; Van Heerden v South African Railways 1946 AD 999 at 1008 and 1012
–1013; Mthembi-Mahanyele v Mail & Guardian Ltd 2004 (6) SA 329 (SCA) paras 37–38 and 39.
26 Mthembi-Mahanyele v Mail & Guardian Ltd 2004 (6) SA 329 (SCA) paras 33, 42 and 43.
27 Argus Printing & Publishing Co Ltd v Esselen’s Estate 1994 (2) SA 1 (A) at 28.
28 1998 (3) SA 1114 (ZS).
29 Section 239 of the Constitution of the Republic of South Africa, 1996 provides that an organ of state is (a) any
department of state or administration in the national, provincial or local sphere of government, or (b) any other
functionary or institution (i) exercising a power or performing a function in terms of the Constitution or a provincial
constitution, or (ii) exercising a public power or performing a public function in terms of any legislation.
30 2015 (5) SA 317 (GJ).
31 Para 37.
32 Isparta v Richter 2013 (6) SA 529 (GNP) para 35.
33 Hassen v Post Newspapers (Pty) Ltd 1965 (3) SA 562 (W) at 564–565; Afrika v Metzler 1997 (4) SA 531 (NM) at
535.
34 Moolman v Slovo 1964 (1) SA 760 (W) at 762–763.
35 1960 (3) SA 109 (O).
36 In Masiu v Dos Ramos [2014] JOL 31608A (FB) paras 10–12, the Court held that sending a letter to a third party in
which it is said that one suspects that a person has been engaging in corrupt behaviour is not merely a thought, but
rather an act with real consequences, for which damages can be claimed by the person identified in the letter.
37 Delta Motor Corporation (Pty) Ltd v Van der Merwe [2004] 4 All SA 365 (SCA); 2004 (6) SA 185 (SCA).
38 Tsichlas v Touch Line Media (Pty) Ltd 2004 (2) SA 112 (W).
39 Godfrey v Demon Internet (Britain) [1999] 4 All ER 342; see also Byrne v Deane [1937] 1 KB 818; [1937] 2 All ER
204.
40 Whittington v Bowles 1934 EDL 142.
41 Tothill v Foster 1925 TPD 857.
42 See Heroldt v Wills 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ) paras 10–23, where Willis J quotes
extensively from an academic article by Roos (‘Privacy in the Facebook Era: A South African Legal
Perspective’ (2012) 129(2) SALJ at 375–402), in which she defines and explains social-media and social-networking
sites, and describes the popularity of such online platforms (at 382–385). In footnote 1, Willis J describes Facebook as
follows:

Facebook is a popular free, social-networking website on the internet which

enables registered users to send messages to one another, upload photographs
and videos, keep in touch with one another and send information about
themselves (and others) to other registered users.

43 See Roos ‘Privacy in the Facebook Era: A South African Legal Perspective’ (2012) 129(2) SALJ 375–402 at 382.
44 Roos and Slabbert ‘Defamation on Facebook: Isparta v Richter 2013 6 SA 529 (GP)’ (2014) 17(6) Potchefstroom
Electronic Law Journal 2844–2868 at 2847.
45 These terms are not always interchangeable. As Roos and Slabbert point out at 2848, ‘social media’ refers to any
online platform to which a person may post content. For example, Twitter, which enables users to post 140-character
messages/ statements (‘microblogging’), is an example of social media. YouTube is also an example of social media.
In contrast, LinkedIn is a social-networking site, as it allows users of the site to meet or renew contact and interact
with other users. Facebook is both a social-media and a social-networking site. Thus, social media is a broader
concept that includes social networking.
46 See Singh ‘Social Media and the Actio Iniuriarum in South Africa – An Exploration of New Challenges in the Online
Era’ (2014) 35(3) Obiter at 616–628, where the author commends the court for taking a pragmatic approach to cases
involving defamation on social-networking sites.
47 The courts have also shown willingness to adapt procedural law to take into account the widespread use of social
media. For example, in CMC Woodworking Machinery (Pty) Ltd v Pieter Odendaal Kitchens 2012 (5) SA 604 (KZD)
the Court held that service of legal process could take place via Facebook. See Singh ‘Welcome to Facebook, Pieter
Odendaal: you have been served! (2013) 2:380 Journal of SA Law.
48 2014 (2) SA 569 (GJ).
49 Paras 47–55.
50 2012 (6) SA 201 (GSJ).
51 Paras 32–41, 45 and 49.
52 2013 (6) SA 529 (GNP).
53 Para 35.
54 2015 (1) SA 270 (KZP).
55 Para 20.
56 Para 28.
57 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ).
58 Paras 31–32 and 35.
59 Para 38.
60 Para 39.
61 Para 47.
62 Para 43.
63 At fn 1 and paras 10–23.
64 Para 8.
65 Whittington v Bowles 1934 EDL 142.
66 Williams v Shaw (1884–1885) 4 EDC 105.
67 Whittington v Bowles 1934 EDL 142.
68 1960 (3) SA 109 (O).
69 Sutter v Brown 1926 AD 155 164.
70 Vermaak v Van der Merwe 1981 (3) SA 78 (N) at 83H.
71 Sindani v Van der Merwe 2002 (2) SA 32 (SCA) para 10.
72 Sindani v Van der Merwe 2002 (2) SA 32 (SCA) para 11.
73 Sutter v Brown 1926 AD 155 at 162; Ketler Investments CC t/a Ketler Presentations v Internet Service Providers’
Association 2014 (2) SA 569 (GJ) at 581–583.
74 1980 (1) SA 835 (A) at 842H.
75 HRH King Zwelithini of KwaZulu v Mervis 1978 (2) SA 521 (W).
76 Channing v South African Financial Gazette Ltd 1966 (3) SA 470 (W) at 474.
77 Demmers v Wyllie 1980 (1) SA 835 (A) at 848.
78 Suid-Afrikaanse Uitsaaikorporasie v O’Malley 1977 (3) SA 394 (A) at 408.
79 Geyser v Pont 1968 (4) SA 67 (W) at 69.
80 Chesterton v Gill 1970 (2) SA 242 (T) at 246.
81 Mthembi-Mahanyele v Mail & Guardian Ltd 2004 (6) SA 329 (SCA).
82 [1936] 2 All ER 1237 (HL) at 1240.
83 1996 (1) SA 673 (A).
84 [1936] 2 All ER 1237 (HL).
85 Botha v Marais 1974 (1) SA 44 (A) at 49.
86 Mtyhopo v South African Municipal Workers Union National Provident Fund 2015 (11) BCLR 1393 (CC) para 30; Le
Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (3) SA 274 (CC)
para 91; The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae) 2011 (4) SA 191 (CC); 2011 (8)
BCLR 816 (CC) para 19; Demmers v Wyllie 1980 (1) SA 835 (A) at 842A–C.
87 Mohamed v Jassiem 1996 (1) SA 673 (A) at 706; Sokhulu v New Africa Publications Ltd 2001 (4) SA 1357 (W) para
7.
88 1996 (1) SA 673 (A).
89 At 704.
90 Mohamed v Jassiem 1996 (1) SA 673 (A) at 709, Van der Walt and Midgley Principles of Delict 4 ed (2016) para
104.
91 Pont v Geyser 1968 (2) SA 545 (A) at 558.
92 Gayre v SA Associated Newspapers Ltd 1963 (3) SA 376 (T).
93 Mangope v Asmal 1997 (4) SA 277 (T); Le Roux v Dey (Freedom of Expression Institute and Restorative Justice
Centre as Amicus Curiae) 2011 (6) BCLR 577 (CC) para 107.
94 Prinsloo v SA Associated Newspapers Ltd 1959 (2) SA 693 (W).
95 Sokhulu v New Africa Publications Ltd 2001 (4) SA 1357 (W) at 1359.
96 For greater detail, see Burchell The Law of Defamation in South Africa (1985) at 103–126.
97 Hassen v Post Newspapers (Pty) Ltd 1965 (3) SA 562 (W).
 98 Mthembi-Mahanyele v Mail & Guardian Ltd 2004 (6) SA 329 (SCA).
99 Pitout v Rosenstein 1930 OPD 112; Botha v Mthiyane 2002 (1) SA 289 (W); Du Plessis v Media 24 t/a Daily Sun
2016 (3) SA 178 (GP).
100 For example, an allegation that a person has venereal disease: Tothill v Foster 1925 TPD 857. An allegation that
someone is HIV positive or has Aids would similarly be defamatory.
101 Such as an allegation of insanity: Masters v Central News Agency 1936 CPD 388 at 393.
102 Johnson v Beckett 1992 (1) SA 762 (A).
103 See African National Congress v Democratic Alliance 2014 (3) SA 608 (GJ) at 619; Cele v Avusa Media Ltd [2013] 2
All SA 412 (GSJ) para 23; Malema v Rampedi 2011 (5) SA 631 (GSJ) at 634–635; Argus Printing and Publishing Co
Ltd v Inkatha Freedom Party 1992 (3) SA 579 (A) at 588.
104 2015 (6) SA 389 (FB).
105 Para 33.
106 HRH King Zwelithini of KwaZulu v Mervis 1978 (2) SA 521 (W).
107 2002 (2) SA 32 (SCA).
108 Para 1.
109 Sindani v Van der Merwe 2002 (2) SA 32 (SCA) para 11.
110 Para 12.
111 Para 14.
112 Para 15.
113 2011 (3) SA 274 (CC).
114 Para 99.
115 Para 98.
116 Para 107.
117 Para 109.
118 ‘Heteronormativity’ is defined by the authors as ‘the set of cultural practices and assumptions that privilege
heterosexuality and assume that the monogamous heterosexual couples represent the principle of social union
itself’ (at 408).
119 Barnard-Naude and De Vos ‘The Heteronormative Observer: The Constitutional Court’s Decision in Le Roux v
Dey’ (2011) 128(3) SALJ at 407–419. For additional, different criticism, see Neethling and Potgieter ‘Defamation of
School Teachers by Learners – Le Roux v Dey 2011 (3) SA 274 (CC)’ (2011) 32(3) Obiter at 721–730.
120 For example, National Coalition for Gay and Lesbian Equality v Minister of Justice 1999 (1) SA 6 (CC) para 134;
Minister of Home Affairs v Fourie (Doctors for Life International, Amici Curiae); Lesbian and Gay Equality Project v
Minister of Home Affairs 2006 (1) SA 524 (CC) para 138.
121 Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as Amici Curiae) 2011 (6) BCLR 577
(CC); 2011 (3) SA 274 (CC) para 154.
122 Para 173.
123 Para 167.
124 Para 179.
125 [2013] 2 All SA 412 (GSJ).
126 Paras 31 and 37.
127 Para 28.
128 Paras 23 and 24.
129 Para 31.
130 Para 46. See the discussion of this aspect of the case in the chapter on infringements of dignity in Chapter 27.
131 A Neumann CC v Beauty Without Cruelty International 1986 (4) SA 675 (C) at 679–680.
132 Williams v Van der Merwe 1994 (2) SA 60 (E) at 63–64.
133 SA Associated Newspapers Ltd v Estate Pelser 1975 (4) SA 797 (A) at 810.
134 1992 (3) SA 912 (A).
135 At 919.
136 1986 (4) SA 675 (C).
137 At 679–680.
138 At 683–684.
139 Suid-Afrikaanse Uitsaaikorporasie v O’Malley 1977 (3) SA 394 (A) at 401–403; Khumalo v Holomisa 2002 (5) SA
401 (CC) para 18.
140 2002 (1) SA 289 (W) para 51.
141 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 357; Modiri v Minister of Safety and
Security [2012] 1 All SA 154 (SCA) para 10.
142 Neethling v Du Preez, Neethling v The Weekly Mail 1994 (1) SA 708 (A) at 767–769.
143 Khumalo v Holomisa 2002 (5) SA 401 (CC).
144 [2012] 1 All SA 154 (SCA).
145 Neethling and Potgieter (2015) at 363.
146 1934 AD 43.
147 1965 (3) SA 562 (W).
148 1982 (3) SA 146 (A).
149 1998 (4) SA 1196 (SCA).
150 Neethling and Potgieter at 365; Neethling, Potgieter and Visser Neethling’s Law of Personality 2 ed (2005) at 166 and
167.
151 Burchell ‘Media freedom of expression scores as strict liability receives the red card: National Media Ltd v Bogoshi
(1999) 116(1) SALJ at 1; Burchell Personality Rights and Freedom of Expression: the Modern Actio Injuriarum
(1998) at 315 et seq.
152 2004 (6) SA 329 (SCA).
153 Para 46.
154 Midgley ‘Media liability for defamation’ (1999) 116(2) SALJ at 211; Midgley ‘Intention remains the fault criterion
under the actio injuriarum’ (2001) 118(3) SALJ at 433.
155 Midgley ‘Fault under the actio iniuriarum: Custer’s last stand?’ in Boezaart and De Kock (Eds) Vita perit, labor non
Moritur: Liber Memorialis: PJ Visser (2008) at 187; National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) at 1213
–1214.
156 1965 (3) SA 562 (W).
157 At 576.
158 At 577.
159 1977 (3) SA 394 (A).
160 1998 (4) SA 1196 (SCA) at 1205.
161 At 1213–1214.
162 At 1214.
163 2002 (5) SA 401 (CC) para 20.
164 2007 (5) SA 250 (CC).
165 Para 55.
166 Para 173.
167 2015 (5) SA 317 (GJ).
168 Paras 70, 84 and 103.
169 Para 64.
 Chapter 31

Grounds of justification associated with

infringements of personality interests

31.1 Introduction

31.2 Truth for public benefit

31.2.1 Truth
31.2.2 Public benefit

31.3 Fair comment

31.3.1 Comment or opinion and not a statement of fact
31.3.2 Based upon substantially true facts that are known
31.3.3 Related to a matter of public interest
31.3.4 The comment must be fair

31.4 Privileged occasion

31.4.1 Absolute privilege
31.4.2 Qualified privilege
|
31.5 The defence of reasonable publication

31.1 Introduction
A court’s conclusion on whether conduct is wrongful is an expression of what society considers
acceptable or unacceptable conduct. Therefore, the conclusion is an expression of society’s legal
convictions or boni mores as to whether there could be a claim in delict. Over time, categories of
circumstances have surfaced in which courts have not allowed certain claims because a defendant’s
conduct is justified in the circumstances, provided that certain criteria have been satisfied. Usually,
these grounds indicate the circumstances in which a defendant’s right to freedom of expression
should trump any personality rights that a plaintiff might have in the circumstances. Grounds of
justification are thus defences that a defendant can use to rebut a wrongfulness presumption and
prove that, in the circumstances of the case, there should be no claim.
In actio iniuriarum cases the most common grounds of justification are:

• Truth for public benefit (or public interest)

• Fair comment
• Privileged occasion.

More recently, the defence of reasonable publication (which applies specifically to the media) has
also come to the fore.
The first two grounds (truth for public benefit and fair comment) are ‘truth-based’ defences,
while the latter two (privileged occasion and reasonable publication) are not. So, defendants can
rely on the latter grounds even in circumstances where the material in question is not true.
Defendants have traditionally used these defences for defamation claims, but, where appropriate,
they apply equally to claims for invasion of dignity, privacy or identity.

31.2 Truth for public benefit

The legal convictions of the community dictate that, as a matter of policy, a defendant should be
protected against a claim if the material in question is true and has been made known for the benefit
of the public.1 This defence, which is sometimes called the defence of justification, has two parts to
it: (i) truth, and (ii) public benefit. For the defence to take effect, both these parts must be present.
This means that truth alone cannot be a defence, and so, a statement that is true but not made in the
public interest may lead to liability in delict.2 However, as we describe below, an untrue statement
in the public interest may be justified in some instances, provided that its publication is considered
reasonable, or was made on a privileged occasion.

31.2.1 Truth
The defence of truth for public benefit operates only in circumstances where a defendant makes
factual allegations; a defendant cannot use it where the material in question amounts to comment. A
statement is either true or it is not true, irrespective of what anyone might think or might have
meant to convey. So, the truth of any statement is a matter of fact and is proved just like any other
factual allegation is proved: by producing evidence to that effect. This part of the enquiry is not
concerned with whether the statement is reasonable or fair, and so does not involve any normative
decisions or value judgements.
In line with the presumption of wrongfulness, the law presumes that a statement is false, unless
the defendant proves otherwise by showing that the substance of the material is true. In other
words, a defendant does not have to prove that the material is true in all respects – a person could
exaggerate, for example – but the material allegations must be substantially true. It appears,
however, that where fraud, dishonesty or criminal conduct is alleged as a fact, then every aspect of
such an allegation must be true, not merely its gist.3
Du Plessis v Media 24 t/a Daily Sun 4
A newspaper article alleged that the plaintiff had bound an employee’s wrists with a plastic
strip and locked him in a cold-storage room for two hours after he had seen the employee
stealing an onion. The article also alleged that when he was finally released from the cold-
storage room, the employee’s hair was frozen, his teeth were chattering, and he could barely
speak. There were further allegations that the employee’s nose had bled all night and that he
had required medical treatment. The Court found the article to be exaggerated and
 substantially untrue5 and, accordingly, the defence of truth for the public benefit could not
succeed in the circumstances.6

31.2.2 Public benefit

The second part of the defence requires that the material must have been published for public
benefit, or in the public interest. The primary meaning of these phrases differs: ‘benefit’ means that
there must be some advantage that accrues, while ‘interest’ indicates that the material must be ‘of
interest, or have curiosity value’. 7 However, both phrases are intended to convey the notion of
public concern, in the sense that the material is important and relevant, and that the public is made
aware of the information because the knowledge may be of interest in the public domain. The
underlying reason for such an interest may be that society would benefit in some way, but the
defence is not limited to those instances where a person can show actual benefits.
A defendant, therefore, has to show that the information will be of some significance to
society, and that it is reasonable for society to have that information. Courts will make the decision
based on the nature of the material, as well as the time, manner and place of its publication.
Information about public figures is normally of public interest, but the disclosure of private
morality, even of public figures, would not usually be in the public interest, except to the extent that
this might reflect on their fitness to fulfil their public duties. In Heroldt v Wills 8 the Court
considered the defence in a case involving a defamatory statement that had been posted to
Facebook and made the following remarks:
The ‘truth plus public benefit/interest’ test will generally protect both public figures and
those who write about them provided it is remembered that it is not in the public interest
that every titbit of information and not every morsel of salacious gossip about a public
figure be made publicly known. There is legitimate public interest in the affairs of public
figures. Legitimate interest in what they do does not overshadow the fact that public figures
have the same human rights as everyone else. They too enjoy a constitutional right to
privacy. Our law protects every person’s right, not only to dignitas (inner tranquillity) but
also to fama (reputation).9

Previous conduct should also be forgiven, as a number of old cases illustrate. For example, in Lyon
v Steyn 10 the Court found that a reference to a true incident that happened 30 years previously 11
was not one that the public needed to know about. The law of defamation protects people against
the raking-up of their past.

COUNTER Spite and malice

POINT Should a court take into account the fact that a defendant was prompted
by spite or malice when it assesses whether material is true or for the
public benefit? Malice or improper motive defeats the defences of fair
comment12 and privileged occasion,13 but apparently not the defence of
truth and public benefit.14 Burchell states:15

The truth is the truth no matter what the motives of the publisher
are and the publication of truth for the public benefit does not
cease to be for the public benefit simply because the publisher
is prompted by some improper or ulterior motive.

However, if malice can indicate that a privileged occasion was abused,

why could it also not indicate that the defence of truth and public benefit
was abused? After all, a ground of justification should not be seen in
isolation from the overall wrongfulness criterion. It should embody
 society’s belief that the disclosure in the circumstances is a reasonable
one. Would society not consider a fact that is true and for the public
benefit, but published maliciously, to be unreasonable? Is this not akin to
an abuse of rights?

Johnson v Rand Daily Mail16

A newspaper published a letter to the editor that criticised the catering arrangements at the
annual agricultural show. Johnson was in charge of the catering arrangements. The criticisms
included that the caterer was incompetent, the food was disgraceful because the eating areas
were filthy, the table linen was greasy and dirty, and that the floor was littered with refuse that
had fallen or had been swept from the tables. The newspaper pleaded truth for the public
benefit.
The Court had to decide whether these allegations were substantially correct, or whether
they were so greatly exaggerated ‘as to leave a wrong impression on the mind of the reader’.
This does not mean that the defendant has to prove every allegation as described. The
evidence indicated that the conditions in the tea room were filthy, but the Court thought that
‘to describe them as indescribably filthy might appear somewhat overstated’. However, the
sting of the defamation was ‘that the tea rooms and the luncheon room were in a state of
uncleanliness, that scraps of meat and vegetables were lying about the floor, and that the
tablecloths were unduly stained with grease and other stains’.17 The Court concluded:

The fact that there is some exaggeration in the language used does not deprive a plea
of justification of its effect. The test is whether the exaggeration leaves a wrong
impression on the reader’s mind to the detriment of the plaintiff … . Even, therefore, if
there was some exaggeration in the use of words such as ‘indescribable’, or in saying
that the tablecloth had to be turned back on account of the grease, then yet the
justification is proved, for only ‘as much must be justified as meets the sting of the
charge and if anything be contained in the charge which does not add to the sting of it,
that need not be justified’.

Modiri v Minister of Safety and Security18

The Daily Sun published an article in which it alleged that the appellant was suspected of
being involved in drug dealing, cash-in-transit heists and car theft. The article quoted a police
superintendent as saying that the appellant used other people to do his ‘dirty work’ for him
and calling on members of the public to help the police apprehend the appellant. The
appellant sued the Minister and the newspaper for defamation, but on appeal the appellant
accepted that the police superintendent had not made the defamatory statement attributed to
him by the article (he had been misquoted) and therefore he had no claim against the police
respondents.19
Regarding the claim against the media respondents, the Court held that the defence of
truth for the public benefit should succeed, as the gist or sting of the article was correct – the
evidence of the police respondents indicated that the appellant was suspected of having
been involved, over a number of years, in criminal activities.20 The fact that the article was
incorrect or inaccurate in some minor respects did not mean it was untrue for the purposes of
the defence of truth for the public benefit.21 The Court also found that publishing the
defamatory matter about the appellant was in the public interest, as the suspicions about the
appellant were strong and publishing the defamatory matter would assist the police in
procuring witnesses and evidence that could be used to prosecute a person suspected of
having engaged in serious criminal activities.22 Accordingly, the newspaper was able to
escape liability on the basis of the defence of truth for the public benefit.23
The fact that the media respondents had not led any evidence regarding the basis for the
defence did not preclude them from using it. As a wrongfulness-based defence, truth for the
public benefit was established on the basis of objective facts, which had been provided by
 evidence given on behalf of the police respondents.24 In addition, the Court rejected the
argument that the media respondents could not rely on the information of the appellant’s
alleged criminal activities testified to by the police respondent’s witnesses because it had not
been demonstrated that the article was based on that information. As far as the Court was
concerned, it did not matter where the information had come from, as long as the sting of the
article was in fact true.25

31.3 Fair comment

Freedom of expression is a prized right in a constitutional democracy, and so public policy requires
the law to give a considerable amount of freedom to citizens expressing opinions. Courts should
protect honest and fair criticism, as well as opinions and comments that are fair. The defence of fair
comment, therefore, protects:
the right of the citizen honestly to express his genuine opinion on a matter of public interest,
however wrong, exaggerated or prejudiced that opinion may be.26

The basis of this defence is that a comment or an opinion exists, unlike truth and public benefit,
which is based upon allegations of fact. Comments and opinions express value judgements, which
by their very nature cannot be true or false, so the protection extends to instances where views and
opinions are honestly held and fairly made. The criteria for assessing what constitutes fair comment
are:
• The material must amount to comment or opinion, and not a statement of fact.
• The facts upon which the comment or opinion is based are substantially true, and either widely
known or incorporated by reference.
• The comment or opinion pertains to a matter of public interest.
• The comment or opinion is fair.

31.3.1 Comment or opinion and not a statement of fact

A person making a comment is expressing an opinion, not a fact. The difference is not always easy
to determine for, as Burchell points out, sometimes an opinion can masquerade as a fact.27 In
Crawford v Albu28 Crawford and eight others were deported to England after being arrested in
connection with labour disturbances. Albu believed the government had acted correctly in
deporting them, and at a meeting shortly after their deportation said:
All this strife has been caused by men who are fanatics – no they are not fanatics – they are
criminals in the fullest sense of the word.

It may seem obvious that this statement is a comment in the form of a criticism based on certain
facts, but one eminent judge of appeal, Innes CJ, considered this to be a statement of fact.
The test for determining whether a statement is an expression of fact or opinion is the opinion
of the reasonable reader, listener or viewer.29 The use of phrases such as ‘it seems to me’ or ‘in my
view’ help to establish the distinction between fact and comment, although not conclusively.
Sometimes the context in which the words are used will also help. However, if a court cannot
determine, on the face of the statement, whether it is a comment or a statement of fact, it will
presume the statement is factual.

31.3.2 Based upon substantially true facts that are known

The law requires that the comment must be based on facts, and that those facts must be known to
the reasonable reader, listener or viewer, either because they were stated expressly when making
the comment, or because they were so well known that they constituted common knowledge. In the
latter instance, the facts are implied.
Fair comment is a ‘truth-based’ defence, and so, while the comment cannot be true or false, the
facts upon which the comment is based must be true. Obviously, an opinion based upon false or
distorted facts cannot be a reasonably held opinion. As with the defence of truth for the public
benefit, every detail of the facts does not have to be true.

31.3.3 Related to a matter of public interest

In this context, ‘public interest’ has the same meaning that ‘public benefit’ has in the defence of
truth for the public benefit. If a matter may affect other people, so that they may be legitimately
interested in it or concerned about how it might affect them or others, it is a matter of public
interest.

31.3.4 The comment must be fair

To be a fair comment, the comment must fall within prescribed limits. The issue is not whether the
comment is valid, impartial or balanced. The criterion for setting these limits is whether the
comment is a genuine expression of opinion, relevant, honest, and free from malice (not prompted
by an improper motive).30 The comment could be extravagant or exaggerated, and possibly even
prejudiced, provided the nature of the prejudice does not offend constitutional values. This also
applies to comments on social media: in Heroldt v Wills 31 the Court linked the defence to the right
to freedom of expression 32 and said: 33
The ‘fair comment’ test will generally come to the aid of those who wish to express
themselves lavishly and perhaps even extravagantly. Trenchant commentaries on the
performances of politicians as politicians, entertainers as entertainers, musicians as
musicians, artists as artists, writers as writers, poets as poets, sports stars as sports stars will
generally pass legal muster, even if posted in the social media. When it comes to freedom of
expression in South Africa, there are oceans in which to swim and upon which to sail as
freely as the wind blows.

It stands to reason that proof of malice would defeat the defence of fair comment.
Delta Motor Corporation (Pty) Ltd v Van der Merwe34
Van der Merwe owned a four-wheel-drive vehicle that Delta Motor Corporation had
manufactured. The vehicle developed a bent chassis, which Van der Merwe believed was the
result of a manufacturer’s defect. He attempted, in vain, to have the vehicle replaced or
repaired at Delta’s expense. Delta maintained that the bent chassis was due to overloading,
bad driving and owner abuse. The result was that Van der Merwe sent email messages to 27
people that contained photographs of the vehicle with the words (translated): ‘Worst 4×4×far’.
One of the issues was whether this statement constituted fair comment.
The criticism was based on the fact that the vehicle’s chassis had bent on a gravel road
and whatever had caused the chassis to bend must have occurred during the trip on the
gravel road. The condition of the road would not have damaged the chassis of an ordinary
vehicle.
The comment was a parody of a well-known advertisement of another product, which
called itself ‘The best 4×4×far’. Van der Merwe’s adaptation was, of course, an exaggeration,
but this did not make the comment malicious, nor did this change its nature to something
other than a genuine expression of opinion that the vehicle was defective. Any inference that
other vehicles of the same make might have similar defects could not be regarded as unfair.
PAUSE FOR Truth and fair comment
REFLECTION In Crawford v Albu 35 the Court summarised the basic difference between
‘truth in the public interest’ and ‘fair comment’ as:

Under the former the defendant must justify every injurious fact
and imputation in whatever shape expressed. Under the latter
he must justify the facts, but he need not justify the comment; it
is sufficient if he satisfies the Court that it is ‘fair’.36
unpersuasive. Untrammelled debate enhances truth-finding and
enables us to scrutinise political argument and deliberate social
value.38

The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others; Amici Curiae)37
McBride sued The Citizen newspaper, its editor and journalists for defamation. The Citizen
had published a series of articles and editorials questioning McBride’s suitability for the post
of Ekurhuleni Metro Police Chief. It expressed the view that McBride was unsuitable for the
post because (i) he was a ‘criminal’ and a ‘murderer’, (ii) he was not contrite for the deaths he
caused, and (iii) he had dubious dealings with alleged gun dealers in Mozambique in 1998.
The articles referred to McBride’s conviction in the apartheid era for the planting of a bomb
that had killed a number of people in a bar. McBride alleged that these articles neglected to
add a crucial fact: McBride had been granted amnesty in terms of the Promotion of National
Unity and Reconciliation Act 34 of 1995, and that in terms of section 20(10) of the Act his
conviction was therefore deemed ‘for all purposes’ not to have taken place. He contended
that the labels ‘criminal’ and ‘murderer’ therefore did not apply to him. The Supreme Court of
Appeal upheld McBride’s claim and found that the false statements could not sustain the
defence of fair comment.
A majority in the Constitutional Court found that the Reconciliation Act did not render the
fact that McBride committed murder untrue. The Citizen’s comments were based on an
adequate exposition of the relevant facts and constituted comment on McBride’s suitability
for an important and prominent public post.
The majority emphasised that protected comment need not be ‘fair or just at all’, in the
sense in which these terms are commonly understood. Criticism is protected even if extreme,
unjust, unbalanced, exaggerated and prejudiced, so long as it expresses an honestly held
opinion, without malice, on a matter of public interest on facts that are true. The Citizen was
thus entitled to express views on McBride’s suitability for the post, and even though the Court
considered The Citizen’s coverage to be vengeful and distasteful, it was nonetheless entitled
to legal protection.
The judgments in this case raise fundamental issues. Cameron J, who wrote the majority
judgment, remarked:

So to dub the defence ‘fair comment’ is misleading. If, to be protected, comment has to
be ‘fair’, the law would require expressions of opinion on matters of fact to be just,
equitable, reasonable, level-headed and balanced. That is not so. An important
rationale for the defence of protected or ‘fair’ comment is to ensure that divergent
views are aired in public and subjected to scrutiny and debate. Through open contest,
these views may be challenged in argument. By contrast, if views we consider wrong-
headed and unacceptable are repressed, they may never be exposed as
unpersuasive. Untrammelled debate enhances truth-finding and enables us to
scrutinise political argument and deliberate social value.38

Perhaps it would be clearer, and helpful in the understanding of the law, if the defence
were known rather as ‘protected comment’. A new name would not change the
 requirements. At common law it was rightly held that ‘fairness’ in fair comment must
draw on the general legal criterion of reasonableness. In our constitutional state,
comment on matters of public interest receives protection under the guarantee of
freedom of expression. Hence the values and norms of the Constitution determine the
boundaries of what is protected. To call the defence ‘protected comment’ may
illuminate the constitutional source and extent of the protection.39

Ngcobo CJ did not share this view, on the basis that the requirement of fair comment
maintains a delicate balance between the need to protect the right of everyone, including the
press, to freedom of expression and the need to respect human dignity:

The requirement that a comment must be fair is consistent with the values that
underlie our constitutional democracy. It underscores the need to balance freedom of
expression, on the one hand, and the need to protect human dignity, on the other. By
insisting that a comment must be fair, the common law demands that comment be fair
having regard to the right to human dignity. The comment must be relevant to the
matter commented upon and it must not be actuated by malice. It underscores the
proposition that freedom of expression does not enjoy a superior status to other rights
enshrined in the Constitution. Indeed, it gives effect to the constitutional commitment
this Court articulated in Mamabolo40 to ‘three conjoined, reciprocal and covalent values’
that are foundational to our Republic, namely, human dignity, equality and freedom.41

In my view, the requirement of fair comment is consistent with the need to respect and
protect dignity. It maintains a delicate balance between the need to protect the right of
everyone, including the press, to freedom of expression and the need to respect
human dignity. This is the balance that the Constitution requires be struck. I do not,
therefore, share the view expressed by Cameron J that the word ‘fair’ is misleading. It
must now be understood in the light of our Constitution, in particular the foundational
values of human dignity and freedom upon which our constitutional democracy rests
and the need to strike a balance between ensuring that freedom of expression is not
stifled and insisting on the need to respect and protect human dignity.42

Cameron J seems to suggest a less exacting standard for establishing the defence (protected
comment) and Ngcobo CJ a more demanding standard (fair comment in the context of a
constitutional commitment to freedom of expression and the value of human dignity).

Democratic Alliance v African National Congress43

The Democratic Alliance sent a text message to more than 1,5 million voters in Gauteng
during the run-up to the 2014 general elections. The text message consisted of the following
statement:

The Nkandla report shows how Zuma stole your money to build his R246m home.
Vote DA on 7 May to beat corruption. Together for change.

The public protector had the day before released a report on security upgrades at President
Jacob Zuma’s private home in Nkandla. The African National Congress claimed that the
content of the text message was a false statement that was in violation of section 89(2) of the
Electoral Act 73 of 1998 and item 9(1)(b) of the Electoral Code, which make it an offence to
disseminate false information with a view to influencing the conduct or outcome of an
election.44 It sought an interdict compelling the Democratic Alliance to retract the statement by
sending a text message to this effect to all the persons to whom the original text message
had been sent, and to apologise for the message. Relying on its right to freedom of
expression,45 the Democratic Alliance claimed the text message was an opinion that
constituted a fair comment in the circumstances.46
While this case was not a delictual claim for damages, but rather one concerning statutory
interpretation,47 the Court considered whether the text message was a factual statement or
 an opinion, so as to determine whether it fell within the ambit of the legislative proscription.
The judgment accordingly draws on and elucidates certain key concepts and defences in the
law of delict.48 The majority of the Court emphasised that penal provisions in statutes should
be restrictively interpreted, so as not to unreasonably interfere with the liberty of citizens 49
but, crucially, the Court held that the statutory provisions in question criminalised only
statements of fact, not opinions.50 The text message was an opinion, as it referred to the
public protector’s report as the factual basis for its claim,51 and opinions could generally not
be said to be false.52 In any event, as the text message was an opinion, it did not fall within
the ambit of the legislative provisions in question,53 so it was not necessary to determine
whether the text message was true or false.54

31.4 Privileged occasion

Unlike truth for public benefit and fair comment, the defence of privileged occasion may protect
both true and untrue statements, and it does not matter whether they are statements of fact or
opinion. The emphasis is on circumstances in which the law recognises that the free flow of
information is more important than a person’s reputation.55 The public interest demands that courts
should not impede freedom of expression, even where that expression consists of defamatory or
untrue statements.
Sometimes this defence is called a ‘privilege’, but it is important to note that it is not the
statement or its content that is privileged, but the circumstances in which the statement is made. For
this reason, it is better to speak of a ‘privileged occasion’ than of ‘privilege’. This also highlights
the need for a statement to be relevant to such an occasion before a person can claim the defence.
Courts distinguish between occasions of absolute privilege – which effectively create
situations of immunity in which everything, no matter what, is protected – and those where the
privilege is qualified, and where protection is extended only if certain criteria are met.

31.4.1 Absolute privilege

In the interests of democracy, free speech and full and effective deliberation,56 statements made
while participating in parliamentary proceedings and those of provincial legislatures are accorded
absolute protection against actions under the actio iniuriarum.57 Provincial legislatures may grant
similar immunity in respect of municipal councils and their members.58 The protection adheres,
irrespective of the truth of or the motive behind making the statement, provided that:
• The person making the statement is a member of the legislature or council.
• It forms part of the business of the legislature and its committees, and the way in which issues
are communicated in conducting its business.

Non-members cannot claim protection, nor can members who have acted outside legislative or
council business.59 However, it has been suggested that, unless a statute specifically excludes
protection,60 the law should protect outsiders who have been called to testify before a committee
(the rationale for protection applies equally to them), and that the law should extend the protection
to all legislative and council business, whether conducted within or outside formal proceedings.61

31.4.2 Qualified privilege

A defendant may rebut the presumption of wrongfulness by showing that the communication is
protected because it was made for moral, social or legal reasons – in other words, in circumstances
public policy dictates should not attract liability.62 So, the defendant’s aim is to negative the
wrongfulness element of the delict. This is unlike absolute privilege, which protects even those
communications that meet all the elements of a delict, including the element of wrongfulness. As
with absolute privilege, the rationale for the defence is to protect freedom of expression. However,
because the public policy factors are not as strong, malicious communication is not protected.
The core issue, as with all wrongfulness questions, is whether the criterion of reasonableness
has been met, and to do so the communication must:

• Be made on an occasion that the law recognises as privileged:

◆ Statements made because there is a moral, social or legal duty to do so, or to further a
legitimate interest or in the discharge of a duty. The key issue here is reciprocity. A person
must not only further an interest or discharge a duty, but the recipient of the information
must have an interest in receiving the information. For example, in Vincent v Long 63
Vincent, a veterinary surgeon, sued for defamation arising from a letter that Long had
written to the South African Veterinary Council. In the letter, he alleged that his mother’s
cat had been treated incorrectly and that several people had lost their pets through
Vincent’s negligent surgery. At issue was whether Long had a legal, moral or social duty to
make the statement to the Council. The cat did not belong to him, he had no mandate to
lodge the complaint on his mother’s behalf or on behalf of the other aggrieved parties, and
he was not responsible for paying the veterinarian’s bill. The Court held that an interest in
bringing complaints that concern professional people to the attention of their disciplinary
body is not restricted to a person who has been directly prejudiced. Similarly, in O v O 64
the defendant made a defamatory statement to a priest over the dinner table. At issue was
whether the occasion was privileged. The Court found that a minister, priest, rabbi or
moulana performs an important function in our society, and that it is in the public interest to
uphold the right to speak frankly to one’s advisers, even at risk to others’ reputations.65 In
Byrne v Masters Squash Promotions CC 66 it was held that an employer has a legal, moral
and social duty to inform an employee of the reasons for his dismissal. Accordingly, the
publication of defamatory facts to the typist of a letter to be sent to the employee informing
him of the reasons for his dismissal was privileged.67
• Statements made during judicial or quasi-judicial proceedings. Judicial or quasi-judicial
proceedings constitute a privileged occasion for judges, magistrates, advocates, attorneys,
litigants and witnesses.68 The ambit of protection is far wider for judicial officers than for the
other participants, since public interest in the due administration of justice requires that they be
given free rein to speak their minds in the exercise of their judicial functions without incurring
liability for damages.
• Statements made in reports on court and parliamentary proceedings and those of public bodies.
In a democratic society, the public needs to know what is happening in its governmental
institutions. So, fair and substantially accurate reports of judicial or parliamentary proceedings
(which are protected) that contain defamatory material, for example, will also be protected.69
• Be relevant or germane to the occasion:70 There is no set rule for proving the relevance of a
statement. In Van der Berg v Coopers and Lybrand Trust (Pty) Ltd 71 the Court held that the
assessment amounts to a value judgement, based on reason and common sense, as to what
would be reasonably necessary to protect the interest or to discharge the duty upon which the
privilege is founded. However, the truthfulness or otherwise of statements has no bearing on
whether statements are relevant.72
• Not be malicious:73 Malice or improper motive will defeat any claim to the defence of
privileged occasion, because society believes that there can be no legal, moral, or social duty
to publish matter for malicious reasons. For liability to arise there would have to be some
abuse of the occasion – for example, where a witness has no reasonable grounds for believing
the truth of a statement.74 (When an untrue statement is made, one can infer that it was made
with malice, unless the circumstances indicate otherwise.)75 The general rule is that when an
 infringement of a personality right is proved, a presumption of wrongfulness arises and the
onus shifts to the defendant to disprove wrongfulness. A bald denial is not enough to rebut the
inference.76 The defendant usually does this by establishing the first two criteria: that the
occasion was privileged, and that the statement was relevant to the occasion. The onus then
shifts back to the plaintiff to show that the occasion was abused, by proving malice. However,
the situation with respect to presiding officers (judges and magistrates) is different. Presiding
officers are presumed to have acted lawfully when they make defamatory statements while
performing their duties,77 which means that, contrary to the general rule, proving the
infringement of a personality right does not raise presumption of wrongfulness. (The normal
rules apply with respect to other participants in legal proceedings.)

PAUSE FOR Media privilege

REFLECTION The media do not occupy a special position in society and there is no
such thing as media privilege in our law.78 If a newspaper were to rely on
the defence of privileged occasion, it would have to comply with the usual
criteria. While there is scope for arguing that the press has a right, and a
duty, to inform the public on matters of public interest, and that the public
has a corresponding right to receive that information,79 given our courts’
reluctance to grant special dispensation to the media, a better avenue
would be to pursue the defence of reasonable publication.80

De Waal v Ziervogel81
Ziervogel was a minister in the Dutch Reformed Church. He sued De Waal, a school
headmaster, for damages arising from an allegation that Ziervogel had an affair with De
Waal’s wife. The wife had confessed to being unfaithful to her husband and to having had an
intimate relationship with Ziervogel. De Waal published the information to various persons,
among them a magistrate, who was a member of the church but not in the same
congregation as the plaintiff and defendant. He also sent the information to four others, three
of whom were elders of the church, and one a former elder.
The Court quoted with approval 82 an earlier Appellate Division statement in Ehmke v
Grunewald:83

Where a person publishing the defamatory matter is under a legal, moral or social duty
to do so or has a legitimate interest in so doing and the person to whom it is published
has a similar duty or interest to receive it then the occasion of the publication would be
privileged.

When deciding whether the occasion was privileged, the Court had to determine from all the
circumstances whether a moral or social duty or interest existed that entitled a person in the
position of the defendant to make the communications that he or she did, and whether a
corresponding interest existed in respect of the persons to whom the communications were
made to receive the information. The Court stressed that it is the occasion on which the
statement is made that is privileged, not the subject matter, although the subject matter
should be relevant to the occasion.84
A court must decide whether the occasion is privileged in terms of the general rule that
relates to duty and interest, from the circumstances of the case, independently of the motives
that prompted the defendant to publish the defamatory matter:

In other words, the question which the Court has to decide at this stage is not was the
defendant in fact speaking from a sense of duty but did the circumstances in the eyes
of a reasonable man create a duty or an interest which entitles the defendant to speak.
 This does not mean that the state of mind or actuating motive of the defendant is
immaterial in the ultimate result of the case because it becomes very relevant in the
next stage of the enquiry when the question arises whether a privileged occasion has
been abused.85

The Court concluded that the recipients of the communications were not all on the same
footing. Three of them were elders, one of them was a former elder, and the other a
magistrate with no official position in the church or in the congregation. The communication to
the elders was privileged because they had an interest in receiving the information, but there
was no such interest in respect of the former elder, who was in no better position than an
ordinary member of the congregation, and the magistrate.

31.5 The defence of reasonable publication

The defence of reasonable publication, confirmed in National Media Ltd v Bogoshi,86 was expressly
stated for the first time in Khumalo v Holomisa: 87
(The defence of reasonable publication) permits a publisher who can establish truth in the
public benefit to do so and avoid liability. But if a publisher cannot establish the truth, or
finds it disproportionately expensive or difficult to do so, the publisher may show that in all
the circumstances the publication was reasonable.

This defence is particularly important in the context of false defamatory statements published in the
press,88 although it could also apply in other areas.
National Media Ltd v Bogoshi89
Bogoshi had sued for defamation arising from the publication of a series of articles. National
Media Ltd alleged that the publication was lawful and therefore objectively reasonable based
on its right to freedom of expression. The Supreme Court of Appeal reaffirmed that public
policy grounds paved the way for recognising new situations that could render a defendant’s
conduct lawful. The criterion to apply in each case is:

the general criterion of reasonableness based on considerations of fairness, morality,

policy and the Court’s perception of the legal convictions of the community.90

The Supreme Court of Appeal specifically addressed the conditions under which they would
consider the publication of false defamatory matter lawful.91 In trying to determine whether the
media deserve some form of protection for publishing false information, the Court seemed to
refer to principles similar to those for the defence of privileged occasion. It recognised that
within this category it may sometimes be reasonable to publish particular untrue facts in a
particular way and at a particular time, depending on all the circumstances of the case.92
Courts should consider the following factors, subsequently confirmed in Mthembi-Mahanyele
v Mail & Guardian Ltd,93 when assessing whether any mistake or ignorance is objectively
reasonable and amounts to a ‘justifiable publication’:94
• The interest in the public being informed
• The manner of publication
• The tone of the material published
• The extent of the public concern in the information
• The reliability of the source
• The steps taken to verify the truth of the information (this factor would also play an important
role in considering whether there was negligence on the part of the press, assuming that the
publication was found to be defamatory)
 • Whether the person defamed was given the opportunity to comment on the statement before
publication. In cases where information is crucial to the public, and is urgent, it may be
justifiable to publish without providing an opportunity to comment.

Notwithstanding this innovation in the law, the Supreme Court of Appeal cautioned that this
did not mean that journalistic standards could be lowered and held that:

members of the press should not be left with the impression that they have licence to
lower the standards of care which must be observed before defamatory matter is
published in a newspaper.95

Courts still expect a high degree of caution from the media, particularly because of the
important role that it plays in a properly functioning democratic society.96 The Court, therefore,
did not reduce or limit the test for reasonableness; it simply extended the circumstances in
which the test can be applied.

According to Van der Walt and Midgley the Bogoshi judgment establishes the following rule:97

Publication in the press of false defamatory material in which the public has an interest
will not be unlawful if, upon a consideration of all the circumstances of the case, it is
found to have been reasonable to publish the particular facts in the particular way and
at the particular time.

They contend that the enquiry is located firmly in the concept of objective reasonableness,
which in turn is based upon society’s legal convictions, and that public interest will play an
important role in determining the contours of the defence.

In summary, prior to the Bogoshi decision ignorance of the law and mistake did not avail a
defendant of a justification ground to rebut unlawfulness. The Bogoshi decision paved the way for a
court to determine whether society would regard a defendant’s mistake or ignorance as objectively
reasonable or justifiable.

1 Heroldt v Wills 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ) para 27; Ketler Investments CC t/a Ketler
Presentations v Internet Service Providers’ Association 2014 (2) SA 569 (GJ) paras 56–83.
2 For a general historical discussion of the defence of truth for the public benefit, see Descheemaeker ‘A Man of Bad
Character Has Not So Much to Lose: Truth as a Defence in the South African Law of Defamation’ (2011) 128(3)
SALJ at 452–478.
3 Johnson v Rand Daily Mails 1928 AD 190 at 205.
4 2016 (3) SA 178 (GP).
5 Para 17.
6 Para 29.
7 National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA).
8 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ).
9 Para 45 (footnotes omitted).
10 1931 TPD 247.
11 One person had accused another of being a Boer verraaier, or traitor to the British, during the Anglo-Boer War.
12 Marais v Richard 1981 (1) SA 1157 (A).
13 De Waal v Ziervogel 1938 AD 112 at 122–124; Mohamed v Jassiem 1996 (1) SA 673 (A) at 711.
14 Kleinhans v Usmar 1929 AD 121 at 126.
15 Burchell Personality Rights and Freedom of Expression: The Modern Actio Injuriarum (1998) at 276.
16 1928 AD 190.
17 At 205–207.
18 [2012] 1 All SA 154 (SCA).
19 Para 8.
20 Paras 13–18.
21 Paras 17 and 18.
22 Para 26.
23 Para 26.
24 Para 12.
25 Para 19.
26 Telnikoff v Matusevitch [1991] 4 All ER 817 at 826.
27 Moyse v Mujuru 1999 (3) SA 39 (ZS) at 47–48.
28 1917 AD 102 at 107.
29 Marais v Richard 1981 (1) SA 1157 (A) at 1168G–H.
30 Crawford v Albu 1917 AD 102 at 115; African National Congress v Democratic Alliance 2014 (3) SA 608 (GJ) at 620
–621.
31 2013 (2) SA 530 (GSJ); [2013] 2 All SA 218 (GSJ).
32 Section 16 of the Constitution of the Republic of South Africa, 1996.
33 Para 45.
34 [2004] 4 All SA 365 (SCA); 2004 (6) SA 185 (SCA).
35 1917 AD 102.
36 1917 AD 102 at 117.
37 2011 (4) SA 191 (CC); 2011 (8) BCLR 816 (CC).
38 The Citizen 1978 (Pty) Ltd v McBride (Johnstone and Others, Amici Curiae) 2011 (4) SA 191 (CC); 2011 (8) BCLR
816 (CC) para 82.
39 Para 84.
40 S v Mamabolo (E TV and Others intervening) 2001 (3) SA 409 (CC).
41 Para 157H.
42 Para 158.
43 2015 (2) SA 232 (CC).
44 Paras 2, 15 and 17.
45 Para 32.
46 Paras 3, 16 and 17.
47 Para 119.
48 Para 30.
49 Paras 129–130.
50 Para 144.
51 Paras 150–153.
52 Paras 145–146.
53 Para 153.
54 Para 167.
55 Burchell The Law of Defamation in South Africa (1985) at 238.
56 Dikoko v Mokhatla 2006 (6) SA 235 (CC) para 39.
57 Sections 58(1) and 117(1) of the Constitution of the Republic of South Africa, 1996; Poovalingam v Rajbansi 1992
(1) SA 283 (A) at 293.
58 Section 161 of the Constitution and section 28 of the Local Government: Municipal Structures Act 117 of 1998;
Dikoko v Mokhatla 2006 (6) SA 235 (CC).
59 Dikoko v Mokhatla 2006 (6) SA 235 (CC) para 32.
60 See, for example, the North West Provincial Legislature’s Powers, Privileges and Immunities Act 5 of 1994.
61 Dikoko v Mokhatla 2006 (6) SA 235 (CC) paras 35, 39 and 41.
62 Ketler Investments CC t/a Ketler Presentations v Internet Service Providers’ Association 2014 (2) SA 569 (GJ) para
84.
63 1988 (3) SA 45 (C).
64 1995 (4) SA 482 (W).
65 At 492.
66 2010 (1) SA 124 (GSJ).
67 Paras 11–12.
68 May v Udwin 1981 (1) SA 1 (A); Joubert v Venter 1985 (1) SA 654 (A).
69 Benson v Robinson & Co (Pty) Ltd 1967 (1) SA 420 (A) at 428.
70 Mohamed v Jassiem 1996 (1) SA 673 (A) at 710; Van der Berg v Coopers and Lybrand Trust (Pty) Ltd 2001 (2) SA
242 (SCA) para 17.
71 2001 (2) SA 242 (SCA) para 26. See also National Education, Health and Allied Workers Union v Tsatsi [2006] 1 All
SA 583 (SCA) para 12.
72 Naylor v Jansen; Jansen v Naylor 2006 (3) SA 546 (SCA) para 11. See also National Education, Health and Allied
Workers Union v Tsatsi [2006] 1 All SA 583 (SCA) para 12.
73 De Waal v Ziervogel 1938 AD 112 at 120–123; Mohamed v Jassiem 1996 (1) SA 673 (A) at 710.
74 Borgin v de Villiers 1980 (3) SA 556 (A) at 578–579; Joubert v Venter 1985 (1) SA 654 (A) at 702.
75 Naylor v Jansen; Jansen v Naylor 2006 (3) SA 546 (SCA) para 11, with reference to Borgin v de Villiers 1980 (3) SA
556 (A) at 578H.
76 Naylor v Jansen; Jansen v Naylor 2006 (3) SA 546 (SCA) para 11.
77 May v Udwin 1981 (1) SA 1 (A) at 19.
78 Neethling v Du Preez; Neethling v The Weekly Mail 1994 (1) SA 708 (A) at 776 and 777.
79 Zillie v Johnson 1984 (2) SA 186 (W).
80 A defence that succeeded in National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) at 1212.
81 1938 AD 112.
82 At 121.
83 1921 AD 575 at 581.
84 At 122.
85 At 123.
86 1998 (4) SA 1196 (SCA).
87 2002 (5) SA 401 (CC) para 43.
88 Malema v Rampedi 2011 (5) SA 631 (GSJ) at 636.
89 1998 (4) SA 1196 (SCA).
90 At 1204D–E.
91 At 1204G–I.
92 At 1212G–H.
93 2004 (6) SA 329 (SCA) para 68.
94 For a more detailed account of the factors, refer to the Bogoshi judgment National Media Ltd v Bogoshi 1998 (4) SA
1196 (SCA) at 1211–1213; Gold Reef City Theme Park (Pty) Ltd v Electronic Media Network Ltd; Akani Egoli (Pty)
Ltd v Electronic Media Network Ltd 2011 (3) SA 208 (GSJ) paras 80–83.
95 National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) at 1212J–1213A.
96 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 133.
97 Van der Walt and Midgley (2016) para 133.
 Chapter 32

Strict liability

32.1 Introduction

32.2 Why strict liability?

32.3 Liability for harm caused by animals

32.3.1 The actio de pauperie for harm caused by domestic and domesticated
animals
32.3.1.1 Type of animal
32.3.1.2 Person who could be liable
32.3.1.3 Type of behaviour – the contra naturam test
32.3.1.4 Elements| of the action
32.3.1.5 Defences

32.4 The actio de pastu for harm caused by grazing animals

32.5 The actio de effusis vel deiectis and the actio positis vel suspensis

32.6 Statutory instances of strict liability

32.6.1 Aviation
32.6.2 Damage to telecommunication lines and call boxes
32.6.3 Genetically modified organisms

32.1 Introduction
In this chapter, we discuss the common-law actions that deal with harm caused by animals, and
those that deal with the liability of occupiers of buildings, as well as some statutory instances of
strict liability. We deal with vicarious liability separately in the following chapter.
32.2 Why strict liability?
The law of delict is primarily concerned with fault-based liability. Actions involving liability
without fault include:
• The actio de pauperie for harm caused by domestic animals
• The actio de pastu for harm caused by grazing animals
• Th actio de effusis vel deiectis and the actio positi vel suspensi for recovering damages from
the occupier of a building from which something was thrown (effusis vel deiectis) or from
which something fell down (positi vel suspensi).

Vicarious liability of employers, which is dealt with in the following chapter, could also be seen as
a form of strict liability. This is where the employer is held liable without fault for the wrongdoing
of an employee, and the liability of the employee, determined according to the normal principles of
delict (including fault), is transferred to the employer. Other instances of strict liability are provided
for in legislation, for example in respect of damage caused by aircraft,1 and damage to
telecommunication lines.2
The core moral principle of delict is that it reflects personal responsibility, based on the idea
that there can be no liability without fault. Strict liability expresses a different morality, in that
society determines that the nature of the conduct, or the risk associated with the conduct is such that
the responsible person or entity should compensate anyone who suffers harm as a result of the
conduct. Hence Cane’s view that strict liability represents ‘personal responsibility for outcomes’,
and that strict liability operates as a form of a tax or a loss-distribution mechanism.3 The ethical
imperative that underlies strict liability is to compensate for harmful outcomes of the relevant
activity. Viewed in this way, strict liability is a type of tax on activities that attract such liability
rather than a penalty for engaging in it. Strict liability is the product of a modern era when
industrial and technological development exposes individuals to widespread risks. There is often no
direct relationship between those creating or spreading a risk, and those exposed to the risks. The
economic and socio-political objective of no-fault liability for harmful outcomes is that there
should be loss-distribution: those that create or spread a risk are often better placed to carry the
burden of harm than the individuals who suffer the harm in the first place.

32.3 Liability for harm caused by animals

Certain Roman actions that involve strict liability for harm caused by animals still form part of
modern South African law.

32.3.1 The actio de pauperie for harm caused by domestic and

domesticated animals
The owner of a domestic or domesticated animal is strictly liable for the harm that the animal
causes to another person when it acts contrary to its nature (contra naturam sui generis) and from
inward excitement or vice (sponte feritate commota).
Strict liability for harm caused by animals is based on the consideration that animals create a
source of danger in the human environment – they kick, butt, gore, stray on busy streets, attack,
jump, chase and bite.4 Many other things also create danger, but the special risk involving animals
is their instinctive, unpredictable behaviour and their mobility.
The law of liability for harm done by animals developed in South Africa on the basis of the
strict liability regime of the Roman actio de pauperie and edictal remedies.5 After initial doubt
about whether the actio de pauperie applied in South African law,6 the Appellate Division held in
O’Callaghan NO v Chaplin 7 that it does form part of South African law, on the basis of a
presumption of fault.8 Courts have now generally accepted it as an action based on strict liability
arising out of the mere ownership of the animal.9

32.3.1.1 Type of animal

The actio de pauperie in Roman law applied to livestock and domestic animals, and was gradually
extended to include all types of domesticated animals. Our law now recognises pauperian liability
for all types of domestic animals and livestock, and also for domesticated wild animals or other
wild creatures employed for domestic use, such as bees or meerkats.10
The special risk of animals in a human environment is not only their aggression, but also their
instinctive, unpredictable behaviour and mobility. Therefore, the policy that underlies recognising
strict liability for harm done by animals does not restrict liability to vicious or aggressive behaviour
of animals. However, courts have tended to restrict liability to such behaviour, excluding liability
for behaviour that one would describe as over-friendly, for example, where a dog stands up against
a frail person and causes that person to fall. There is no requirement that the animal must have
shown previous signs of viciousness or unpredictable behaviour. We refer to this aspect below in
the context of the type of behaviour giving rise to liability.

32.3.1.2 Person who could be liable

The person liable is the owner of the animal at the time when the injury or harm was caused.
Ownership is determined in accordance with the accepted principles of property law. Control of the
animal is not sufficient for liability in respect of the actio de pauperie, but the person in control
may be liable under the Aquilian action.
The special risk of having animals in a human environment is their instinctive, unpredictable
behaviour and mobility. The keeper of an animal is a party to creating or maintaining this risk, and
so is in a position to contain the risk by controlling or restraining the animal. The keeper may also
benefit from keeping the animal. Therefore, the policy considerations that underlie recognising
strict liability for harm done by animals, favour the view that liability should include keepers and
not be restricted to owners. The owner of an animal is liable even if he or she was not in possession
of the animal when it caused harm. However, the owner can rely on the negligence of the keeper of
the animal at the time as a defence.11

32.3.1.3 Type of behaviour – the contra naturam test

For liability under the actio de pauperie the animal must have acted contra naturam sui generis. In
Roman law the implication of this requirement was that the animal must have acted spontaneously
and not in response to an external factor, such as provocation. However, the owner was liable if the
animal reacted to stroking or patting. The contra naturam rule has been described as a conveniently
flexible concept to determine where the risk of damage or injury should lie. It was not meant to
focus on the disposition of the individual animal (for example, whether it was a generally placid or
unruly horse), nor on the disposition of the species concerned (for example, whether a horse or dog
was by nature aggressive), but to indicate uncharacteristic behaviour of a domesticated animal in a
human environment.12
Courts interpret the contra naturam rule inconsistently. Some cases refer to the innate
wildness, viciousness or perverseness of the particular animal (a subjective approach), while others
refer to what one could expect of a well-behaved animal of its type (an objective or ‘reasonable
animal’ approach).13 A third approach, often used, is to consider both subjective and objective
factors and to apply common sense to a particular situation.
Da Silva v Otto14
The appellant had been walking his dog on a leash in a public road in a residential area,
when the respondent’s dog ran out of an open side gate on the respondent’s premises and
attacked the dog on the leash. The appellant, who carried a whip, hit the respondent’s dog,
which then bit him. The Court found that the respondent had not acted negligently as the dog
had never bitten a person before and he had taken sufficient precautions against the
possibility of the dog escaping from his premises. Furthermore, the Court held that the
appellant had also not acted negligently as he had not ‘provoked’ the dog by hitting it and was
entitled to protect his dog against an attack. The Court held that in such circumstances,
where the injured person had acted lawfully and reasonably, the dog was presumed to have
acted contra naturam sui generis. The judgment was that an objective test of the reasonable
dog applied and the dog was expected to distinguish between a lawful attack and an unlawful
attack.

Green v Naidoo15
A four-year-old girl and her older sister were visiting the defendant’s son. At some stage
during the visit the girl pulled a Chow dog’s nose while it was eating. It retaliated by biting the
child in the face. The Court rejected a suggestion that the two girls were trespassers, in that
they were accepted as guests in the household and were not intruders. Without expertise in
dog psychology and in the absence of expert evidence in that regard, the Court had some
difficulty with the contra naturam sui generis requirement, however, especially with the
suggest rationale that domesticated animals should behave as well-behaved animals of their
kind would16 and be able to control themselves; and if they do not, they are to be regarded as
having acted contra naturam sui generis.17 Instead it resorted to applying the following
approach:

Notwithstanding that there are so many configurations of four legged animals, genetic
tendencies and learned behaviour, I can do no more than approach the dog in
question, Taz, as yet another exemplar of a pet dog and apply my own common sense
and limited experience to the questions to be answered.

The crisp question is whether it is contrary to the nature of a dog which is a household
pet to bite a child in the garden of that house? The crisp answer must be in the
affirmative.

The Court thereafter considered the defence of provocation. In the course of finding that the
child’s conduct constituted an external interference that caused the dog’s unnatural behaviour
and thus exonerated the owner of the dog,18 it noted:19
• While a doli incapax child ‘cannot be held liable in law for an act she may have performed
but it does not mean that the act was not performed’
• The enquiry ‘is directed at ascertaining the conduct expected of a reasonable dog acting in
accordance with its nature when faced with such a situation’
• A dog should not be expected ‘to distinguish whether or not his assailant is doli or culpa
capax or incapax’
• External factors are not limited to human conduct and vis maior is part of this defence.

PAUSE FOR Reasonable dog behaviour

1. Should courts require expert evidence in each case to determine whether the
REFLECTION animal behaved naturally? Is this a factual issue or a conclusion of law?
Compare the concern expressed in Green v Naidoo 20 with Thysse v Bekker 21
where expert evidence was presented and where the Court said:22
 [I]t may be that [the expert, Prof Odendaal’s] hypothesis, based
as it is upon scientific principles, gives a reasonable explanation
of the dog’s aggressive behaviour. But two things must be made
clear. One is that it is my function, not Prof Odendaal’s, to
decide whether or not the dog acted contra naturam sui generis.
While I may find guidance in Prof Odendaal’s expert opinion, I
must come to my own conclusion. The second is that Prof
Odendaal’s conclusion is based on scientific criteria which are
different from the legal criteria which I must apply. Professor
Odendaal explains the behaviour of this particular dog from its
point of view. I must evaluate the dog’s behaviour against a
different and more general standard – the standard of behaviour
which the law expects of a domesticated animal generally. The
issue is not whether Prins [the dog] behaved according to its
own nature, which is the test applied by Prof Odendaal, or to the
nature of its breed. It is whether the dog behaved in a manner
which the law considers acceptable by animals which share the
human environment with human beings because they have over
the ages become domesticated.

2. How should a disciplined dog behave when provoked or attacked? Did the
courts in Da Silva v Otto 23 and Green v Naidoo 24 have different views on what
a reasonable dog can be expected to distinguish?

The policy that underlies recognising strict liability for harm caused by animals suggests that courts
should not restrict liability for injury to the vicious or aggressive behaviour of animals.
To apply the contra naturam requirement, the focus should not be on the particular species or
breed to which the animal belongs, but should be more general, on uncharacteristic behaviour of a
domesticated animal in a human environment. However, courts tend to have a narrower focus,
understanding this requirement to mean that the animal must have acted contrary to the nature of its
genus.25 In other words, if a Pitbull Terrier bites a person, the question is whether the dog acted
contrary to the nature of a well-behaved dog (which does not normally bite people) and not whether
it acted contrary to the nature of a Pitbull (which might do so).

32.3.1.4 Elements of the action

To succeed in this action, the plaintiff will have to prove six essential elements:26
• Harm
• Conduct by a domesticated animal
• A causal link between the conduct and the harm
• The defendant must have been the owner of the animal at the time of the injury
• The animal must have acted contrary to the nature of its breed – contra naturam sui generis
• The animal must have acted from inward excitement or vice – sponte feritate commota.

32.3.1.5 Defences
The following are recognised defences to an actio de pauperie:
• The defendant can rely on an external factor that provoked the animal’s harmful behaviour.
This can occur by a natural phenomenon (vis maior); for example, when an animal reacts to a
sudden thunder clap. ‘The authorities contemplate instances of animals discomforted by loose
 harnesses, alarmed by lightning and thunder and other natural forces.’27 Such a reaction is
natural (not contra naturam sui generis) and does not stem from inward excitement or vice.
• Culpable conduct by the plaintiff that resulted in the harm will defeat a claim, for example,
where the plaintiff provoked a dog and was then bitten.28 The level of the plaintiff’s fault must
amount to ‘substantial negligence or imprudence’, but must amount to more than stroking or
petting a strange horse or dog.29 Contributory negligence will not result in courts apportioning
damages in terms of the Apportionment of Damages Act 34 of 1956. To date, courts have
adopted an all-or-nothing approach, although not specifically excluding the possibility of
apportionment.30
• Voluntary assumption of the risk of harm by the plaintiff will also defeat the claim; for
example, where the plaintiff knowingly and deliberately runs the risk of injury by taking a dare
and entering an area that holds a fierce bull.
• The owner of the animal can rely on a prior contractual undertaking not to claim damages in
the event of harm occurring (pactum de non petendo); for example, where the plaintiff signed
such an undertaking before going horse-riding on the owner’s estate.31
• The owner can rely on the fact that the plaintiff was unlawfully present at the place where the
harm occurred. There are different interpretations of this defence. Some cases require a ‘legal
right’ for the plaintiff to be lawfully present (for instance by invitation or with consent of the
owner or occupier), and others only a ‘lawful purpose’ (for instance to make a delivery). In
most cases, the two approaches lead to the same result if the court accepts that the owner or
occupier of the place tacitly consented to the person entering for a lawful purpose.
• The owner can rely on the negligence of the keeper of the animal at the time of the incident.32
Normally, the owner of an animal is liable even if not in possession of the animal at the time
when it caused harm. However, negligence on the part of a keeper will defeat the claim against
the owner.33
Lever v Purdy34
The plaintiff (respondent) had been bitten by a dog that belonged to the owner (appellant).
The incident occurred while the owner was overseas, leaving a custodian in charge of his
house and dog. The custodian had summoned the plaintiff, a repairman, to the house and
had advised him that there was a vicious dog on the premises. The plaintiff had requested the
custodian to lock the dog away before his arrival at an agreed time. When the plaintiff arrived
at the house and entered the premises, the dog had not been locked away, and it bit him. It
was common cause that the plaintiff had established the essentials of the actio de pauperie
against the owner, but the question was whether the owner was exonerated by the
negligence of the custodian.
Both Joubert ACJ35 and Kumleben JA36 listed categories of culpable third-party conduct
that constituted a defence to pauperian liability. The first category involves inciting or
provoking an animal to behave contra naturam sui generis, by striking, wounding, scaring or
annoying the animal. The causa causans of the harm suffered is the inciting behaviour and
not the animal’s vicious propensities. Courts generally treat this as an ‘established exception’
to pauperian liability. The second category involves culpable conduct, which contributes to
the injury, but falls short of being the causa causans of it, for example, where a visitor
facilitates injury by leaving open the owner’s gate so that a vicious dog can get out. The third
category involves culpable lack of control of a vicious animal by the person to whom control
of the animal has been entrusted, as in the Lever case. The majority of the Appellate Division
held that the negligence by the custodian afforded the absent owner a defence to pauperian
liability.
Arguably, the injury was caused by the custodian’s negligent conduct rather than by the
materialising of a typical risk of owning a dog37 and in this case illustrates that the issues of
strict liability of a custodian, and joint and several liability of the owner and the custodian,
remain unresolved in South African law.
 Fourie v Naranjo38
Fourie’s dog had attacked and savaged his domestic worker. Naranjo hastened to the
worker’s assistance and was in turn attacked and bitten by the dog and also by another dog
(not owned by Fourie) that was on the premises for the purposes of mating with Fourie’s dog.
Naranjo’s wife witnessed the incident. Naranjo sued in respect of the injuries he sustained
and his wife sued for emotional shock and related medical expenses.
The Court found that Fourie’s dog had acted contra naturam sui generis according to the
objective test of the reasonable dog, that Naranjo did not consent to the risk of injury when
he entered the property (the defence of volenti non fit iniuria requires knowledge of,
appreciation of, and consent to harm or the risk of harm), and that Naranjo had not been
negligent in acting as he had done (this would have been a complete defence against the
actio de pauperie).39
The interesting and unique feature of this case was the award of damages for emotional
shock in the context of the pauperian action. Fourie had contended that the wife’s claim for
emotional shock was not competent, as she had not been attacked by the dog. However, the
Court noted that it is trite that an action in delict may lie for patrimonial or sentimental
damages arising out of the infliction of emotional shock.40
The Court pointed out that a person bitten by a dog is entitled to damages not only for the
direct injury sustained, but also for subsequent physical disorders caused by the nervous
shock; that there is support for the view that damages may be claimed by a dependant for
loss of support if an animal has caused the death of a breadwinner; and that the extent of
Fourie’s liability should be limited only in accordance with the flexible criterion of legal
causation. There was thus no objection in principle to the actio de pauperie being extended
to recover damages for emotional shock.41

32.4 The actio de pastu for harm caused by grazing animals

South African law, following the Roman actio de pastu, recognises strict liability for the owner of a
domesticated animal that trespasses onto another’s land and causes harm by eating plants. The
damages that may be claimed are not restricted to those resulting from the grazing activity only, but
also includes damages in respect of any collateral damage caused while grazing, such as trampling
on plants and breaking fences. As in the case of the actio de pauperie, liability under the actio de
pastu is based on ownership of the animal.42
To succeed, a plaintiff will have to prove the following:
• The defendant is the owner of a domesticated animal that trespassed on the plaintiff’s property
and caused harm to the plaintiff.
• The harm was caused by grazing, and occurred in the form of damage to plants, and can
include collateral harm such as breaking fences. 43
• The animal must have acted of its own accord and was not provoked or prompted into
behaving in this way.

The defences to the action are the same as for the actio de pauperie, but culpable conduct on the
part of an outsider has not been recognised as a defence for the owner.

PAUSE FOR The actio de pauperie and the actio de pastu

REFLECTION What are the differences, if any, between the elements of the actio de
pauperie and the actio de pastu, and how do they compare with those of
the Aquilian action? Is there any merit in the contention that the elements
of liability for human behaviour are mirrored in the elements for liability in
respect of domesticated animals? Are the situations sufficiently
analogous for such comparisons to be drawn? If yes, should the law
 adapt the requirements for these actions to mirror the elements of the lex
Aquilia more closely? To answer these questions it may be useful to draw
a grid, similar to the one below.

32.5 The actio de effusis vel deiectis and the actio positis vel
suspensis
South African law has retained remnants of the Roman law actio de effusis vel deiectis (available
against the occupier of a building for harm caused by throwing or pouring an object or substance
out of a building) as well as the actio de positis vel suspensis (used to impose a criminal fine on the
occupier when an object falls from a building to the street below). Even though these actions have
not been rejected, there has however been no significant development of strict liability by applying
these actions in modern law.44
The few claims brought in recent years have not had success: In Colman v Dunbar 45 a claim
based on the actio de effusis vel deiectis failed. This was not because the Court refused to recognise
the applicability of the Praetor’s Edict, but rather because the action had been brought against a
builder working on the premises (against whom there was no evidence of negligence) and only an
occupier could be held liable.
A claim based on the actio de effusis vel deiectis was similarly unsuccessful in Bowden v
Rudman,46 again because the facts of the case did not meet the terms of the Edict. In this case, the
plaintiff claimed damages for an injury suffered due to a gate swinging open across a pavement,
admittedly without negligence on the part of the occupier. The Court ruled that the actio was not
available since it applied only to objects falling after being placed on, or suspended from a
building.

32.6 Statutory instances of strict liability

Modern legislation has created certain additional forms of strict liability. The common feature of
these forms of liability is that they concern activities involving a considerable risk of harm.

32.6.1 Aviation
The Aviation Act 74 of 1962 imposes strict liability on the owner 47 of an aircraft where material
harm or loss is caused by the aircraft in flight, on taking off or landing, by any person in such
aircraft, by any article falling from such aircraft, or to any person or property on land or water.48
The owner is not liable, however, if the harm or loss is caused, or contributed to by the negligent or
wilful act of the claimant.49 The intention appears to be to create a complete defence and to exclude
the provisions of the Apportionment of Damages Act 34 of 1956.
32.6.2 Damage to telecommunication lines and call boxes
The Post and Telecommunication-related Matters Act 44 of 1958 50 imposes strict liability on any
person who directly or indirectly injures or destroys any telecommunications lines or call boxes.51
However, where the telecommunications company is of the opinion that it is justified, it may bear
the whole or part of the expenses necessary to make good such injury or destruction.52 If a person’s
negligent conduct caused the injury or destruction, then that person will also be guilty of an offence
for which a court can impose a fine.53

32.6.3 Genetically modified organisms

The Genetically Modified Organisms Act 15 of 1997 provides that the user concerned should be
liable for harm caused by using or releasing a genetically modified organism. No mention is made
as to whether fault needs to be proven, nor to any presumption of fault.54 The Act provides that
liability for harm caused by the use or release of a genetically modified organism shall be borne by
the user concerned.55 However, when the harm is caused while the organism is in the possession of
an inspector appointed under the Act, the user will not be liable unless the harm was foreseen or
foreseeable and preventable, and the user failed to take reasonable action to prevent the harm.56
Users are also required to ensure that appropriate measures are taken to avoid an adverse
impact on the environment that may arise from using genetically modified organisms.57

1 Aviation Act 74 of 1962, section 11.

2 Post and Telecommunication-Related Matters Act, section 108.
3 Cane The Anatomy of Tort Law (1997) at 51–52.
4 Zimmermann The Law of Obligations: Roman Foundations of the Civilian Tradition (1990) at 1095; and see the
remarks of Kumbleben JA in Lever v Purdy 1993 (3) SA 17 (A) at 29C–G.
5 The historical development of liability for harm done by animals is discussed in detail by Van der Merwe Skuldlose
Aanspreeklikheid vir Skade Veroorsaak deur Diere Thesis/Dissertation University of South Africa (1971).
6 Parker v Reed (1904) 21 SC 496.
7 1927 AD 310.
8 O’Callaghan NO v Chaplin 1927 AD 310 at 378. See also Bristow v Lycett 1971 (4) SA 223 (RA).
9 SAR & H v Edwards 1930 AD 3 at 9; Green v Naidoo 2007 (6) SA 372 (W) para 10.
10 Van der Merwe and Rabie LAWSA Animals (2014) Vol 1 (3 ed) para 403.
11 Lever v Purdy 1993 (3) SA 17 (A) at 19.
12 Zimmermann (1990) at 1103.
13 See, generally, Van der Merwe and Rabie (2014) para 406.
14 1986 (3) SA 538 (T).
15 2007 (6) SA 372 (W).
16 SAR & H v Edwards 1930 AD 3 at 9–10.
17 Solomon v De Waal 1972 (1) SA 575 (A) at 582E.
18 Green v Naidoo 2007 (6) SA 372 (W) para 33.
19 Para 24.
20 2007 (6) SA 372 (W).
21 2007 (3) SA 350 (SEC); [2007] 4 All SA 1309 (SE).
22 Para 9 (footnotes omitted).
23 1986 (3) SA 538 (T).
24 2007 (6) SA 372 (W).
25 Loriza Brahman v Dippenaar 2002 (2) SA 477 (SCA).
26 Visagie v Transsun (Pty) Ltd [1996] 4 All SA 702 (Tk).
27 Green v Naidoo 2007 (6) SA 372 (W) para 24.
28 O’Callaghan NO v Chaplin 1927 AD 310 at 329.
29 Van der Merwe and Rabie (2014) para 409; O’Callaghan NO v Chaplin 1927 AD 310 at 329.
30 See Portwood v Svamvur 1970 (1) SA 144 (R) at 145; Swart v Honeyborne 1981 (1) SA 974 (C) at 976B; Fourie v
Naranjo [2007] 4 All SA 1152 (C); 2008 (1) SA 192 (C) para 19.
31 Walker v Redhouse 2007 (3) SA 514 (SCA).
32 Lever v Purdy 1993 (3) SA 17 (A).
33 Walker v Redhouse 2007 (3) SA 514 (SCA).
34 1993 (3) SA 17 (A)
35 At 21C–22A.
36 At 26E–H.
37 Van der Merwe ‘The defence of conduct of a third party in view of the rationale for strict liability in terms of the
pauperien action revisited’ (1994) 111(1) SALJ 47 at 52–53.
38 [2007] 4 All SA 1152 (C); 2008 (1) SA 192 (C). See also para 3.1 above.
39 See paras 15–19.
40 Para 23.
41 Para 24.
42 Van Zyl v Van Biljon 1987 (2) SA 372 (O).
43 Potgieter v Smit 1985 (2) SA 690 (D) at 695.
44 Zimmermann (1990) at 1126 states: But although all four remedies were incorporated in the ius commune, none
of them can be said to have prospered.
45 1933 AD 141.
46 1964 (4) SA 686 (N).
47 In terms of section 11(6), if the owner bona fide leases or hires out the aircraft to another person for a period
exceeding 14 days and no operative members of the crew are in the owner’s employ, then the person to whom the
aircraft was leased or hired out is responsible on the same basis as the owner would have been in terms of section 11
(2).
48 Section 11(2).
49 Section 11(3).
50 Previously known as the Post Office Act.
51 Section 108; and see Esterhuizen v Minister van Pos & Telekommunikasiewese 1978 (2) SA 227 (T) at 229; Telkom
(SA) Ltd v Duncan 1995 (3) SA 941 (W) at 943–945.
52 Section 108.
53 Section 108.
54 Section 17(2).
55 Section 17(2).
56 Section 17(2).
57 Section 17(1).
 Chapter 33

Vicarious liability

33.1 Introduction

33.2 General rule and justification for vicarious liability

33.3 Who qualifies as an employee?

33.4 An employment or akin-to-employment relationship must exist

33.5 Employee of two employers

33.6 Independent contractors

|

33.7 The delict must be committed by the employee while acting within
the course and scope of employment

33.8 Intentional wrongdoing

33.9 Vicarious liability and the Protection of Personal Information Act 4 of

2013

33.1 Introduction
Vicarious liability of employers is a form of strict liability. The employer is held liable without
fault for an employee’s wrongdoing and the delictual liability of the employee is transferred to the
employer. The doctrine of vicarious liability, as it is applied in South Africa, is predominantly
English in derivation.1 In the context of South African law, it has been described as ‘perhaps the
most comprehensive and far-reaching innovation we have taken from English law’. 2 South African
and English courts approach the two cardinal features of the doctrine – the concept of an employee
and that of acting in the course and scope of employment – in a similar way, although there are
differences in detail.

33.2 General rule and justification for vicarious liability

The general rule is that an employer is vicariously liable, regardless of the absence of fault on the
employer’s part, for employees’ delicts when these are committed in the course and scope of their
employment.3 The employer is bound in solidum (jointly liable) with the employee to compensate
the person who suffers harm as a result of the employee’s delict.4
There have been a number of theories to explain and justify imposing liability on an employer
in this way:
• It is sometimes suggested that employers are in control of the behaviour of the employees. This
suggestion is no longer convincing as many employees perform skilled tasks and employers
are often incapable of controlling or even understanding what the employee is doing.
• The rule is also said to be ‘founded upon public policy and convenience, for in no other way
could there be any safety to third persons in their dealings either directly or indirectly with him
through the instrumentality of agents’. 5
• The employer can be considered to be the actor where he or she acts through an employee –
qui facit per alium facit per se.
• Employers can be said to be the best ‘risk-absorbers’, because they can distribute the risk of
accidents cost-effectively by insurance, and take precautionary measures by allocating
resources or incentive devices tailored to their type of enterprise.
• The doctrine can also be justified on a moral basis, as the employee inflicts loss on the victim
while pursuing the employer’s business interests. Since the employer obtains a benefit from
the employee’s work, the employer should also bear the costs of accidents arising out of such
work.
• In the context of the employment relationship, the employer creates a risk for his or her own
ends and the employer’s liability is therefore ‘co-extensive and identical in every respect with
the liability of the servant’. 6 The ‘creation of risk’ theory is a common justification for
vicarious liability. The judgment of the Constitutional Court in K v Minister of Safety and
Security 7 indicates that the creation of risk can also play a direct role in establishing whether
courts should impose strict liability in a particular situation. Where the employer creates a risk
that third parties may be harmed whilst the employee is promoting the business or interests of
the employer, it is fair to hold the employer liable.8

33.3 Who qualifies as an employee?

Courts often regard the employment relationship or contract of service (locatio conductio
operarum) and the contract of services (locatio conductio operis) as distinguishable for purposes of
vicarious liability. This is because in the former instance an employee is under the control of the
employer in respect of the nature of the work and the manner in which it is to be done, whereas an
independent contractor is not subject to the control of the person paying for the services.9 The same
applies to mandatories.
The current approach is to regard the right to exercise control and direction over the employee
as merely one of the indicators of an employment relationship.10 Courts apply a multi-faceted test,11
taking into account all the circumstances, such as the nature of the work, the manner of payment,
the manner of performing the work, state of social and economic interdependence, authority to give
instructions, membership of medical or pension fund, paid vacation, working hours, use of
employer’s premises and equipment, and whether the work and the person performing it form an
integral part of the business organisation.12 Employees are subject to the control or directives of
their employers and their services are terminated by death or on expiration of the agreed period. On
the other hand, independent contractors have contracts to perform certain specified work and are
ordinarily not subject to the directives of persons who employ them. Therefore, a person is liable
for the acts of an independent contractor only in respect of operations where there is a ‘non-
delegable’ duty, in other words, where engaging a contractor does not absolve the employer from a
duty not to harm third parties, such as when the operations involve an abnormal level of danger.13

33.4 An employment or akin-to-employment relationship

must exist
Vicarious liability requires a delict by an employee who is acting in accordance with the
instructions of and on behalf of the employer. Vicarious liability arises from all types of
employment relationships, including those that involve skilled or professional employees who act
with a large measure of independence, such as medical doctors.14
Vicarious liability can also arise from relationships other than employment:

• Individual members of a partnership are jointly and severally liable to a person other than a
partner for any delict that a partner commits while acting on partnership business or with the
authority of co-partners.15 Therefore, for vicarious liability, it is not a person’s capacity as a
partner, but the capacity in which the person acts for the partnership that is important.16
• As a general rule, a principal is not liable for the unauthorised acts of his or her agent, even if
the act was part of the mandate,17 unless:
◆ The agent was acting in the capacity as agent and within the scope of his actual or
ostensible authority 18
◆ The agent was also an employee and acting within the course and scope of his or her
employment.19
• The owner of a motor vehicle who allows another person to drive it may be vicariously liable
for the latter’s negligent driving if:
◆ The vehicle is being driven in the owner’s interest 20
◆ The relationship between the owner and the driver is such that the owner retains the right to
control the way in which the vehicle is driven.

In the leading case, Messina Associated Carriers v Kleinhaus,21 the Court held that the true enquiry
is whether the relationship between the owner and the driver, and the interest of the owner in
driving the vehicle, are sufficiently similar to an employee driving the vehicle in the course of and
within the scope of employment.
Spouses and parents are generally not vicariously liable for delicts of the other spouse 22 or a
child,23 unless the spouse or child was acting in one of the capacities just described.24

33.5 Employee of two employers

A person who is employed by one employer may temporarily become the employee of another for
vicarious liability purposes, typically where an employee is lent or hired by one employer to
another. In Midway Two Engineering & Construction Services v Transnet BPK 25 a labour broker
supplied a driver to a client. The Supreme Court of Appeal had to consider whether the broker or
the client was vicariously liable at common law for damage caused by the driver while driving for
the client. Since the Court regarded the temporary employer as the one most closely associated with
the risk-creating event, the Court found the temporary employer to be vicariously liable. The fact
that the agreement between the two employers labelled the driver as the employee of the labour
broker, was therefore not decisive.
The onus is on the general employer to show that, in the circumstances, it did not have control
over its employee. The principle is that the general employer retains responsibility unless it can
establish that in the transfer, another person temporarily became the employer.

33.6 Independent contractors

Generally in South African law, the employer of an independent contractor is not vicariously
responsible for the wrongful conduct of the contractor.26 This is because the independent contractor
carries out a specific piece of work (locatio conductio operis) 27 and is not normally subject to the
control or directions of the employer when performing this work.28 A defendant cannot avoid
vicarious liability by stating in a contract that a person is appointed as an independent contractor, if
the relationship is in fact that of employment.29
Stein v Rising Tide Productions CC30
The Court had to consider the vicarious liability of a company for a delict committed by a
technical crew hired by Rising Tide (the company) to assist during a film shoot. Rising Tide
hired the crew to provide technical skills that the company did not possess. The crew brought
its own equipment and used it at its own discretion without any direction from the company.
Its employer subsequently invoiced the company and received payment from the company.
The Court identified a number of factors that indicated that the technical crew was not an
employee of the company, including:
• The nature of the work
• The existence or non-existence of a right of supervision on the part of the employer
• The manner of payment (for example, whether the employee is paid a fixed rate or a
commission)
• The relative dependence or freedom of action of the employee in the performance of his or
her duties
• The employer’s power of dismissal
• Whether the employee is precluded from working for another
• Whether the employee is required to devote a particular amount of time to his or her work
• Whether the employee is obliged to perform his or her duties personally
• The ownership of the working facilities and whether the employee provides his or her own
tools and equipment
• The place of work
• The length of time of the employment
• The intention of the parties.

Courts determine the liability of an employer for harm done by an independent contractor by the
ordinary rule that a person must exercise the degree of care demanded by the circumstances.31 This
will, for instance, require that courts consider the nature and extent of the danger, the degree of
expertise available to the employer and the independent contractor respectively, and the reasonably
practicable means available to the employer to avert the danger.32 In Langley Fox Building
Partnership (Pty) Ltd v De Valence 33 the Court had to decide on the liability of the owner of a
building for damage caused when a building contractor left a temporary beam extending over a
pavement. Goldstone AJA said the following:
In my opinion, it follows from the aforegoing that in a case such as the present, there are three
broad questions which must be asked, viz:
 (1) would a reasonable man have foreseen the risk of danger in consequence of the work he
employed the contractor to perform? If so,
(2) would a reasonable man have taken steps to guard against the danger? If so,
(3) were such steps duly taken in the case in question?
Only where the answer to the first two questions is in the affirmative does a legal duty arise, the
failure to comply with which can form the basis of liability.

Therefore, the liability of an independent contractor’s employer is governed by the ordinary

principles of negligence, and the question is whether the harm was reasonably foreseeable and
preventable by the employer.
Chartaprops 16 (Pty) Ltd v Silberman34
The plaintiff, Mrs Silberman, slipped and fell in one of the passageways of a shopping mall,
where there was a pool of a slippery substance that had been spilt on the floor. The shopping
mall was owned by, and under the control of Chartaprops, which had engaged a contractor,
Advanced Cleaning, to keep the floors of the mall clean. Advanced Cleaning had a system in
place for cleaning the floors, so that every part of the floor should ordinarily have been
passed over by one or other of their cleaners at intervals of no more than five minutes.
Chartaprops itself regularly checked the contractor’s performance. Each morning, its centre
manager met the cleaning supervisor and personally inspected the floors of the mall to
ensure that they had been properly cleaned. If he encountered litter or spillage, he would
arrange for its immediate removal. The Trial Court found that the substance had been on the
floor for thirty minutes or more when the incident occurred.
Mrs Silberman sued both Chartaprops and Advanced Cleaning for damages. The
questions were:
• Was the owner, Chartaprops, negligent?
• Was the independent contractor, Advanced Cleaning, negligent?
• If Advanced Cleaning was negligent, was Chartaprops vicariously liable as the employer of
Advanced Cleaning, even if Chartaprops itself was not negligent?

The majority of the Court held that the correct approach to determine the liability of a
principal, such as Chartaprops, for the negligence of an independent contractor, such as
Advanced Cleaning, is to apply the fundamental rule that obliges a person to exercise the
degree of care that the circumstances demand. In this case, the Court should not apply the
general policy of the law that the economic cost of the wrong should be borne by the legal
entity immediately responsible for it. It should not shift the economic cost of negligence from
Advanced Cleaning, the independent contractor with primary responsibility, to Chartaprops,
because of the legal fiction of a non-delegable duty of the owner-employer to ensure safety in
the shopping mall. Most operations entrusted to an independent contractor by an owner-
employer are capable of being sources of danger to others. If courts were to hold the owner-
employer vicariously liable in all such cases, the distinction between ‘employee’ and
‘independent contractor’ will all but disappear.
This was plainly not a case where the Court could say that Chartaprops negligently
selected an independent contractor, that it interfered with the work, or that it authorised or
ratified the wrong. The harm was caused solely by the act or omission of the independent
contractor, Advanced Cleaning, or its employees. Chartaprops did not merely content itself
with contracting Advanced Cleaning to perform the cleaning services. It did more. Its centre
manager met the cleaning supervisor every morning and personally inspected the floors of
the mall on a regular basis to ensure that they had been properly cleaned. Chartaprops did
all that a reasonable person could do to make sure that the floors of the mall were safe. The
duty to make sure that the premises were safe could not have been discharged better than
by employing a competent contractor, as Chartaprops did. Advanced Cleaning had a legal
duty in delict to third parties, to ensure that the floors were clean and safe, and negligently
failed to discharge that duty. Therefore, only Advanced Cleaning was held liable to Mrs
Silberman.
 Note: Also study the minority judgment of Nugent JA, and consider the points of difference
between the majority and minority judgments.

33.7 The delict must be committed by the employee while

acting within the course and scope of employment
Generally, an employer is liable for any wrongful conduct by employees while acting in the course
and within the scope of their employment.35 Broadly speaking, employees are acting in the scope of
their employment when they carry out tasks authorised by the employer, even if they carry out the
tasks by an unauthorised method.36 The problem arises when the employee engages in conduct that
could be described as contrary to, or a deviation from the tasks for which the employee was
appointed. If the employee abandons the tasks of the employer entirely, that is, engages in activities
completely unrelated to, or not in furtherance of the employer’s work, and as a result of these
activities harms an innocent third party, the employer is not liable.
Minister of Police v Rabie 37
To determine whether an employee acted within or outside the scope of employment in the
aforementioned situations, courts apply the ‘standard test’, which in this case was
authoritatively set out as follows:

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 by a servant does so fall, some
reference is to be made to the servant’s intention. The test in this regard is 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.

It is important to note that an employer will be liable:

… even for acts which he has not authorised provided that they are so connected with
acts which he has authorised that they may rightly be regarded as modes – although
improper modes – of doing them … . On the other hand if the unauthorised and
wrongful act … is not so connected with the authorised act as to be a mode of doing it,
but is an independent act, the master is not responsible; for in such a case the servant
is not acting in the course of his employment, but has gone outside of it.38

In the Rabie case, a sergeant in the South African Police, who was employed as a mechanic,
had seriously assaulted, wrongfully arrested and detained, and fabricated charges against
Rabie. The assault and arrest had taken place when the sergeant had been off-duty, dressed
in civilian clothing, in his private vehicle and at the scene clearly pursuing his own private
interests. The sole issue on appeal was whether the Minister of Police was vicariously liable
for the sergeant’s wrongful conduct.
The essential question facing the Court was whether Rabie had proven that the sergeant
was acting in the course and scope of his employment as a servant of the State, that is
whether he was doing the State’s work, namely, police work, when he committed the wrongs
in question.39 Conversely, for the Minister to escape liability, the Minister needed to show that
the nature of the sergeant’s duties was such that he had not been acting as a servant of the
State when arresting Rabie.
The Court reasoned that, even though the sergeant’s work as a mechanic was limited to a
time and place, his work as a policeman – which included questioning, arresting, escorting to
a police station, and charging a suspect – was not so restricted. In the absence of specific
instructions to the contrary, the sergeant could at any time and place perform his functions as
 a policeman, and it is conceivable that in some instances he might have been called upon to
do so in the line of duty.40 On the facts, the sergeant had identified himself as a policeman to
Rabie when he arrested him, and in the circumstances it seemed reasonable and fair to infer
that the sergeant had intended to exercise his authority as a policeman, and was therefore
acting in the course and scope of his employment.41
The Court reached this decision despite the fact that the sergeant had known that Rabie
was innocent and that in these circumstances, there were no grounds for exercising his
authority as a policeman over Rabie. It was clear from the nature of the arrests and the
laying of the false charges that the sergeant acted in this way out of malice. So, we could say
that the sergeant’s conduct was completely self-serving and in bad faith. It followed that the
sergeant, whatever his ostensible conduct, had not and could not in reality have been
performing any of the functions set out in the Police Act 7 of 1958.42 Yet the Court imposed
vicarious liability, because there was a sufficiently close link between the servant’s conduct
for his own interests and purposes and the business of his master.43
The decision is also important for its reference to the issue of risk. The Court reasoned as
follows:

By approaching the problem whether the (the policeman’s) acts were done ‘within the
course or scope of his employment’ from the angle of the 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.

The Court reasoned that giving the policeman certain powers creates a risk of harm to others
for which the State, as the employer, must take responsibility. It followed then that the State
had indeed created the risk by allowing the sergeant to abuse his powers as a policeman for
his own purposes, and the Minister had to be held liable for the sergeant’s wrongs.

So, an employer may be liable even if the employee acted solely for his or her own interests and
purposes in a situation occasioned by his or her employment. The test is both subjective, in that the
employee’s intention is taken into account, and objective, in that the existence of a sufficiently
close link between the employee’s act for his or her own purposes and the business of his or her
employer may render the employer vicariously liable. An employer is liable for acts that it did not
authorise, provided that the acts are connected to acts that it did authorise, in such a way that we
can regard them as modes – although improper modes – of doing authorised acts. If the employee’s
conduct falls outside the work or particular class of work that he or she is employed to do, the
employer is not liable.
Bezuidenhout NO v Eskom44
There is no liability when the employee has been carrying out tasks wholly unauthorised by
the employer. In this case the Court held that an employee’s conduct in transporting a
passenger in contravention of an express prohibition was not performed in the course and
within the scope of his employment. The employer had expressly forbidden the employee
from carrying unauthorised passengers and had thus placed a limitation on the scope of
employment. It was not merely an instruction of how to perform the employer’s business.
After considering all the relevant circumstances, the Court said that it was unfair to hold the
employer liable to a passenger who had in effect assumed the risk of association with the
forbidden conduct of the employee. Also, in terms of both the subjective state of mind of the
employee and the objective test of a sufficiently close link between the employee’s acts for
his own purposes and the business of the employer, the Court held that vicarious liability
should not be imposed.

The difficulty in cases has often been to distinguish between:

• An unauthorised mode of carrying out authorised work (where the employer is liable)
• Unauthorised conduct falling outside the employee’s work (where the employer is not liable)
• Conduct that involves the employer’s time or equipment but that is aimed solely at furthering
the employee’s interests (a frolic of the employee’s own, not rendering the employer liable –
there is no vicarious liability where the employee is engaged in his or her personal affairs,
separate from the employment).45

If, for own purposes, an employee deviates from the employer’s business, the question is whether
the digression is so great in terms of space or time that we cannot reasonably say that the employee
is still acting within the course and scope of his or her employment.
• Courts have held an employer liable where the employee caused harm while smoking on duty,
notably in Viljoen v Smith.46 In this case an employee climbed through a fence to neighbouring
property to relieve himself during working hours, and negligently started a fire when smoking
there.
• In Feldman (Pty) Ltd v Mall 47 a delivery vehicle was used to deliver parcels. The driver
deviated from the job to go for drinks and then negligently caused an accident while returning
to work. The Court held the employer liable.
• In Carter & Co (Pty) Ltd v McDonald 48 the employee collided with a pedestrian while cycling
to the market for his own purposes during working hours. The Court held that the employee
had acted in his own interests and the employer was therefore not liable. The employer does
not necessarily escape liability because he prohibited the harmful conduct of the employee. It
depends on all the circumstances.
• In Bezuidenhout NO v Eskom 49 (previously referred to in more detail) the Court held that an
employee’s conduct in transporting a passenger in contravention of an express prohibition by
his employer was not performed in the course and within the scope of his employment.

In summary, the ultimate question is whether the employee not only subjectively promoted his or
her own interests, but objectively was also dissociated from the business affairs or work of the
employer. In other words, in terms of the ‘standard test’ formulated in the Rabie case, the question
is whether the employee intended to promote his or her own interests, and in so doing deviated
from the work of the employer to such a degree that there was no longer a sufficiently close link
between the employee’s acts and the scope of employment.

33.8 Intentional wrongdoing

A difficulty arises when an intentional wrong is conducted in the course of employment. In
Salmond’s often-cited formulation, vicarious liability arises when the employee has done
‘fraudulently that which he was authorised to do honestly’ but not when the employee’s conduct
was insufficiently ‘connected with the authorised act as to be a mode of doing it’. 50 This distinction
depends upon the close scrutiny of the employee’s authorised functions, resolving often into a
question of ‘degree’. 51 For example, in Ess Kay Electronics (Pty) Ltd v First National Bank of
Southern Africa Ltd 52 the Court found the bank not liable when one of its employees
misappropriated blank bank drafts. The employee’s work had in fact included the control and
administration of the bank’s foreign exchange department. However, the Court ruled that there was
no liability, because the employee had been promoting his own interests to the exclusion of the
bank’s, and in acting wrongfully, he had effectively disengaged himself from his appointed
functions. In this case, the Court rejected the plaintiff’s argument that there was a sufficiently close
link between the unauthorised and wrongful acts of the employee and the work he was authorised
to do. Although the employee exploited opportunities offered by his employment, his self-interest
in doing so removed his conduct from the category of authorised actions.
 A similar situation occurred in Absa Bank Ltd v Bond Equipment (Pretoria) (Pty) Ltd,53 where
cheques payable to the employer were paid into a bank account operated by the employee. Despite
the fact that it was the employee’s duty to collect and deposit cheques for his employer, the Court
ruled that in stealing the cheques for himself, he could not be said to be performing his duties at all:
he was on a frolic of his own.
Clearly intentional wrongdoing does not always remove the employee from the scope of
employment, as illustrated by cases on State liability for police transgressions.54 The State (like
other employers) can in certain circumstances be liable for intentional misconduct of its employees.
In the case of policemen, the approach of courts has been to establish whether there is some
connection between their misconduct and carrying out police duties.55
The Constitutional Court has considered whether the common-law principles on vicarious
liability conform with the spirit, purport and objects of the Bill of Rights, as intended in section 39
(2) of the Constitution.56 The Court held that both the principles of vicarious liability, and their
application, must conform to the normative framework of the Constitution. On the issue of whether
the State should be liable for rape committed by police officers while on duty, the Court found the
principles of vicarious liability to be consistent with constitutional norms. When applying these
principles, the Court found that, subjectively viewed, the policemen had acted in pursuit of their
own objectives, but objectively, their conduct was sufficiently linked to their employment as
policemen, and held the State liable.
In Minister of Safety and Security v Luiters 57 both the Supreme Court of Appeal and the
Constitutional Court decided that they should impose vicarious liability on the State when an off-
duty policeman pursued persons who had attempted to rob him and shot an innocent third party.
The Courts applied a two-stage test:
Whether the employee’s acts were committed solely for employee’s purposes, and if so
whether there was a sufficiently close link between employee’s acts and the employer’s
purposes and business.

The Courts held that in pursuing would-be robbers the policeman had acted in both his own
interests and those of the police service. He intended to perform police duties. The fact that the
policeman did not adhere to the rules of criminal procedure or police standing orders was no bar to
liability.
In Minister of Finance v Gore NO 58 the Court imposed vicarious liability on the State for the
conduct of public officials who fraudulently awarded a government tender. The Court held that the
conduct of the officials closely resembled duties performed in the course and within the scope of
their employment. There was a sufficiently close link between their conduct and their employment,
and policy considerations supported the imposition of vicarious liability for fraud.
The measure of a ‘close and direct connection’ depends on a diverse range of subsidiary
criteria, including the nature of the relationship between the employer and the injured third party.
An employer who has taken responsibility for the plaintiff’s person or possessions owes the
plaintiff duties that are more extensive than those owed by the public at large. The fact that there is
a pre-existing relationship between the employer and the injured third party has been an important
factor in some cases. Where employees have misappropriated goods entrusted to their care by their
employers, the employers are generally held accountable for such wrongdoing, and courts often
take into account the employer’s pre-existing duty to the owner of the goods.59
South African courts follow a multi-faceted approach based on policy considerations and
subsidiary factors when determining the closeness of the connection between the harmful conduct
and the employment. The policy considerations and subsidiary factors include the following:
• Whether the employer is, in the circumstances, the most able to spread the risk and whether the
employer equitably should be the ‘risk-absorber’
• The time and place of the act in relation to the duties of employment
• The use of the employer’s property or equipment by the employee
• The extent to which the employer had authorised or prohibited the employee’s conduct
• The extent to which the respective interests of the employer and employee were served by the
harmful act
• Whether a pre-existing relationship between the employer and the plaintiff create reliance on
safety or security
• Whether the employee’s wrongdoing constituted the realisation of a risk created by the
employer by entrusting the work to the employee.
K v Minister of Safety and Security60
The Court had to decide whether the State was vicariously liable for three policemen raping a
young woman. The facts were that Kern and her boyfriend had a fight in a nightclub. He
refused to take her home. She left the nightclub on her own and walked to a nearby petrol
station to phone her mother to collect her. The telephone at the petrol station could only
receive incoming calls. At this point, three policemen in uniform pulled up outside the petrol
station in a police vehicle. One of them, who overheard her conversation with the attendant,
offered her a lift home. She accepted his offer, but he did not take her home. Instead the
policemen took her elsewhere and each of the policemen raped her in turn at knifepoint. She
had to be treated for injuries and severe trauma. She subsequently claimed damages from
the Minister of Safety and Security. The crucial question was whether the three policemen
were acting in the course and scope of their employment when they raped the plaintiff.
In the Supreme Court of Appeal, Scott JA considered this to be a deviation case, posing
the legal issue in the following way:

… whether the deviation was of such a degree that it can be said that in doing what he
or she did the employee was still exercising the functions to which he or she was
appointed or authorised to do or still carrying out some instruction of his or her
employer.

The Supreme Court of Appeal declined to hold the Minister vicariously liable for the wilful or
intentional misconduct of the policemen, even though they were in uniform and on duty
escorting Kern home when they had raped her. In addition, Scott JA reasoned that the Court
could not establish a sufficiently close link between the deviant acts of the policemen in
pursuit of their own selfish interests and the business of the Minister. The Court made this
decision on the basis that the deviant conduct of the policemen was solely self-serving and
was not authorised by the Minister, even though the wrongful acts were committed whilst
working for the Minister.
In the Constitutional Court, O’Regan J expressed the view that the business of the Minister
of Police does create a foreseeable risk that people would trust policemen, especially those
on duty in uniform, and who are in a police vehicle offering assistance to a member of the
public in the early hours of the morning. Members of the public are likely to trust policemen in
these circumstances. This does not mean that the State would be liable in all instances in
which policemen wilfully cause harm to innocent third parties. In general, the harm must be
foreseeable and there must have been a close connection between the acts of the employee
and the risk created by the business of the employer, or the instruction given by the employer
in furtherance of the business interests. O’Regan J held that there was such a close
connection. The policemen had raped the plaintiff whilst on duty. They were patrolling the
area to protect the public from precisely the wrongs they had committed. Taking care of
public safety and security, which includes escorting someone to their home, and preventing
crimes such as rape, is within the ambit of their authorised duties. They were able to fulfill
their self-interests by abusing a power vested in them, driving a police vehicle and in police
uniform. It is reasonable to infer that the plaintiff would not have accepted the lift home had
they been dressed in ordinary civilian clothes and driving an ordinary vehicle. The Court
 could easily have established a close connection to justify imposing vicarious liability on the
Minister.
O’Regan J said that courts should avoid hiding beneath semantic discussions of the
meanings of ‘the course and scope’ and ‘mode of conduct’ requirements, when imposing
vicarious liability. Instead, they should interpret these principles with the spirit and objects of
the Constitution in mind. In other words, courts must take into account ‘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’.61
In Minister of Safety and Security v F 62 the plaintiff, F, found herself stranded late at night,
and an off-duty policeman offered to drive her home, using a police vehicle. Instead, he
drove to a remote spot where he raped her. The important distinction between this case and
the K case is that on this occasion the policeman was off-duty and for this reason the
Supreme Court of Appeal held that this case fails the test for vicarious liability that was
articulated in K. On appeal the Constitutional Court overturned the decision.63 Mogoeng J,
who wrote the majority judgment, found that there was a sufficiently close link between the
policeman’s employment and F’s assault and rape: The police vehicle facilitated the
commission of the rape. F placed her trust in the policeman because he was a police official,
and the State has a constitutional obligation to protect the public against crime.
Consequently, the Minister was held vicariously liable for the harm that F had suffered.
Froneman J, in a separate concurring judgment, agreed with the outcome, but found that
the Minister’s liability was direct rather than vicarious, because the actions of State officials
are in effect the State’s own actions, and the normative considerations for determining
liability may be appropriately assessed as part of the wrongfulness inquiry.
In their minority judgment, Yacoob J and Jafta J applied the same test as the majority, but
concluded that there was not a sufficient link between the delict and the employment of the
policeman concerned: The policeman’s criminal conduct was too far removed in space and
time from his employment to render the use of the police vehicle and the presence of police
dockets in vehicles sufficient to establish vicarious liability, even if due regard was had to the
victim’s vulnerable situation.

PAUSE FOR Are the outcomes in K and F justifiable?

REFLECTION The ‘close connection’ test requires a value judgement on the degree of
closeness, but the Court is not absolved from the need for an open and
structured process of reasoning, with reference to specific factors and
policy considerations. There is a close connection between the rape and
the policemen’s employment in K, because they were on duty, in uniform,
and using a police vehicle. In these circumstances, K placed her trust in
them to take her home safely, and therefore a connection exists between
their employment as policemen at that time and the rape. The facts
relevant to application of the ‘close connection’ test were different in F,
because the policeman was off-duty and not in uniform, and the car was
not marked as a police car. Consider whether the outcome should
nevertheless be the same.

Costa da Oura Restaurant (Pty) Ltd t/a Umdhloti Bush Tavern v Reddy64
The Court had to decide whether a barman acted within the scope of his employment when
he assaulted a patron outside the bar. The reason for the assault was that the patron had
made remarks about the barman’s efficiency. The patron afterwards tipped a second barman
excessively in the presence of the first one. The first barman saw this as provocation,
followed the patron when he left the restaurant, and attacked him.
Consider the following questions:
 • Did the assault occur after the barman had abandoned his duties?
• Was the assault still sufficiently closely linked to the barman’s employment?
• The Court held that the attack was the result of the ‘personal vindictiveness’ of the barman.
How can we explain this in terms of the ‘standard test’ as set out in the Rabie case, which
requires a court to consider both subjective and objective factors? Does the ‘standard test’
imply that personal vindictiveness automatically severs the link between the employee’s acts
and the employer’s business?
• Did the Court give sufficient weight to policy considerations? What are the relevant policy
considerations?

33.9 Vicarious liability and the Protection of Personal

Information Act 4 of 2013
The Protection of Personal Information Act (‘POPI’) was enacted to protect personal information
that individuals would ordinarily wish to keep private. It allows individuals to institute civil actions
for damages against people who unlawfully process this personal information. It also imposes a
form of vicarious liability on an employer for an employee’s unlawful processing of personal
information.65
In circumstances where an employee has unlawfully processed an individual’s personal
information, that individual (or the information regulator, at the request of the individual) may sue
for damages on the basis of common-law vicarious liability, or in terms of the statutory vicarious
liability created by section 99(1) of POPI. If an individual chooses to pursue a statutory action, an
employer can escape liability only in terms of the specific defences created in section 99(2) of the
Act, which reads:
(2) In the event of a breach the responsible party may raise any of the following defences against
an action for damages:
(a) Vis major;
(b) consent of the plaintiff;
(c) fault on the part of the plaintiff;
(d) compliance was not reasonably practicable in the circumstances of the
particular case; or
(e) the Regulator has granted an exemption in terms of section 37.

It would seem that the employer is unable to raise any common-law defences, and this may have
the effect of unreasonably prejudicing an employer who has diligently educated employees
regarding the requirements of POPI, as well as an employer who has:
done everything reasonably and practicable in its power to ensure that its employees comply
with the requirements of the protection of personal information.66

POPI, therefore, has the effect of imposing liability on an employer for an employee’s
contravention, even if the employee wilfully disobeys instructions or exceeds the scope of his or
her employment.

1 For the historical development of vicarious liability in South Africa, and the influence of English law, see Jordaan
‘Employment relations’ in Zimmermann and Visser (Eds) Southern Cross: Civil Law and Common Law in South
Africa (1996) 389 at 397–401. See further Reid and Loubser ‘Strict liability’ in Zimmermann, Visser and Reid (Eds)
Mixed Legal Systems in Comparative Perspective: Property and Obligations in Scotland and South Africa (2004) at
605; McQuoid-Mason ‘Vicarious and strict liability’ LAWSA 2ed (2011) Vol 30 at 289.
2 Boberg ‘Oak Tree or Acorn: Conflicting Approaches to our Law of Delict’ (1966) 83(2) SALJ 150 at 169.
 3 Crown Chickens (Pty) Ltd t/a Rocklands Poultry v Rieck 2007 (2) SA 118 (SCA).
4 Botes v Van Deventer 1966 (3) SA 182 (A).
5 Gifford v Table Bay Dock and Breakwater Management Commission (1874) 4 Buch 96.
6 Botes v Van Deventer 1966 (3) SA 182 (A) at 206.
7 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC).
8 F v Minister of Safety and Security 2012 (1) SA 536 (CC).
9 Smit v Workmen’s Compensation Commissioner 1979 (1) SA 51 (A) at 61.
10 Smit v Workmen’s Compensation Commissioner 1979 (1) SA 51 (A) at 62.
11 Midway Two Engineering & Construction Services v Transnet Bpk 1998 (3) SA 17 (SCA) at 23; Stein v Rising Tide
Productions CC 2002 (5) SA 199 (C).
12 Smit v Workmen’s Compensation Commissioner 1979 (1) SA 51 (A) at 62; Gibbins v Williams, Muller, Wright &
Mostert Ingelyf 1987 (2) SA 82 (T) at 90.
13 Langley Fox Building Partnership (Pty) Ltd v De Valence 1991 (1) SA 1 (A) at 11.
14 See, for example, Esterhuizen v Administrator Tvl 1957 (3) SA 710 (T).
15 Holland v Simenhoff1923 AD 676 at 679; Rodrigues v Alves 1978 (4) SA 834 (A) at 842ff; Lindsay v Stofberg NO
1988 (2) SA 462 (C) at 467.
16 Rodrigues v Alves 1978 (4) SA 834 (A) at 839.
17 Eksteen v Van Schalkwyk 1991 (2) SA 39 (T) at 45ff.
18 Colonial Mutual Life Assurance Society Ltd v Macdonald 1931 AD 412 at 442.
19 Eksteen v Van Schalkwyk 1991 (2) SA 39 (T) at 45.
20 Even if a limited interest of a social nature, such as driving to have a meal together, as in Roman v Pietersen 1990 (3)
SA 350 (C).
21 2001 (3) SA 868 (SCA).
22 Grove v Ellis 1977 (3) SA 388 (C) at 390; Opperman v Opperman 1962 (3) SA 40 (N) at 45; Hamman v South West
African People’s Organisation 1991 (1) SA 127 (SWA) at 139.
23 De Beer v Sergeant 1976 (1) SA 246 (T) at 251; Hamman v South West African People’s Organisation 1991 (1) SA
127 (SWA) at 139.
24 Mkize v Martens 1914 AD 382.
25 1998 (3) SA 17 (SCA). At 23, the Court refers to the English case of Mersey Docks and Harbour Board v Coggins &
Griffith (Liverpool) Ltd [1947] AC 3 (HL) as authoritative in this regard, reflecting the same position as in Scotland.
26 Colonial Mutual Life Assurance Society Ltd v Macdonald 1931 AD 412; Auto Protection Insurance Co Ltd v
Macdonald (Pty) Ltd 1962 (1) SA 793 (A) at 799; Smit v Workmen’s Compensation Commissioner 1979 (1) SA 51
(A) at 61; Langley Fox Building Partnership (Pty) Ltd v De Valence 1991 (1) SA 1 (A).
27 Colonial Mutual Life Assurance Society Ltd v Macdonald 1931 AD 412; Auto Protection Insurance Co Ltd v
Macdonald (Pty) Ltd 1962 (1) SA 793 (A) at 799; Smit v Workmen’s Compensation Commissioner 1979 (1) SA 51
(A) at 61; Stein v Rising Tide Productions CC 2002 (5) SA 199 (C).
28 Colonial Mutual Life Assurance Society Ltd v Macdonald 1931 AD 412 at 433; Dukes v Marthinusen 1937 AD 12 at
17 and 23; Munarin v Peri-Urban Areas Health Board 1965 (1) SA 545 (W) at 549; Smit v Workmen’s Compensation
Commissioner 1979 (1) SA 51 (A) at 61; Stein v Rising Tide Productions CC 2002 (5) SA 199 (C).
29 Goldberg v Durban City Council 1970 (3) SA 325 (N) at 331.
30 2002 (5) SA 199 (C).
31 Langley Fox Building Partnership (Pty) Ltd v De Valence 1991 (1) SA 1 (A) at 11.
32 Langley Fox Building Partnership (Pty) Ltd v De Valence 1991 (1) SA 1 (A) at 13.
33 1991 (1) SA 1 (A) at 12H–J.
34 2009 (1) SA 265 (SCA).
35 Crown Chickens (Pty) Ltd t/a Rocklands Poultry v Rieck 2007 (2) SA 118 (SCA).
36 Costa da Oura Restaurant (Pty) Ltd t/a Umdhloti Bush Tavern v Reddy 2003 (4) SA 34 (SCA) (assault by a restaurant
employee on a customer, outside the establishment); Kasper v Andrè Kemp Boerdery CC 2012 (3) SA 20 (WCC) (an
employee who had been ordered to transport certain weeds to another location chose instead to burn them, causing a
fire that damaged a neighbouring property).
37 1986 (1) SA 117 (A) at 134.
38 See also Feldman (Pty) Ltd v Mall 1945 AD 733 at 774.
39 At 132G–H.
40 At 133D–E.
41 At 133G–H.
42 At 133–134.
43 At 134I–J.
44 2003 (3) SA 83 (SCA).
45 Witham v Minister of Home Affairs 1989 (1) SA 116 (ZH) at 125; Minister van Wet en Orde v Wilson 1992 (3) SA
920 (A) at 927ff; Minister of Law and Order v Ngobo 1992 (4) SA 822 (A) at 828; Costa da Oura Restaurant (Pty)
Ltd t/a Umdhloti Bush Tavern v Reddy 2003 (4) SA 34 (SCA).
46 1997 (1) SA 309 (A).
47 1945 AD 733.
48 1955 (1) SA 202 (A).
49 2003 (3) SA 83 (SCA).
50 Heuston and Buckley (Eds) Salmond and Heuston on the Law of Torts 21 ed (1996) at 443.
51 Feldman (Pty) Ltd v Mall 1945 AD 733 at 756.
52 2001 (1) SA 1214 (SCA).
53 2001 (1) SA 372 (SCA).
54 Minister van Veiligheid en Sekuriteit v Japmoco Bk h/a Status Motors 2002 (5) SA 649 (SCA); Minister van
Veiligheid en Sekuriteit v Phoebus Apollo Aviation Bk 2002 (5) SA 475 (SCA); Giesecke & Devrient Southern Africa
(Pty) Ltd v Minister of Safety and Security 2012 (2) SA 137 (SCA); Von Beneke v Minister of Defence 2012 (5) SA
225 (GNP).
55 Minister of Police v Rabie 1986 (1) SA 117 (A); Minister van Veiligheid en Sekuriteit v Japmoco Bk h/a Status
Motors 2002 (5) SA 649 (SCA); Minister van Veiligheid en Sekuriteit v Phoebus Apollo Aviation Bk 2002 (5) SA 475
(SCA); Masuku v Mdlalose 1998 (1) SA 1 (SCA); K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005
(6) SA 419 (CC).
56 K v Minister of Safety and Security 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC).
57 2006 (4) SA 160 (SCA); 2007 (2) SA 106 (CC) at 110A–B.
58 2007 (1) SA 111 (SCA).
59 See, for example, the Scottish case of Central Motors (Glasgow) Ltd v Cessnock Garage and Motor Co 1925 SLT
563; 1925 SC 796 in which the Court held that the employer had a personal duty to look after a car left in its garage
for safekeeping. Other relevant English cases are Lloyd v Grace, Smith & Co [1912] AC 716; and Morris v CW
Martin & Sons Ltd [1965] 1 QB 716, both referred to in the South African case Absa Bank Ltd v Bond Equipment
(Pretoria) (Pty) Ltd 2001 (1) SA 372 (SCA).
60 2005 (9) BCLR 835 (CC); 2005 (6) SA 419 (CC) para 9.
61 Para 52.
62 2011 (3) SA 487 (SCA).
63 F v Minister of Safety and Security 2012 (1) SA 536 (CC).
64 2003 (4) SA 34 (SCA).
65 Millard and Bascerano ‘Employers’ Statutory Vicarious Liability in Terms of the Protection of Personal Information
Act’ PER/PELJ 2016(19):1 DOI available online at http://dx.doi.org/10.17159/1727-3781/2016/v19i0a555.
66 Supra 30.
Chapter 34

Remedies

34.1 Introduction

34.2 Damages
34.2.1 Purpose of a damages award
34.2.2 Future loss and the once-and-for-all rule
34.2.3 Once-and-for-all rule
34.2.4 Exceptions to the once-and-for-all rule
34.2.4.1 Common-law exceptions
34.2.4.2 Statutory exceptions
34.2.5 Accounting for benefits and collateral sources
34.2.6 General factors that |influence an award
34.2.6.1 Time with reference to which assessment is made
34.2.6.2 Taxation
34.2.6.3 Interest
34.2.6.4 Inflation
34.2.6.5 Currency
34.2.6.6 Contingencies
34.2.7 Lex Aquilia: Assessment and quantification of damages for patrimonial harm
34.2.7.1 Assessment of harm: General principles
34.2.7.2 Quantifying the damages: Damage to property
34.2.7.3 Quantifying the damages: Personal injury
34.2.7.4 Quantifying the damages: Expenses
34.2.7.5 Quantifying the damages: Loss of earnings and earning capacity
34.2.7.6 Quantifying the damages: Basic method for calculating
loss of future income
34.2.7.7 Quantifying the damages: Illegal earnings
34.2.7.8 Quantifying the damages: Loss of support
34.2.7.9 Quantifying the damages: Mitigation of loss
34.2.8 Germanic remedy: Assessing reparation for infringements of physical-mental
integrity
34.2.8.1 Introduction
34.2.8.2 Pain and suffering
34.2.8.3 Loss of amenities of life
 34.2.9 Actio iniuriarum: Determining the amount that would provide appropriate
satisfaction
34.2.9.1 Assault
34.2.9.2 Sexual abuse
34.2.9.3 Deprivation of liberty
34.2.9.4 Dignity, privacy, and identity
34.2.9.5 Defamation

34.3 Retraction and apology

34.4 Interdict

34.5 Conclusion

34.1 Introduction
Once a plaintiff has proved the elements of a delict, the delictual remedies become available to the
plaintiff . Th e general purpose of these remedies is either to compensate for the harm suff ered or
to prevent harm or further harm from ensuing. Th is chapter concentrates on three specifi c
remedies: damages (or compensation), retraction and apology, and interdict.

TERMINOLOGY Remedy
The term ‘remedy’ is often used to describe a particular action. For
example, one finds that the lex Aquilia, the actio iniuriarum, and
the action for pain and suffering (the Germanic remedy) are
referred to as ‘remedies’. We acknowledge the use of this term for
these actions, but in this chapter, we use the term ‘remedy’ to
describe the form of relief that persons can obtain once they have
shown that a delict has been committed, for example, damages,
retraction and apology, or in the case of threatened harm, an
interdict.

34.2 Damages
A plaintiff may recover damages by means of one or more of the following actions:
• The actio legis Aquiliae for compensation for patrimonial loss
• The actio iniuriarum for satisfaction for infringements of personality rights
• The Germanic remedy for pain and suffering for claiming reparation for infringements of
physical-mental integrity
• The condictio furtiva, by means of which a person with an interest in a thing, usually the
owner, seeks satisfaction for having been deprived of its possession
• Th actio de pauperie and the actio de pastu, which aim to compensate pauperies (patrimonial
loss, and pain and suffering caused by animals).

There might also be other possibilities, but these are the actions that we discuss in this book.
 Calculating damages involves a two-step process. First, one assesses the harm by identifying
and establishing the nature of harm that has been suffered, and also its extent. For example, a
plaintiff has had to pay R10 000 medical expenses to date and is likely to incur a further R30 000 in
the future. Once one has assessed the harm, the next step is to quantify the harm. This is the process
in which courts determine the amount of damages, whether they need to make any adjustments to
the amount, to reach a final figure for the award amount. In our example of medical expenses,
inflation might influence the situation and courts also have to consider the fact that the plaintiff, in
receiving the money sooner than it was needed, would earn some interest that strictly speaking is
not due to him or her. Courts also adjust the amount for contingencies, that is, the possibility that
the plaintiff might not incur the claimed expenses in the future. The final amount that a court
awards for future expenses in this example might therefore be adjusted, say, to R20 000, to allow
for these extraneous factors. The demarcation between these steps is not always clear, but they are
nonetheless followed in some way or another in all cases.

TERMINOLOGY Harm and damages

As already discussed, harm is an element of a delict. Harm is also
sometimes referred to as ‘loss’ or ‘damage’. However, the term
‘damage’ must be distinguished from the term ‘damages’. The
former is an element of delictual liability; the latter is a remedy
available to plaintiffs once they have established liability. We use
the term ‘damages’ for the amount of money that courts award to
the plaintiff to compensate or satisfy the harm suffered as a result
of delictual conduct.

General and special damages

In the law reports, the headings called ‘damages’ often include
items such as ‘special’ and ‘general’ damages. Courts do not
always use these terms consistently. Furthermore, courts often
use the terms ‘general’ and ‘special’ damages to refer to harm or
loss, instead of the amount to be awarded.
Special damages are damages that are specially pleaded and
proven, in other words, those items of loss that one can specify,
list and quantify. They may cover all forms of patrimonial loss, both
past and future, but as a rule, cover expenses or earnings prior to
the date of trial. Courts award general damages in respect of both
patrimonial and non-patrimonial loss for harm that arises from the
conduct as a natural and probable consequence, and that has not
materialised at the date of trial into a certain, quantifiable amount.
In layperson’s terms, courts cannot specify or detail such loss, and
so they usually have to estimate a lump sum. Included in this
category are damages for loss of future earnings capacity and
future medical expenses, as well as compensation for non-
patrimonial loss, such as pain and suffering, emotional shock, loss
of amenities of life, and compensation for infringement of
personality rights (solatium) under the actio iniuriarum.

The assessment of general damages is a matter within the discretion of the court of first instance
and depends on the unique circumstances of each case. An appeal court is not likely to interfere
with an award by the trial court unless there is evidence of irregularity or misdirection.
Minister of Safety and Security (Now Minister of Police) v Scott1
 Scott, a professional hunter, and his company Scottco sued the Minister for damages arising
from his alleged unlawful arrest and detention. Scott had been arrested after he and his
friends assaulted someone outside a pub and as a result he had been unable to take an
American group for a planned hunting trip at his ranch. His failure to conduct the hunting trip
caused an American magazine to terminate its agreements with Scott and his company.
Essentially, the magazine would no longer run Scottco’s advertisements and would also no
longer bring its own clients to Scott’s ranch. The High Court awarded Scott general damages
amounting to R75 000 in respect of the unlawful arrest and detention and R577 610 to
Scottco for wasted advertisement costs and loss of contractual income and profits. The
Minister appealed against both awards.
On appeal Scottco’s damages were set aside on the basis that there had been no
intentional interference in the contractual relationship. In determining whether the award to
Scott was appropriate, the Court emphasised that an appeal court is generally reluctant to
interfere with the award of the trial court but will do so where the appeal court is of the
opinion that no sound basis exists for the initial award or where there is a striking disparity
between that award and the award which the appeal court considers ought to have been
made.2 After considering similar cases, the Court found that the High Court’s award was
grossly excessive and instead awarded R30 000 for general damages.

34.2.1 Purpose of a damages award

The primary purpose of an award of damages is to compensate an injured party. In the case of
damages claimed under the actio legis Aquilia, a court will, as far as possible, put the injured party
in the position that he or she would have been in had the delict not been committed. This is also
referred to as the ‘negative interest’ or sum-formula. Minister of Defence v Jackson 3 clearly
illustrates this principle. The Court held that in awarding damages for past and future lost earnings,
it had to make an allowance for the fact that the plaintiff would as a result of his accident not have
to pay tax. The Court deducted from his claim the amount of taxes he would have paid to ensure
that he was not over-compensated. In the case of the actio iniuriarum, the damages award serves to
provide satisfaction (or genoegdoening) insofar as money is able to,4 although there is still a view
that the actio iniuriarum has retained an element of punishment. The purpose of the Germanic
action is to increase the plaintiff’s feeling of happiness. Neethling and Potgieter 5 call it ‘imperfect
compensation’, while Van der Walt and Midgley 6 use ‘reparation’. What is clear is that it is not
possible to measure pain and suffering in monetary terms, and the amount of damages paid does
not compensate in the sense of restoring the pre-delict situation.
Our law does not recognise the concept of punitive or exemplary damages, which is well
known in countries such as the United States of America and England. The purpose of these
damages is to punish a defendant. In these countries a plaintiff could, in addition to compensation
for loss, also claim an additional amount of damages which is intended to penalise the defendant
for the unacceptable behaviour. Views differ on whether our law should recognise that some
delictual damages serve a punitive function.

COUNTER Punitive damages

POINT In the United States of America, it is not unusual for plaintiffs to receive
more than double the amount of their damages in punitive damages,
even in instances where plaintiffs claim compensation for patrimonial
loss. A well-known case is that of Liebeck v McDonald’s Restaurants 7 in
which a jury awarded $200 000 in compensatory damages and $2,7
million in punitive damages to an elderly woman who had suffered third-
degree burns when hot coffee spilt over her legs at a drive-through
restaurant. Although the amounts were eventually reduced, the punitive
damages award was still three times the amount awarded for
 compensatory damages. It is clear that there is no relation between the
extent of the loss and the amount of punitive damages.
English law distinguishes between aggravated damages and
exemplary or punitive damages. Courts award exemplary or punitive
damages in instances where the conduct of the defendant was
outrageous, as in Kuddus (AP) v Chief Constable of Leicestershire
Constabulary.8 On the other hand, courts award aggravated damages to
victims as compensation for mental distress caused by injuries that result
from tortuous conduct. Exemplary and aggravated damages were part of
early South African law as a result of the English influence. However,
apart from what was said in Fose v Minister of Safety and Security 9 (see
below), our legal system does not recognise these damages at present.
The actio iniuriarum has been described as an actio vindictam
spirans. Historically this action was purely punitive, but over the centuries
it lost a great deal of its punitive nature. Nonetheless, there is a view that
the actio is not purely compensatory. Visser and Potgieter,10 for example,
describe the action as ‘maintain(ing) a position somewhere between
compensation and punishment’.
In contrast, Van der Walt and Midgley suggest that even though the
actio iniuriarum’s original purpose was to penalise a wrongdoer, and
even though there is case law that indicates that the actio serves a
punitive purpose, ‘this function is out of place in the modern law of
delict’.11 They argue that instead:

where defendants behave maliciously, or where other

aggravating circumstances are present, a larger solatium –
‘aggravated damages’ – is required to assuage the plaintiff’s
feelings. The size of the award may end up being the same, but
the focus is properly on the plaintiff, not the defendant.12

Fose v Minister of Safety and Security13

The plaintiff had been arrested, detained, and subsequently assaulted by members of the
South African Police Service. He thereafter claimed delictual damages from the Minister and,
in addition, punitive damages as relief for the violations of his fundamental rights. While his
claims for delictual damages were sustained, the Constitutional Court (per Ackermann J)
confirmed that punitive damages were not part of South African law:14

… I have come to the conclusion that we ought not, in the present case, to hold that
there is any place for punitive constitutional damages. I can see no reason at all for
perpetuating an historical anomaly which fails to observe the distinctive functions of
the civil and the criminal law and which sanctions the imposition of a penalty without
any of the safeguards afforded in a criminal prosecution.

I agree with the criticisms of punitive constitutional damages … . Nothing has been
produced or referred to which leads me to conclude that the idea that punitive
damages against the government will serve as a significant deterrent against individual
or systemic repetition of the infringement in question is anything but an illusion.
Nothing in our own recent history, where substantial awards for death and brutality in
detention were awarded or agreed to, suggests that this had any preventative effect.
To make nominal punitive awards will, if anything, trivialise the right involved.

For awards to have any conceivable deterrent effect against the government they will
have to be very substantial and the more substantial they are the greater the anomaly
 that a single plaintiff receives a windfall of such magnitude. And if more than one
person has been assaulted in a particular police station, or if there has been a pattern
of assaults, it is difficult to see on what principle, which did not offend against equality,
any similarly placed victim could be denied comparable punitive damages. This would
be the case even if, at the time the award is made, the individuals responsible for the
assaults had been dismissed from the police force or other effective remedial steps
taken.

In a country where there is a great demand generally on scarce resources, where the
government has various constitutionally prescribed commitments which have
substantial economic implications and where there are ‘multifarious demands on the
public purse and the machinery of government that flow from the urgent need for
economic and social reform’, it seems to me to be inappropriate to use these scarce
resources to pay punitive constitutional damages to plaintiffs who are already fully
compensated for the injuries done to them with no real assurance that such payment
will have any deterrent or preventative effect. It would seem that funds of this nature
could be better employed in structural and systemic ways to eliminate or substantially
reduce the causes of infringement.

• The Court specifically mentions ‘constitutional punitive damages’ and not ‘delictual punitive
damages’. Does this make a difference?
• What is the ‘historical anomaly’ that is referred to?
• What are the criticisms that the Ackermann J refers to?15
• The Court canvasses another suggested purpose of delictual damages: deterrence. What is
the Court’s view in this regard? Does it accept or reject deterrence as an underlying
purpose?
• The Court also mentions a social policy reason for not recognising punitive damages in this
instance. Does this policy factor apply to all claims under the actio iniuriarum?
• Earlier in the judgment,16 the Court noted that an award of aggravated damages is still
basically compensatory, but the particular circumstances of the case justify a substantial
award. Would full recognition of aggravated damages not resolve the current debate
regarding the actio’s punitive function?

The condictio furtiva is a damages remedy available to the owner of a thing or someone with an
interest in a thing against a thief and persons who obtained the thing from the thief. An essential
requirement is that the object in question must have been stolen or used without permission; in
other words, the remedy lies in both instances of theft or furtum usus (the appropriation of the use
of another’s property without the intention to appropriate the thing permanently).17 The rationale for
this delictual remedy is that risk passes to the person who deprives the owner or possessor of the
property and its purpose is to compensate the owner or possessor for the loss of the property in
those circumstances where it cannot be restored through rei vindicatio.18
Chetty v Italtile Ceramics Ltd19
Chetty entered into a joint venture and franchise agreements with Italtile Ceramics, and
became manager of a warehouse and retail store. During the course of his employ Chetty
engaged in certain practices that resulted in Italtile losing stock. He had sold stock on credit
instead of in cash to benefit certain customers who would pay at the end of the month; and he
rolled stock over (taking the stock off the computer system at the beginning of the month only
to reverse that entry at the end of the month) with the intention to disguise losses that were
being suffered. These actions were fraudulent. On discovering these practices Italtile
terminated the agreements and sought to recover damages for missing stock on the basis of
the condictio furtiva.
The High Court found that the Chetty’s practices of a delivery book (credit) system, false
writeoffs and reversals of missing stock resulted in patrimonial loss to Italtile. The SCA
 upheld the appeal on the ground that legal causation was absent. In the course of the
judgment the Court pointed out that:
• Patrimonial loss was not the issue and that liability could arise on account of the furtum usus.
• The intention to appropriate the thing permanently, as in the case of criminal theft, is not a
requirement of the condictio where furtum usus is concerned.
• In this instance, the conduct complained of did not constitute the use of Italtile’s property.
What Chetty did was to post false entries to the accounts to mislead Italtile. That could well
have amounted to fraud, but it was not use of the stock.
• Italtile’s loss did not arise directly from the use of the goods, but from Chetty’s failure to take
steps to collect the outstanding debts after termination of the agreement with the appellant.

34.2.2 Future loss and the once-and-for-all rule

Harm also includes prospective or future loss, which is harm that may arise after the delict and even
after the date of the trial (when courts assess the harm and quantify the damages). At the end of a
trial, courts know whether harm is likely to arise in the future. Future harm may consist of future
patrimonial harm or future pain and suffering. Future patrimonial loss includes loss of future
earnings (also referred to as loss of earning capacity), future medical expenses, future loss of
support, and loss of profit.
When someone is seriously injured in an accident and is completely unable to work, this
person’s loss includes not only the medical and hospital costs, and the earnings already lost, but
also future earnings lost and future medical costs. At the date of the trial, the plaintiff has not
suffered these future losses, but it is clear that they will arise. Assessing such harm involves a
certain degree of speculation on the part of the court, because at the time of the trial, the extent of
this harm, as well as other factors that could influence it in the future, are unknown. Courts have to
engage in this speculative exercise because of the once-and-for-all rule, which prevents a plaintiff
from claiming damages before the harm actually materialises and the fact that delictual claims
become prescribed three years after the date of the delict.

34.2.3 Once-and-for-all rule

The once-and-for-all rule originated in English law. According to this rule, a person has to claim
compensation for all loss, past and future, and patrimonial and non-patrimonial, in a single action,
if the harm is based on a single cause of action. This means that the plaintiff will claim damages for
harm that has not yet materialised, but that will in all probability arise in the future. The effect of
this rule is that courts award damages in a lump sum, even though some of the harm has not yet
arisen, for example, a plaintiff who suffers severe injuries as a result of an accident, and will incur
ongoing medical expenses in the future after the date of the trial. A court will award compensation
for that harm at the trial, even though the harm (the costs for the future medical treatment) will
arise in the future.
A cause of action and an action are not the same. An action is a process for claiming a remedy,
such as the actio legis Aquiliae or the actio iniuriarum. A cause of action is the set of facts upon
which an action is based. There are two possible approaches to the concept ‘cause of action’:

• The ‘single cause’ approach, in terms of which every harm-causing event constitutes one cause
of action irrespective of the nature of the claim
• The facta probanda approach, in terms of which a cause of action will arise as soon as all the
material requirements for a particular type of claim are present.
Evins v Shield Insurance Co Ltd20
The plaintiff was injured and her husband killed in a motor vehicle accident on 30 March
1972. The plaintiff claimed compensation from the defendant, an insurance company, in
terms of the Compulsory Motor Vehicle Insurance Act 56 of 1972, the predecessor of the
current Road Accident Fund Act 56 of 1996. In May 1973, the plaintiff delivered the claim for
compensation in the prescribed form to the defendant. The form was completed properly in
respect of the claim for compensation for bodily injuries, but not properly in respect of the
claim for loss of support. In August 1973, the plaintiff served summons on the defendant.
Over three years later, in January 1977, the plaintiff issued a second summons in which only
a claim for loss of support was pursued. The plaintiff next obtained an unopposed order to
amend the first summons by deleting all reference to the claim for loss of support. Eventually,
the defendant filed a special plea alleging that the plaintiff’s claim for loss of support had
become prescribed in that it arose on 30 March 1972.
At issue was whether the claims for bodily injuries and loss of support were based on a
single cause of action or two different causes of action. If there were two causes of action,
the defendant’s special plea that the claim for loss of support had prescribed would be
correct. To resolve the issue, the Court first looked at the meaning of the term ‘cause of
action’. It explained the facta probanda approach and applied it to the facts of the case:21

‘The proper legal meaning of the expression “cause of action’’ is the entire set of facts
which gives rise to an enforceable claim and includes every fact which is material to be
proved to entitle a plaintiff to succeed in his claim. It includes all that a plaintiff must set
out in his declaration in order to disclose a cause of action. Such cause of action does
not ‘‘arise’’ or ‘’accrue’’ until the occurrence of the last of such facts and consequently
the last of such facts is sometimes loosely spoken of as the cause of action … .’

In the case of an Aquilian action for damages for bodily injury (and here I use the term
Aquilian in an extended sense to include the solatium awarded for pain and suffering, loss of
amenities of life, etc, which is sui generis and strictly does not fall under the umbrella of the
actio legis Aquiliae: …, the basic ingredients of the plaintiff’s cause of action are (a) a
wrongful act by the defendant causing bodily injury, (b) accompanied by fault, in the sense of
culpa or dolus, on the part of the defendant, and (c) damnum, i.e. loss to plaintiff’s patrimony,
caused by the bodily injury. The material facts which must be proved in order to enable the
plaintiff to sue (or facta probanda) would relate to these three basic ingredients and upon the
concurrence of these facts the cause of action arises. In the usual case of bodily injury arising
from a motor accident this concurrence would take place at the time of the accident. On the
other hand, in the case of an action for damages for loss of support, the basic ingredients of
the plaintiff’s cause of action would be (a) a wrongful act by the defendant causing the death
of the deceased, (b) concomitant culpa (or dolus) on the part of the defendant, (c) a legal
right to be supported by the deceased, vested in the plaintiff prior to the death of the
deceased, and (d) damnum, in the sense of a real deprivation of anticipated support. The
facta probanda would relate to these matters and no cause of action would arise until they
had all occurred.

From this analysis it is evident that, although there is a measure of overlapping, the facta
probanda in a bodily injury claim differs substantially from the facta probanda in a claim for
loss of support. Proof of bodily injury to the plaintiff is basic to the one; proof of death of the
breadwinner is basic to the other. Proof of a right to support and the real expectation that, but
for the breadwinner’s death, such support would have been forthcoming is basic to the one,
irrelevant to the other. It is evident, too, that even where both claims flow from the same
accident, the cause of action in each case may arise at a different time. As I have said, the
cause of action in respect of bodily injury will normally arise at the time of the accident, i.e.
when the bodily injury and the consequent damnum are inflicted; in the case of the cause of
action for loss of support, this will arise only upon the death of the deceased, which may
occur some considerable time after the accident. Until such death there is, of course, no
wrongful act qua the plaintiff; only a wrongful act qua the person who is later to become the
deceased.
The Court in this instance found that although the claim for damages for bodily injuries and loss of
support were caused by the same harm-causing event, they nevertheless were based on different
causes of action, because the essential facts that the plaintiff had to prove in each instance were
different.

PAUSE FOR Single cause of action

REFLECTION The plaintiff in Evins v Shield Insurance Co Ltd22 did not succeed in
proving that there was a single cause of action and so the claim for loss
of support had prescribed. If the plaintiff had brought her claim within
three years after the date of the delict, her claim would have succeeded.
However, would the plaintiff have been successful if there had been a
single cause of action? If the claims for loss of support and bodily injuries
constituted the same cause of action, would the once-and-for-all rule then
have prevented the plaintiff from instituting action?

The once-and-for-all rule has both advantages and disadvantages. Its main advantage is the fact that
it brings the litigation to an end and prevents the same case from being taken to court again. Its
main disadvantage lies in the fact that the plaintiff has to claim in one lawsuit for both past and
future loss. With future loss, courts have to ascertain how much money to award to the plaintiff for
loss that may or may not arise in the future. This invariably results in some speculation and could
result in either under- or over-compensation.23

PAUSE FOR Assessing loss in absence of the once-and-for-all rule

REFLECTION An example of assessing loss in the absence of a rule such as the once-
and-for-all rule is contained in Article 6:105 of the Dutch Civil Code. This
article creates the possibility for Dutch courts, in those cases where harm
has not yet materialised, either to assess the harm in advance, or to
postpone the assessment partially or completely until such time as the
harm has ensued. Where courts make the assessment prior to the actual
harm ensuing, a judge also has a choice to order payment of damages
either by a lump sum, or by periodic payments. In the latter instance, a
judge may require the defendant to provide security for the future
payments. Such a provision could solve many of the problems that are
currently experienced with the speculation around future losses.
However, the only way in which the law that relates to the once-and-for-
all rule could change would be by statutory intervention.

34.2.4 Exceptions to the once-and-for-all rule

There are both common-law and statutory exceptions to the once-and-for-all rule.

34.2.4.1 Common-law exceptions

• In the case of nuisance, a plaintiff may institute a claim every time the nuisance causes harm,
because one cannot anticipate future harm that may arise out of the nuisance. Examples of
conduct that amount to nuisance include an unpleasant smell, noise, leaves from a neighbour’s
tree falling into a person’s garden, and smoke. In De Charmoy v Day Star Hatchery (Pty) Ltd 24
 the defendant company bred and sold chickens. The plaintiff stayed on the neighbouring
property and sought relief based on the noise and foul odours that emanated from the chicken
hatchery.
• In subsidence cases, where harm is caused by wrongful excavation, a cause of action arises
every time the harm arises. South African law recognises the right of the landowner to lateral
support and if a person causes subsidence on a neighbour’s land as a result of excavations on
the property, he or she could be liable for damages. In D and D Deliveries (Pty) Ltd Pinetown
Borough 25 the defendant caused substantial damage to the plaintiff’s property as a result of
excavations on its property. The plaintiff claimed damages based on the subsidence and then
later amended its particulars of claim and inserted an additional claim in terms of the actio
legis Aquiliae. The question before the Court was whether these constituted separate causes of
action. The Court held that the facta probanda for the two claims were substantially different
and that the second claim had accordingly become prescribed. With regard to claims based on
subsidence the Court held: 26
In subsidence cases it is unnecessary to prove an unlawful act or negligence;
the cause of action is simply damage following upon deprivation of lateral
support. The action lies only against the owner of the adjoining property, and
each successive subsidence gives rise to a fresh cause of action.

• In the case of a continuing wrong there is a cause of action every time harm arises. In
Slomowitz v Vereeniging Town Council 27 the plaintiff claimed damages from the defendant for
harm suffered as a result of a road closure. The defendant submitted that the claim was barred
by a provincial ordinance which stated that claims against local authorities had to be brought
within six months. The defendant’s argument rested on the assumption that the road closure
constituted a single act. The Court held that the fact that the road was kept closed constituted a
continuing wrong and thus gave rise to several causes of actions.

34.2.4.2 Statutory exceptions

One finds statutory exceptions in the Road Accident Fund Act:
• Section 17(1), read together with section 21, creates Road Accident Fund liability for harm
resulting from bodily injury or death ‘caused by or arising from the driving of a motor
vehicle’. The Fund is, however, not liable for damage to property, and so the victim may, on
the same set of facta probanda, sue the driver separately using the actio legis Aquiliae in
respect of damage to the motor vehicle.
• Section 17(4)(a) of the Act provides that in respect of certain expenses the Fund may
compensate the plaintiff only once the expenses have been incurred. This is an exception in
that it deviates from the lump-sum payment rule.
• Section 17(4)(b) of the Act makes provision for payment of loss of future income or support in
instalments.

34.2.5 Accounting for benefits and collateral sources

It often happens that the plaintiff, as a result of the delict, not only suffers harm, but also receives
some benefit. This benefit may assume different forms, such as insurance payouts and pension
payouts. For example, Tom is injured in an accident and as a result he suffers brain damage. He can
no longer do the work he did before. His employer feels sorry for him and allows him to do odd
jobs around the office. He pays him a small sum of money for this work. Tom claims damages for
loss of future earnings. In a case like this, the question is whether the court should deduct from the
total damages claim the money paid to Tom by the employer.
Benefits that courts do not deduct from the damages claim are regarded as collateral sources
and therefore res inter alios acta. This phrase literally means that something that happens between
two parties does not concern anyone else. Some of the benefits that a plaintiff may receive from
other sources after having suffered loss will therefore be ‘nobody else’s business’ and a court will
not consider them when deciding the extent of the loss and eventually the size of the damages
award. Where benefits are deducted, one normally refers to this as ‘accounting for benefits’.
Whether a benefit is deducted usually depends on the source and the nature of the benefit. In
Santam Versekeringsmaatskappy Bpk v Byleveldt,28 Byleveldt was seriously injured and suffered
brain damage that left him unable to work. His wife asked his employer to give him odd jobs to do
and to pay him something. Out of compassion the employer agreed to do so. The defendant claimed
that the amount in question should be deducted from the damages claim.
The problem in these instances is if a person, as a result of a harm-causing event, not only
suffers harm, but also receives a benefit, then that person would be enriched by the event if the
amount of the benefit is not subtracted from the amount of damages. In our example, Byleveldt
received some money from his employer, even though it was more of a gift than a wage. We
already know that the purpose of an award of damages is to compensate, not to enrich the plaintiff.
In this instance, the question is whether a court should deduct that money from the overall award so
that Byleveldt does not receive double compensation. On the other hand, if a court does deduct the
benefits that the plaintiff receives as a result of the accident from the amount of damages, the
wrongdoer gains from someone else’s generosity and is partly absolved from liability. In many
instances it would be unfair for the defendant to benefit in this way. Courts have rejected this latter
argument on the basis that the purpose of damages is to compensate the plaintiff, not to punish the
defendant. In the Byleveldt case, however, the Court held that an amount given to an injured party
as a result of compassion or as a gift should not be deducted.
Standard General Insurance Co Ltd v Dugmore NO29
R, a 43-year-old man, was badly injured in a motor vehicle accident. As a result of his
injuries, he became totally incapacitated and unemployable. The plaintiff, acting as R’s
curator ad litem, claimed damages from the defendant in terms of the Multilateral Motor
Vehicle Accidents Fund Act 93 of 1989, the immediate predecessor of the Road Accident
Fund Act 56 of 1996. The plaintiff claimed damages under a variety of heads. All the claims
were settled except for the claims for past and future earnings. The problem arose because R
had received the following benefits:
• Monthly pension payments from a disability pension to which he was entitled in terms of his
employer’s pension fund
• Benefits from a group accident insurance policy.

The question before the Court was whether these amounts were res inter alios acta, or
whether they had to deduct them from the total amount of damages claimed for past and
future loss of earnings. The Court a quo held that they had to deduct the pension from the
damages, because they were a direct consequence of his contract of employment, but the
insurance policy benefits were res inter alios acta and so need not be deducted. Both parties
appealed, and both appeals were dismissed.

The Court explained the collateral source rule as follows:30

The object of awarding Aquilian damages is to place the plaintiff in the position in
which he would have been had the delict not been committed, thereby redressing the
diminution of his patrimony caused by the defendant’s delict … .
 In calculating the patrimonial position in which the plaintiff would have been had the
delict not been committed, and comparing it with his present position, one has to take
into account not only the detrimental sequelae of the delict, but also the advantageous
consequences thereof: after all, one needs to compare the total patrimonial position of
the plaintiff at present (i.e. post delicto) with the corresponding position ante delicto.

Developed to its logical conclusion, this principle would require the plaintiff to disclose
and deduct from his claim each and every benefit received or receivable as a
consequence of the delict. But it seems evident that the rule cannot be pursued to
such logical conclusion: it is manifestly unjust that the plaintiff should deduct from his
claim, and the defendant profit by, for example gratuitous benefits received by the
plaintiff.

The question thus is one of demarcation only: which benefits are deductible from the
plaintiff’s claim?

The Court attempted to play down the ‘demarcation’, in other words, trying to find a formula or
principle to explain which benefits should be deducted:31

It now seems to be generally accepted that there is no single test to determine which
benefits are collateral and which are deductible. Both in our country … and in England
… it is acknowledged that policy considerations of fairness ultimately play a
determinative role.

In the end, the only ‘principle’ that the Court identified was that of fairness, reasonableness
and justice. In other words, the decision is one of policy, which serves to determine whether
the third-party payment is legally relevant to the plaintiff’s claim.

Examples of benefits that courts do not deduct from the amount of damages include: 32
• Benefits received from insurance contracts: this applies to both indemnity insurance (for
example, accident insurance) and non-indemnity insurance (such as life insurance)
• Benefits received from a medical aid, where the medical aid has the discretion as to whether to
pay the benefit
• Paid sick leave, where the plaintiff’s employer has a discretion as to whether to grant sick
leave
• Ex gratia payments, such as that given in the Byleveldt case
• Pension payments, where the payment of the pension is discretionary
• An award that has been received as a solatium.

Benefits that courts do deduct from the damages claim include: 33

• Medical aid payments received pursuant to a contractual or statutory right
• Sick leave granted pursuant to an employment contract or a statutory right
• Pension paid out pursuant to a contractual or statutory right
• The benefit of free medical care received in a provincial hospital
• Savings on income tax as a result of lost income
• Social security grants paid out to a single mother in respect of her children are deducted from
damages for loss of support.34

PAUSE FOR Accounting for benefits

REFLECTION There is no general principle that our courts can use to decide which
benefits they should account for and which benefits they should regard
as res inter alios acta. South African courts deal with each benefit on a
 case-by-case basis, based on policy considerations. In the Netherlands,
Article 100 of the Civil Code provides that courts must consider a benefit
when assessing damages, insofar as it is reasonable. This principle is so
general that it requires interpretation by the courts. The net effect is that
courts have to decide in each case whether it would be reasonable to
deduct a particular benefit. This has given rise to a body of law that is
almost as casuistic as that in South Africa.35

34.2.6 General factors that influence an award

When courts decide on an amount of satisfaction, they consider certain factors. The factors
described in the following sections do not constitute a closed list and in some instances they may
overlap.

34.2.6.1 Time with reference to which assessment is made

Generally, the date of the delict serves as a point of reference for assessing harm, including future
harm, and for making the comparison between the plaintiff’s pre-delict and postdelict positions,
irrespective of whether the hypothetical or actual method is used.36 However, courts may also
consider events that take place between the date of the delict and the date of the trial. The date of
the trial, therefore, becomes relevant in cases of future losses where courts have to capitalise and
discount the amount of damages to present value.
General Accident Insurance Co SA Ltd v Summers, Southern Versekeringsassosiasie
Bpk v Carstens NO, General Accident Insurance Co SA Ltd v Nhlumayo37
Three appeals were heard in one sitting. The decisions all concerned the date to which
damages had to be discounted to present value. The harm in each instance was prospective
loss. In each of these cases, the actuary who provided expert evidence for the defendant
regarded the date of the delict as the proper reference point for discounting the damages for
loss of future earnings. The plaintiffs’ actuaries suggested using a split method in terms of
which claims for damages were to be discounted to the date of delict, and others, for
example, claims for loss of support and future earnings, were to be discounted to the date of
trial. Discounting damages to the date of the delict would result in a smaller amount of
damages than the split method advanced by the plaintiff.
The Court said:38

It may sound logical to say that, because the cause of action of a victim of delictual
conduct originated on the day of the delict, his compensation has to be determined
with reference to that day. I am of the opinion that this is not necessarily, or in all
circumstances, a correct statement, and especially it should not simply be accepted
that because harm to a thing such as a motor vehicle has to be determined with
reference to the date of the delict, loss of earning capacity or support should also of
necessity be determined in this manner. The two cases are not the same.

When a thing such as a car is destroyed or damaged, the process of destruction or

harm is usually completed immediately. It is easy therefore to understand why in the
case of such harm the owner’s harm is to be assessed with reference to the date of
the delict … .

Where someone is injured and his earning capacity is destroyed or reduced as a result
the position will not be the same. Even though someone’s earning capacity is regarded
as an asset in his estate, the harm caused as a result of the destruction thereof will not
 be completed on the date of the delict. One has to do therefore with harm which
continues after the date of the delict, that is, into the future. The same considerations
apply where harm arises as a result of the death of a breadwinner.

The Court held that although the point of departure regarding the date to which the harm is
assessed will be the date of the delict, there may be instances in which the reference has to
be made to the date of the trial.

34.2.6.2 Taxation
Courts will consider the fact that the plaintiff could, because of inability to work, save on paying
income tax, and they will deduct this saving from the amount of damages.
Minister of Defence v Jackson39
The plaintiff was severely injured in a collision between his vehicle and a military vehicle. He
claimed damages under a variety of heads: for expenses, pain and suffering, and loss of
earnings, both past and future. One of the issues was whether, and to what extent any
savings on income tax affected the damages calculation. (The tax legislation at the time
exempted damages awards from income tax.) The Court referred to the English decision of
British Transport Commissioner v Gourley,40 in which the Court held they should deduct
savings on income tax from the amount of damages when quantifying both past and future
earnings, provided of course that these amounts were exempt from tax. In this regard, there
should be no difference between past earnings (which the Court in Gourley classified as
special damages) and future earnings (which it classified as general damages), because
often whether something was past loss or future loss depends on coincidence, namely the
date of the trial. The Court, therefore, would deduct savings from income tax from damages
for past and future earnings. Underlying this rule was the principle that the purpose for an
award of damages is not to over-compensate the plaintiff or to punish the defendant; it is
merely to put the plaintiff in the position that he or she would have been had the delict not
been committed.

34.2.6.3 Interest
Interest rates are relevant when calculating future losses. When a court makes an award for future
loss of earnings (loss of earning capacity), they discount that amount to present value (calculate the
amount which, if invested now, would grow to the amount of the future loss at the time the loss is
expected to occur). In order to discount an amount to present value, a court has to consider interest
rates. The interest rate that a court chooses depends on expert evidence, often by using actuaries.
Courts also use actuarial or discount tables, which set out the present values of amounts with
reference to different interest rates and over different periods.

34.2.6.4 Inflation
Inflation causes currency to lose its purchasing power. The result is that items that one purchases
become more expensive as time goes by. This means that one can buy less today with a certain
amount of money than one could three years ago, or that a damages amount awarded today will be
worth less in a few years’ time. Assume, for example, that a person who is injured incurs R30 000
of medical expenses and the case is only heard two years after the accident. If the court awards that
person R30 000, this money is worth less than what it was worth two years ago. The question is
whether the Court should consider and make good that loss in value. In the case of future expenses
and loss of earnings, should a court consider inflation and increase the amount accordingly?
 SA Eagle Insurance Co Ltd v Hartley 41
The plaintiff was involved in a motor vehicle accident. Inter alia, he claimed for loss of past
earnings. The Court a quo awarded a sum of money adjusted to ‘compensate the respondent
for the loss of purchasing power of money since the dates upon which his past losses of
earnings had been incurred’. This portion of the damages award was the basis for the appeal.
The Court a quo had applied the Everson adjustment in terms of which inflation to take into
consideration when assessing past losses. Previously, in Everson v Allianz Insurance Ltd,42
the Court considered the diminished purchasing power of the money when assessing past
losses.
However, in the present case, the Appellate Division held that inflation should not be
considered with regard to past losses. The Court based its decision on General Accident
Insurance Co SA Ltd v Summers 43 in which the Court held that it should assess harm with
reference to the date of the delict. With future losses, courts have to discount these to
present value, because the plaintiff receives the money before the future loss materialises
and so can invest the money and benefit from it sooner than would have been possible in the
ordinary course of events. In the case of past losses, courts do not discount the amount, and
the plaintiff receives the full amount. Applying the Everson principle would entail that the
amount of damages is altered according to when the plaintiff claims and receives
compensation:

This result seems to me to be in conflict with the principle of nominalism of currency

which underlies all aspects of South African law, including the law of obligations. Its
essence, in the field of obligations, is that a debt sounding in money has to be paid in
terms of its nominal value irrespective of any fluctuations in the purchasing power of
currency.

Therefore, based on the principle of nominalism, the Court found that it should make no
adjustment for inflation in respect of past losses.

In cases of future loss, such as loss of future income, loss of future support, and future medical
expenses, courts make provision for the depreciation of the value of money that the plaintiff would
require in the future. As a general rule, courts adopt a conservative approach in this regard.
Courts also adjust awards for non-patrimonial loss, often based on past awards, to factor in the
inflation rate.
Jonosky v Road Accident Fund44
Jonosky instituted proceedings against the Road Accident Fund (RAF) for recovery of
compensation arising from a motor vehicle collision. All issues were settled save the amount
of compensation for loss of income. At issue was the correct interpretation and application of
section 17(4)(c) and (4A)(b) of the Road Accident Fund Act:
• Section 17(4)(c) of the Road Accident Fund Act provides that where a claim for
compensation includes a claim for loss of income the annual loss, irrespective of the actual
loss, shall be proportionately calculated to an amount not exceeding R160000 per year in the
case of a claim for loss of income and in respect of each deceased breadwinner in the case
of a claim for loss of support.
• Section 17(4A)(b) on the other hand, provides that the amount, as adjusted quarterly in order
to counter the effect of inflation, shall be the amount set out in the last notice issued prior to
the date on which the cause of action arose. The Road Accident Fund is required, by notice
in the Government Gazette, to adjust the amounts quarterly in order to counter the effect of
inflation.

Jonosky argued that the amount awarded in respect of both past and future loss of income
had to be adjusted to counter the effect of inflation. The RAF argued that no adjustment
should be made for future loss of income as the future rate of inflation was subject to
fluctuations and, as such, inexact, or at best, speculative.
 The Court held that in calculating future loss of earnings beyond the date on which such
calculation was made, an actuary was duty-bound to incorporate a projected future inflation
rate on an annual basis. As a result, the Court awarded the higher amount that Jonosky
sought. The Court noted:
• That section 17(4A)(b) contemplated an adjustment of the amounts stipulated in section 17
(4)(c) in respect of a claim for loss of income as at the date when the loss occurred, this
being the date of the collision. The subsection did not purport to deal with adjustments after
the date on which the cause of action arose.
• However, in respect of future annual loss, reliance had to be placed on the actuarial
calculations of future loss of earnings which, in the past, had always taken into consideration
a projected future inflation rate for each year up to the date of retirement. There was no need
to disturb that methodology when calculating future loss of earnings.

It emphasised that there has always been a speculative ‘looking into a crystal ball’ to come up
with a projected annual inflation rate during the future years up to retirement. Therefore, the
purpose of section 17(4A)(b) was intended to set a starting date for utilising adjustment
amounts when calculating loss of earnings after the accident and should not be used for any
other purpose.

34.2.6.5 Currency
Damages awards are usually made in South African currency, but a plaintiff may in some instances
claim damages in a foreign currency. The principle of nominalism applies and courts will award the
actual amount lost irrespective of the inflation rate or fluctuations in the currency. In Standard
Chartered Bank of Canada v Nedperm Bank Ltd 45 the plaintiff claimed damages in the amount of
$US2,5 million as a result of a negligent misstatement made by the defendant that resulted in the
plaintiff suffering harm. The Court held that the damages could be paid in foreign currency: 46
When one turns to the facts of the present case, it is apparent that the loss suffered by
Stanchart was basically in US dollars, the currency in which the bill of exchange was
expressed and which was paid or credited to Cansulex when the bill was discounted. This
was the currency in which its loss was ‘felt’. The type of transaction which led to the loss
had on previous occasions, it would seem, always been done in US dollars and, having
regard to Nedbank’s state of knowledge about this transaction and Stanchart’s involvement
(as previously elaborated), I am of the view that a loss in dollars was reasonably foreseeable.
In oral argument before us Mr Browde very fairly (and in my opinion very correctly)
conceded that, if the Court was empowered to grant judgment in a foreign currency, he
could not advance any argument that in this case the judgment should not be in the foreign
currency suggested, viz US dollars. I accordingly conclude that the damages to be awarded
in this case should be expressed in US dollars. It is implicit in any order to this effect that the
judgment debt may be satisfied in South Africa by payment in the foreign currency or by
the payment of its equivalent in Rand when paid … . Any other conversion date could
render meaningless the award in the foreign currency.

34.2.6.6 Contingencies
Contingencies are:
uncertainties or circumstances of life relevant to situations, conditions, illnesses or accidents
and which could have a negative or positive effect on the extent of damage, as well as
influencing the earning capacity and/or life expectancy of the affected person.47
Therefore, they are uncertain future events that could affect the amount of damages awarded and
so, once courts have calculated compensation in respect of future losses, they adjust the amounts
for contingencies. Assume, for example, that a plaintiff is injured, is unable to work again, and
claims damages for loss of future earnings. The plaintiff, now 25 years old, would have been
expected to work until the age of 65, but because of the injury, a court calculates that his life
expectancy is now only a further 30 years and that he would be able to work for no more than a
further 20 years. However, these calculations are no more than educated guesswork, and the
plaintiff may end up living and/or working for longer or for shorter periods. A court then has to
make a contingency adjustment to allow for the possibility that its assumptions in making its
calculations might be wrong. The percentage adjustment that the court decides upon depends upon
what it considers to be fair and reasonable in the circumstances.
Some academics do not agree with contingency adjustments. For example, the English writer
McGregor 48 regards a court’s insistence upon reduction for the contingency that the plaintiff dies
sooner than expected as ‘misconceived’ and notes: 49
For there is no more reason to assume that the claimant will die earlier than expected than
to assume that he will outlive his life expectancy, and the eventuality of earlier decease is
already taken into account in the actuarial tables [of the combined annuity and life
expectation variety] themselves.

Nevertheless, our courts have always taken contingencies into consideration.

Some contingencies may be beneficial to the plaintiff and have the effect of increasing the
amount (‘positive’ contingencies), while others (‘negative’ contingencies) serve to reduce the
amount and to prevent a plaintiff from benefiting unnecessarily. Therefore, courts may take into
account, for example, the likelihood that the future harm might be more severe than anticipated,
and increase the amount accordingly; or that the plaintiff’s anticipated harm might not occur, in
which event they adjust the amount downwards. In every instance, courts should always consider
both positive and negative contingencies that might impact on the amount. However, courts have
never made upward adjustments. Invariably, the amounts are reduced, with positive contingencies
merely affecting the extent of the eventual reduction.
There is no fixed formula for making these adjustments and so courts decide contingencies on
a case-by-case basis according to what they consider to be fair. Often courts will use actuarial
evidence to determine the likelihood of these contingencies arising and how they should reduce the
amount. However, a court is not bound by such actuarial evidence, for fairness and reasonableness
are normative concepts that cannot be proven in this way. Ultimately, the decision is in a court’s
discretion.
In Minister of Defence v Jackson 50 the Court said:
The contingencies to which regard is to be had in the making of a discount include such
matters as the possibility of an error in the estimation of the plaintiff’s life expectancy and
retirement age; the likelihood of illness, accident or unemployment which would have
occurred in any event, or which may in fact occur, and so affect the plaintiff’s earning
capacity; and the inflation or depreciation of money in the future. Naturally the amount of
any discount depends on the facts of each particular case. There is no standard rule. The
assessment is largely arbitrary.

In addition to those listed in Minister of Defence v Jackson,51 the following are examples of some
other contingencies that courts have considered:
• That the plaintiff’s life expectancy might be reduced and so he or she might require future
income for a shorter period than expected
• In the case of a widow’s claim for loss of support, the prospect of her marrying again, or the
likelihood that she and her husband might have divorced had he not died
• The prospect of the plaintiff being unemployed at a future date, due to, for example, economic
recession, or labour unrest
• That the plaintiff’s business might have failed
• The likelihood of other non-fatal accidents.

PAUSE FOR Contingency factors

REFLECTION Courts have considered a widow’s prospect of marrying again as a
contingency factor that reduces the amount of compensation awarded for
loss of support. In Shield Insurance Co Ltd v Booysen 52 the Court made
a 50% reduction for contingencies in an award for the prospect of the
widow remarrying. However, where an orphaned child claims for loss of
support, courts make no adjustment in respect of the possibility that the
child might be adopted in the future.53 Are these situations not
comparable? What policy factors would favour similar or different
treatments in these instances?

In Esterhuizen v Road Accident Fund 54 the Court emphasised that inasmuch as a contingency
provision for a remarriage is not unconstitutional, courts have to consider that a second marriage
may not necessarily result in financial support. Moreover, there is the possibility that the second
marriage may not last and that the financial support, if gained may be lost. These issues must also
be considered when a remarriage contingency is determined.55 However, the possible adoption of an
orphan is not considered when assessing loss of support claims; nor are foster care and child
support grants considered when making contingency adjustments. In Coughlan NO v Road
Accident Fund 56 the Constitutional Court overruled previous decisions 57 and instead held that
foster care and child support grants are unrelated to damages for loss of support and hence are
different in nature from compensation. The State acts in the role of caregiver when paying grants,
but indemnifies the wrongdoer when paying compensation via the Road Accident Fund.58

34.2.7 Lex Aquilia: Assessment and quantification of damages for

patrimonial harm
A plaintiff must prove that he or she has sustained a recoverable loss,59 whereafter our law of
damages recognises the principle of complete compensation. The purpose of a damages award is to
place the plaintiff in the position that he or she would have been in had the delict not been
committed – nothing more and nothing less. The principle of complete compensation is tempered
by principles such as the rules that relate to accounting for benefits, and mitigation of loss. We
discuss some of these principles in the following sections.

34.2.7.1 Assessment of harm: General principles

When one assesses the extent of patrimonial harm, one compares two positions: one prior to the
commission of the delict, and the other subsequent to this. If the comparison shows that the plaintiff
is worse off after the delict, the difference between these positions represents the harm that the
plaintiff has suffered. Once one has assessed the harm and its extent, one can begin the process of
quantification to place a monetary value on that difference.
 A comparative method called the sum-formula approach is used to assess harm. This formula
was described by the German jurist Mommsen and involves comparing the actual position in which
the victim of the harmful conduct finds himself or herself after the delict, with the hypothetical
position that the victim would have been in had the delict not been committed. The name ‘sum-
formula’ is derived from the fact that assessing the harm involves comparing an actual sum of
money with a hypothetical sum of money. When the sum-formula is used to quantify damages, it is
also referred to as the ‘negative interesse’ formula.60 It is the method of assessing the amount of
money required to place the plaintiff in the position he or she would have been in had the delict not
been committed.
Visser and Potgieter 61 note, however, that:
[a]lthough the sum formula (with its abstract comparative method) is indeed part of our law
of damages, it would seem that in certain instances our courts pay only lip-service to it and
do not actually follow this method.

In addition, the sum-formula has been subject to academic criticism. Some have argued that it
should be abolished and replaced by a concrete concept of harm in which the actual position of the
plaintiff prior to the delict is compared with the actual position after the delict.62 The negative
difference constitutes the harm suffered by the plaintiff. Santam Versekeringsmaatskappy Bpk v
Byleveldt 63 used this latter method, but nonetheless, the Supreme Court of Appeal recently
confirmed the sum-formula approach in Transnet Ltd v Sechaba Photoscan (Pty) Ltd: 64
It is now beyond question that damages in delict … are assessed according to the
comparative method. Essentially, that method, in my view, determines the difference, or,
literally, the interesse. The award of delictual damages seeks to compensate for the
difference between the actual position that obtains as a result of the delict and the
hypothetical position that would have obtained had there been no delict.

Therefore, it seems clear that despite the criticism, our courts remain committed to the abstract
method of assessing harm. Once a court has assessed the extent of the harm, it can then place a
monetary value on that loss.

34.2.7.2 Quantifying the damages: Damage to property

We will now show how damages are quantified in specific instances.
Where property is either damaged or destroyed, the point of departure is the sum-formula or
negative interest rule. This is supplemented by any one of the following tests:
• Diminution in market value: The plaintiff has to prove both the market value of the property
prior to the delict and the market value of the property after the delict. In Monumental Art Co v
Kenston Pharmacy (Pty) Ltd 65 the Court distinguished between the instances where property
was completely destroyed and where the property was damaged.66 In the former instance one
quantifies the damages with reference to the market value at the time the property was
destroyed. Where the property is damaged, the measure of the damages is the difference in the
market value before and after the harm ensued. The onus is on the plaintiff to prove the market
value of the property at the relevant times.
• Reasonable repair costs: This method is easier and more realistic, because often it is difficult to
prove the market value of a damaged object. This method would not be appropriate where the
property was very badly damaged or destroyed.
In Erasmus v Davis 67 the Court used reasonable repair costs to quantify the damage to a motor
vehicle. In a minority decision, Jansen JA noted 68 that there would be three instances where
reasonable repair costs would not be a good measure:
• Where the cost of repairs is more than the pre-accident market value of the property: In this
instance, the plaintiff would receive too much
• Where the cost of repairs is more than the diminution in value of property: Once again, the
plaintiff would receive too much
• Where the repairs restore the property to its pre-accident condition, but not to its pre-accident
value, simply because of the fact that it has been involved in an accident: In this instance, the
plaintiff would receive too little compensation.

34.2.7.3 Quantifying the damages: Personal injury

Where someone is injured as a result of bodily injury, that person may claim in respect of the
following types of patrimonial loss:
• Medical and hospital expenses
• Future medical and hospital expenses
• Rehabilitation costs (if the injury was very serious)
• Psychiatric expenses (if the person was sufficiently traumatised)
• Loss of earnings
• Loss of future earnings (or loss of earning capacity).
The Road Accident Fund’s liability to compensate the victim of a road accident is limited to
instances where the person has suffered ‘serious injuries’, to be determined by a registered medical
practitioner following a ‘prescribed method’. 69
Road Accident Fund v Duma and Three Related Cases (Health Professions Council of
South Africa as amicus curiae)70
Duma and three others suffered injuries in motor vehicle collisions. The RAF accepted
liability, except in respect of general damages. It contended that the respondents did not
suffer ‘serious injuries’ as required by section 17(1). The RAF argued that:
• The injuries had not been correctly assessed as serious.
• The RAF 4 forms were completed by a ‘health practitioner’, namely an occupational
therapist, and signed by a psychiatrist instead of a ‘medical practitioner’, as required by
regulation 3 of the Road Accident Fund Regulations.
• The respondents failed to undergo a physical examination and that the psychiatrist only
signed the forms after looking at hospital records.
• Pursuant to the RAF’s rejection of the RAF 4 forms, the respondents were supposed to have
referred the dispute to the Health Professions Council within 90 days, whereafter an appeal
would have been instituted. Instead, the respondents proceeded to trial.

The SCA upheld the appeal on the grounds that the respondents had indeed suffered
‘serious injury’ and that regulation 3 raised a challengeable constitutional issue regarding the
assessment of injuries. The Court accordingly postponed the issue of general damages sine
die and gave the respondents 90 days to refer the dispute to the Health Professions Council
for determination.
Brand JA took the opportunity to caution that it is the RAF, not a court that is vested with
the power to decide whether or not the injury to a third party was serious enough to meet the
threshold requirement for an award of general damages.71 Moreover, the fund’s decision to
reject the respondents’ RAF 4 forms constitutes an administrative action, which can be
reviewed under the Promotion of Administrative Justice Act 3 of 2000. This means until the
decision is set aside by a court on review or overturned in an internal appeal, it remains valid
and binding. The fact that the RAF gives no reasons for its decisions, or that the reasons
 given are unpersuasive or not based on proper medical or legal grounds cannot detract from
that principle. Whether the fund’s decisions were right or wrong is of no consequence as they
existed as a fact until set aside, reviewed or overturned in an internal appeal. Therefore, it
not open to a court to disregard the fund’s rejection of the RAF 4 forms on the basis that the
reasons given are insufficient.72

Section 17(4) of the Road Accident Fund Act provides that the fund shall, in an appropriate case,
provide plaintiffs with an undertaking for the costs of future accommodation in a hospital or
nursing home or treatment of or rendering a service or supplying of goods to him or her arising out
of injuries suffered in a motor vehicle collision. This section must be read together with section 36
of the Compensation for Occupational Injuries and Diseases Act 130 of 1993 (COIDA), which
provides that an employee who suffers occupational injuries in the course of employment may
claim compensation in terms of COIDA and may also institute action for damages against a third
party (such as the fund). In awarding damages in a third party claim, the court is required to have
regard to the compensation paid to the plaintiff in terms of COIDA.73

34.2.7.4 Quantifying the damages: Expenses

Expenses from personal injury that the plaintiff may have incurred by the date of the trial include
medical and hospital expenses. Damages under this head will also include the cost of an artificial
limb and modifying a vehicle and home in the case of disability.
As this harm has already materialised by the date of the trial, the plaintiff can present evidence
of these expenses in the form of accounts or invoices. Furthermore, courts can determine whether
the treatment the plaintiff received was necessary and whether the expenses were reasonable.
With future expenses, there is usually no clear evidence at the date of the trial as to the precise
extent of the expenses, yet the once-and-for-all rule requires the plaintiff to claim for future
expenses before they actually materialise. Courts, therefore, have to make an estimate based on the
present costs of treatment and any expert evidence as to the extent of treatment that the plaintiff
may require in the future. In such instances, the plaintiff does not necessarily have to prove future
medical expenses on a balance of probabilities, but will have to prove that there is a reasonable
possibility that such harm will arise.74 The process of quantifying such loss involves some degree of
speculation. However, despite the fact that it is difficult to assess this type of loss, courts will
nevertheless make an award of damages, as they rely on expert medical evidence. When awarding
damages for future expenses, courts capitalise and discount the award to present value.75
Where a dependant is injured and the associated financial loss, usually medical expenses, is
borne by the breadwinner, that breadwinner may claim in respect of such patrimonial loss.

34.2.7.5 Quantifying the damages: Loss of earnings and earning

capacity
If a person is seriously injured, the injury could result in incapacity to work, either temporarily or
permanently. If, at the date of the trial, the person has already lost income due to being unable to
work, the earnings lost up to the date of the trial are referred to as loss of earnings, or past loss of
earnings. It is easier to quantify this loss than loss of future earnings, because courts can easily
ascertain the plaintiff’s earnings at the time of the injury by looking at payslips or obtaining
evidence from the employer. Of course, if the trial takes place some years after the injury and the
plaintiff has been incapacitated for the entire time, this quantification could become problematic, in
that one would then have to factor in variables such as a promotion that the plaintiff could have had
in the meantime. Courts do not take inflation into account when calculating past losses.76
 One of the disadvantages of the once-and-for-all rule is that damages for future harm, such as
loss of earning capacity, have to be recovered before they materialise. The uncertainty in predicting
the future may lead to either over or under compensation, and it may be that someone who is
injured to such an extent that he or she will earn less in the future, or be completely unable to work,
will be compensated for what he or she will not earn in the future.
Southern Insurance Association Ltd v Bailey NO77
The plaintiff was the father of a two-year-old girl who instituted action against the defendant in
terms of the Compulsory Motor Vehicle Insurance Act 56 of 1972, the predecessor of the
Road Accident Fund Act. His daughter had been knocked down by a motor vehicle after
which she had an epileptic seizure and fell into a deep unconscious state, which lasted for a
long time. She eventually recovered consciousness and was released from hospital, but her
severe injuries were permanent and she would never be able to work in the future. Among
the damages claimed for was an amount for loss of earning capacity.
Nicholas JA eloquently described the process of quantifying loss of earning capacity:78

Any enquiry into damages for loss of earning capacity is of its nature speculative,
because it involves a prediction as to the future, without the benefit of crystal balls,
soothsayers, augurs or oracles. All that the Court can do is to make an estimate, which
is often a very rough estimate, of the present value of the loss.

It has open to it two possible approaches.

One is for the Judge to make a round estimate of an amount which seems to him to be
fair and reasonable. That is entirely a matter of guesswork, a blind plunge into the
unknown.

The other is to try to make an assessment, by way of mathematical calculations, on

the basis of assumptions resting on the evidence. The validity of this approach
depends of course upon the soundness of the assumptions, and these may vary from
the strongly probable to the speculative.

It is manifest that either approach involves guesswork to a greater or lesser extent. But
the Court cannot for this reason adopt a non possumus attitude and make no award
….

In a case where the Court has before it material on which an actuarial calculation can
usefully be made, I do not think that the first approach offers any advantage over the
second. On the contrary, while the result of an actuarial computation may be no more
than an ‘informed guess’, it has the advantage of an attempt to ascertain the value of
what was lost on a logical basis; whereas the trial Judge’s ‘gut feeling’ (to use the
words of appellant’s counsel) as to what is fair and reasonable is nothing more than a
blind guess … .

It is true that, in the case of a young child, the assessment of damages for loss of
earnings is speculative in the extreme. Nevertheless I do not think that even in such a
case it is wrong in principle to make an assessment on the basis of actuarial
calculations.

In President Insurance Co Ltd v Mathews 79 the Court said:

There is no reason in principle why, in an appropriate case, the cost of employing a
substitute should not form the basis of a claim for damages arising from a plaintiff’s
inability to carry on his pre-collision trade or profession.80
So, a person who will sustain further loss because he or she has to employ additional, necessary
labour in order to substitute certain aspects of functions which he or she can no longer perform
because of the injuries sustained, is entitled to be compensated for the cost of such substituted
labour on the basis of loss of future earning capacity.81
Raath v Nel82
Raath, an anaesthetist, administered a failed pre-operation intubation procedure on Nel that
resulted in Nel spending more than a month in the intensive care unit of a hospital. His
physical and mental recovery after his discharge was slow and problematic. Nel sued for
damages for the loss caused by his inability to attend to his business affairs as a game
farmer and auto dealer; for future medical and hospital expenses; as well as for general
damages. Raath accepted liability and the trial court awarded damages. The appeal hinged
on the specific amounts awarded.
Regarding loss of income and earning capacity, Raath argued that any loss that may have
been suffered was not suffered by the Nel personally, because for estate planning and estate
duty considerations, he had sold all his assets, including his shares and loan account in his
business, to a trust. The issue, therefore, was whether the loss suffered by Nel’s business
prior to the sale of assets to the trust, and by the trust thereafter, over the relevant period
could be characterised as Nel’s loss.
The Court held that Nel was not a capital beneficiary of the trust but was, in the discretion
of the trustees, a potential income beneficiary; and noted that this type of trust has become
popular for estate planning and tax purposes.83 It found that a trust estate, comprising an
accumulation of assets and liabilities, is a separate entity, but with no legal personality. The
core concept of a trust is the separation of ownership or control from enjoyment. Applied to
the present matter, the separateness of the trust estate had to be recognised and
emphasised. What Nel had sought, in effect, was the advantage of both a reduction in estate
duty and the continued retention of control and advantages of ownership of the trust assets.84
The appeal in respect of the claim for loss of earning capacity after the date of the sale of
assets to the trust was therefore upheld.

34.2.7.6 Quantifying the damages: Basic method for calculating loss

of future income
Courts have devised a basic method for assessing future losses. However, before discussing this
method, we explain the terminology that is used in this regard.

TERMINOLOGY Capitalisation
Capitalisation occurs when the income (or loss of support) that the
plaintiff would have earned over a period is reduced to a lump sum
to be paid out immediately. Had the plaintiff not been injured or
killed, the income or support would have been received monthly
over a number of years. In this case, the amounts that the plaintiff
would have earned or received are added together into a lump
sum, for example, R100 per month over five years is capitalised at
R6 000.
However, if the plaintiff were to receive R6 000 immediately, in
other words, before the amounts that make up the R6 000 are
due, the plaintiff would benefit unduly by investing that money and
earning interest, so that in the end, the plaintiff gains more than
the amount required to compensate for the actual loss (for
example, almost R10 000). So, courts discount the amount to
present value, in other words, they reduce the amount to an
 amount that, if it were to be invested immediately, would ultimately
yield a total value (lump sum plus interest earned) equal to the
capitalised amount. In that way, the plaintiff does not benefit at the
expense of the defendant. In our example, assuming an average
interest rate of 10% over the five-year period, R3 725,53 invested
immediately would ultimately yield R6 000. So, the court would
award the plaintiff damages amounting to R3 725,53.

Courts use a four-step method to determine loss of future income or loss of earning capacity: 85
1. Courts determine the present value of the income that the plaintiff would have earned in an
uninjured condition during the period that he or she would have been working. For example,
assume that the plaintiff is 40 years old at the date of the trial and would have retired at 65. The
plaintiff would therefore, but for the delict, have worked for another 25 years. Based on
evidence as to what the earnings were at the date of the delict, the plaintiff’s likely career path,
what the plaintiff’s promotion prospects were, and the likely inflation rate, a court has to
quantify the plaintiff’s average annual income for those 25 years. A court adds together the
average annual income for all the years to form a capital amount, which it then discounts to
present value. When discounting, a court has to speculate what the interest rate and the
inflation rate are likely to be. Courts often use actuarial tables and other evidence in the
process, but, as can be seen from the extract from Southern Insurance Association Ltd v Bailey
NO,86 courts are not bound by such evidence.
2. Courts now assess the income of the plaintiff in an injured state in the same way as in the first
step. If the plaintiff’s working ability is unaffected by the injury, then this amount will be the
same as in the first step. If the plaintiff is partially incapacitated, this amount will be less than
in the first step, if the plaintiff is completely incapacitated, the amount will be zero.
3. Courts then subtract the amount calculated in the second step from the amount calculated in
the first step. This result represents the plaintiff’s loss.
4. Courts now capitalise the amount calculated in the third step and adjust it for contingencies.

34.2.7.7 Quantifying the damages: Illegal earnings

Where the plaintiff was earning an income from an illegal source, courts do not compensate for the
loss of this income, because it would be against public policy. In Dhlamini v Protea Assurance Co
Ltd 87 the plaintiff had been earning a living as a hawker prior to the accident. He did not have a
hawker’s licence and the Court held that compensation for loss of income derived from such a
source would be against public policy. The illegal nature of the earnings impacts far more on
claims for past loss of earnings than on future losses, because courts can frame the latter as loss of
earning capacity. Van der Walt and Midgley write as follows: 88
The impact [of the illegal nature of earnings] on claims for future losses is less severe. In
such instances the criterion is loss of earning capacity which unlike loss of earnings, cannot
be termed ‘illegal’. According to principle, therefore, if the evidence supporting the
plaintiff’s claim for loss of earning capacity is acceptable, damages ought to be awarded,
even if past earnings were illegal. Illegal earnings can therefore be relied upon as some
indication of earning capacity.

This view accords with that of Boberg, who is of the opinion that compensating loss of earning
capacity instead of loss of future earnings would be a more equitable solution to the problem.
COUNTER Loss of earning capacity
POINT When a person can no longer work, this person has a claim for loss of
earning capacity. Visser and Potgieter89 write the following in this regard:

The precise theoretical explanation for this type of damage is

not clear, but the best explanation seems to be the existence of
a subjective right to earning capacity as legal object (that is, the
ability to earn money). The legal object in question is closely
related to a person’s bodily integrity, but not identical to it. There
is also a view that the loss under discussion is merely the loss of
actual future earnings without it being necessary to identify a
specific patrimonial asset.

Boberg suggests that the ‘loss of earning capacity’ approach is

preferable:90

[I]t is more flexible, and likely to extend compensation to those

not presently utilising their ability to earn money (e.g. the
housewife …, the skilled engineer who preferred his hobby of
watch repairing to his profession …, or the famous doctor who
becomes an unpaid medical missionary …). It should enable an
illegal earner … to recover something for loss of his legitimate
earning capacity if it appears that he may one day have ‘gone
straight’. And if the true rationale of compensation is a loss of
earning capacity, the receipt of collateral benefits is rightly
disregarded, for they do nothing to restore that which was lost.

If one follows Boberg’s approach, one could bypass the difficulty that
cases such as Dhlamini v Protea Assurance Co Ltd 91 raise, where one is
unable to claim compensation for loss of future income, because prior to
the delict one has been earning an income from a ‘morally neutral’ illegal
activity, such as the failure to obtain the necessary licence. The same
applies to claims for loss of support in instances where the deceased was
receiving an income from an illegal source.

Heese NO v Road Accident Fund92

The Court held that if it appears from evidence that a claimant’s earning capacity would likely
be rendered worthless by some or other event in the future period covered by the claim, then
a court could properly conclude that a claim of diminution in earning capacity had not been
established on a balance of probability. The future event could, for example, be a lengthy
imprisonment. It is a factual question whether earning capacity can been rendered worthless
or diminished in value by such a future event.93
The claimant, a German national, was seriously injured in a motor vehicle collision in
South Africa. His curatrix ad litem, Heese, claimed compensation on his behalf from the
Road Accident Fund. The claim was settled on general damages and medical expenses, the
only remaining issue being compensation for loss of earning capacity. The problem was that,
although the claimant was a businessman, his earnings were made illegally under a massive
fraudulent tax-evasion scheme. Not only was he under-declaring his earnings but he was
also inflating his expenses by, for example, deducting fictitious expenses from his income.
The Court held that the claimant was not entitled to compensation for loss of illegal
earnings. It was a factual question whether the claimant’s earning capacity would have been
rendered worthless or diminished in value by a future event such as imprisonment.94
On grounds of public policy a South African court would not make an award for diminution
in earning capacity if the only way in which the earning capacity could remain productive was
 through a failure on the part of the claimant, post-accident, to comply with his or her legal
duties to the tax authorities.95

34.2.7.8 Quantifying the damages: Loss of support

Where a breadwinner is injured or killed, the dependants will suffer patrimonial loss in the form of
loss of support. The claim for loss of support could include both damages for harm suffered prior to
the date of the trial, and damages for future losses, particularly where the breadwinner dies or is
permanently disabled.
For calculating future loss of support, courts have to make an assessment based on facts not yet
known at the date of the trial. Courts use a method similar to the one used in loss of future earnings.
If the dependants stand to inherit from the deceased breadwinner, courts will offset this inheritance
against their claim for loss of support.
Where the deceased’s income came from an illegal source, courts will refuse a claim for loss
of support. In Santam Insurance Ltd v Ferguson,96 the wife of a deceased panel beater instituted a
claim for loss of support. The deceased had practised his trade in contravention of a municipal
ordinance. The Court a quo allowed the claim on the basis of loss of earning capacity: 97
In my view the Roman-Dutch law has advanced far beyond the principle that the sins of the
fathers are visited upon the next generations. Whereas a plaintiff who personally has lost
‘illegal’ income may be frowned upon by the Court and non-suited on that account, non
constat that a plaintiff who has lost earning capacity, i.e. future loss of income, must be
frowned upon. He has been injured and, if the injuries are permanent, his earning capacity
is affected. … But it does not follow that in the future he is necessarily going to carry on his
illegal business or activities. A fortiori his dependants cannot be non-suited for what is their
loss after the deceased’s ‘illegal’ activities have ceased.

To non-suit a widow or child or in the extreme case an unborn child, because the deceased
husband or father was making a living out of the sale of lucerne seed to unauthorised buyers
is, in my opinion, unjust and on principle unsound. The widow or child may know nothing
about the deceased’s activities and, if the illegality was discovered and stopped, it is obvious
that the paterfamilias would have resorted to some other form of livelihood.

This view supports the opinion of Boberg and others that basing the claim on the deceased’s loss of
earning capacity is a better approach. On appeal, however, the Court rejected the claim, because of
the illegal nature of the deceased’s income. The Court held that the deceased’s income was earned
illegally and that this had to influence the claim for loss of support, and that the illegality had to
preclude the claim.98

34.2.7.9 Quantifying the damages: Mitigation of loss

Where someone suffers harm as a result of delictual conduct, that person is expected to take
reasonable steps to limit the loss suffered. This is known as ‘mitigation of loss’. Should a person
incur expenses in attempting to limit the loss, the person can recover these expenses as part of the
damages award. For example, assume that a plaintiff’s delivery truck is badly damaged in an
accident and that the plaintiff, in addition to the damage to his truck, will also suffer loss of income
as a result of not being able to make deliveries. The plaintiff could also lose clients who, during the
period that no deliveries are made, may decide to engage the services of another supplier. To
prevent this from happening, the plaintiff may decide to rent another truck while the damaged truck
is being repaired and claim the expenses as damages.
 Courts do not expect a plaintiff to take unusual steps to limit loss. However, where a plaintiff
does take such steps, the plaintiff can recover compensation for reasonable expenses only. Where a
plaintiff does not take any steps, the plaintiff will not be entitled to recover damages for loss that
could reasonably have been prevented.
The plaintiff does not need to prove that the steps taken were reasonable. The onus is on the
defendant to prove that the plaintiff did not take reasonable steps to mitigate his or her loss, or that
the steps that the plaintiff did take were unreasonable.
Shrog v Valentine99
The plaintiff’s truck was damaged when he drove into a trench that the defendant had dug.
The plaintiff had been using his truck for his business at the time of the incident. In order to
continue with his business, the plaintiff rented another truck and included the cost of renting
this vehicle in his claim for damages.

The Court held as follows:100

A plaintiff is entitled to claim for loss of income or loss of profits which results from his
being deprived of the use of the vehicle while it is being repaired. Since a party is
bound to mitigate his loss a businessman should normally hire another vehicle in place
of the damaged one, and he may then claim the expense of hiring it.

In that event, however, he does not have to prove that the expense was reasonable,
the reasoning being:

‘… the lost profits owing to the deprivation of the use of a vehicle are not
required to be reasonable. They are the actual profits. Here that loss was
avoided by the taking of steps by the plaintiff to hire a lorry to replace the
damaged vehicle. The expense is an expense for which the defendant is
responsible unless he can show that the action of the plaintiff in hiring a lorry
at all, or in hiring a lorry at that cost, was unreasonable.’

PAUSE FOR Overlapping rules

REFLECTION Sometimes different rules overlap. For example, in Smit v Abrahams 101
the facts of the case lend themselves equally to applying the vulnerability
(egg skull) rule, the principles of legal causation, and the rule of
mitigation of loss. In this case, the plaintiff’s vehicle had been damaged
as a result of the defendant’s conduct. The plaintiff had used his vehicle
in his business as a hawker and could not afford to buy a new vehicle.
Renting a vehicle was also too expensive. He rented a vehicle from a
friend at a low price for a while, but thereafter the friend wanted his
vehicle back, and the plaintiff moved his goods by wheelbarrow. As the
plaintiff could not afford to replace his vehicle, he also suffered
considerable loss of income. The magistrate’s court decided the matter
on the mitigation of loss rule and came to the conclusion that in the
circumstances, the plaintiff had taken reasonable steps to mitigate his
loss. The magistrate furthermore took the plaintiff’s weak financial
position into consideration, which prevented him from replacing the
damaged vehicle (egg skull rule). Thirdly, both the Court a quo and the
Supreme Court of Appeal dealt with the matter on the basis of legal
causation, although their approaches differed. The Court a quo held that
legal causation had been established because the harm had been
reasonably foreseeable, while the Supreme Court of Appeal decided the
question of legal causation using the flexible approach.
34.2.8 Germanic remedy: Assessing reparation for infringements of
physical-mental integrity
A plaintiff recovers compensation for negligent infringements of physical-mental integrity using
the Germanic remedy, also known as the action for pain and suffering. In the case of non-
patrimonial loss, it is very difficult to measure the harm in terms of money. Therefore, the process
of quantification is often speculative, because courts have no precise guidelines to assist them in
arriving at a figure that accurately reflect this type of harm.

34.2.8.1 Introduction
In assessing the amount of damages awarded in claims in terms of the Germanic action, courts
place a monetary value on non-patrimonial loss. A number of factors will influence the size of the
awards, such as the seriousness of the harm, the age and lifestyle of the plaintiff, and the duration
of the harm suffered. As a general rule courts tend to be conservative when making awards for non-
patrimonial loss. In De Jongh v Du Pisanie NO 102 the Court said:
I can, however, not agree with the trial court’s point of departure in accordance which it
held that stinginess with compensation for serious injuries does not belong in a civilised
society. Because it is the defendant, and not the society, who pays the compensation,
stinginess on the part of society has nothing to do with the case.

It behoves society equally little to waste the defendant’s money, even if he or she is legally
liable because of his or her negligent conduct. The following dictum of Holmes J in Pitt v
Economic Insurance Co Ltd 1957 (3) SA 284 (N) at 287E–F is therefore appropriate in the
present case:

‘[The] Court must take care to see that its award is fair to both sides – it must
give just compensation to the plaintiff, but must not pour our largesse from the
horn of plenty at the defendant’s expense.’

Conservatism in the award of general damages has its origin in the fact that the defendant
also has to be treated fairly, not in the stinginess of society towards the plaintiff.

Courts take previous awards into consideration, and also the nature of the injuries in the previous
awards. In this regard, the leading case is Protea Assurance Co Ltd v Lamb: 103
It should be emphasised, however, that this process of comparison does not take the form of
a meticulous examination of awards made in other cases in order to fix the amount of
compensation; nor should the process be allowed so to dominate the enquiry as to become a
fetter upon the Court’s general discretion in such matters. Comparable cases, when
available, should rather be used to afford some guidance, in a general way, towards assisting
the Court in arriving at an award which is not substantially out of general accord with
previous awards in broadly similar cases, regard being had to all the factors which are
considered to be relevant in the assessment of general damages. At the same time it may be
permissible, in an appropriate case, to test any assessment arrived at upon this basis by
reference to the general pattern of previous awards in cases where the injuries and their
sequelae may have been either more serious or less than those in the case under
consideration.
34.2.8.2 Pain and suffering
When a person suffers bodily injuries in addition to the loss in the form of medical, hospital and
other expenses, that person also may endure physical and psychological pain. This harm is real, and
therefore the person is entitled to some reparation. When assessing the amount of damages, courts
have to consider the degree and duration of the pain and reach a monetary figure that acknowledges
the suffering and repairs the harm in some way. However, there is no formula that courts can use to
do this and so the result depends on what the court deems fair in the circumstances.104
Pain and suffering includes psychological harm and mental anguish. Where psychiatric injury
is associated with bodily injury to the plaintiff, courts assess general damages as part of the
assessment for pain and suffering.105 (Any actual expenses, for example, the cost of psychiatric
treatment and medication, would be claimed under the lex Aquilia.) However, psychiatric injury
might also give rise to a free-standing claim in the absence of any physical injury, as happened in N
v T,106 where a mother was awarded general damages in respect of stress suffered while coping with
the rape of her daughter.

34.2.8.3 Loss of amenities of life

Courts also award damages for any inability to live one’s life to the full, in the same way that the
person would have been able to prior to the delict. Administrator-General, South West Africa v
Kriel 107 describes this as:
those satisfactions in one’s everyday existence which flow from the blessings of an unclouded
mind, a healthy body, and sound limbs.

The award compensates past and future loss of amenities of life. The purpose of the award is not to
punish the defendant, but to console the plaintiff and to increase his or her feeling of well-being.
The amount of damages will depend on the circumstances of each case, for example, the extent of
any disability and/or disfigurement; whether the person’s life expectancy is reduced, and to what
extent; the effect on the plaintiff, both physically and mentally; the person’s lifestyle before and
after the delict; and whether paramedical aids have been included in the assessment of
compensation under the lex Aquilia. (If a person has been given money for a wheelchair, or an
artificial leg, then that would affect the award for loss of amenities of life, since a person who is
mobile is able to enjoy life better than one who is not.)108
Ngubane v South African Transport Services109
In this bodily injury case, the Court had to guard against duplication of awards. Often, items
claimed under one head of damages can also serve to compensate harm under another
head. In this case, the plaintiff claimed damages after having fallen out of a moving train. He
claimed the following:
• Past loss of income
• Loss of earning capacity
• The purchase and use of a motor vehicle (this was necessary to transport him from the rural
area where he lived to receive medical services)
• Future medical expenses and adaptive aids
• The cost of an attendant
• General damages for pain and suffering, loss of amenities of life, and disability.

The Court noted, in making the award for general damages, that it had to guard against
awards overlapping and the plaintiff being overcompensated:110
 It remains to consider an appropriate award under this head of damages. The trial
Court considered that R85000 would be fitting. [It was] …, however, submitted that it
was excessive and that the award should be no more than R60000. He correctly
pointed out that one must guard against overlapping and a resultant duplication of
awards for general damages and cited what was said by Hoexter JA in Administrator-
General, South West Africa, and Others v Kriel 1988 (3) SA 275 (A) at 286C–D:

‘As pointed out by this Court in Southern Insurance Association Ltd v Bailey
NO 1984 (1) SA 98 (A) at 113E–F, where (as here) damages for bodily
injuries are awarded not in a globular amount but under separate heads, a
trial Court should guard against the danger of duplication as a result of an
overlapping between separate awards.’

In the present case, in addition to the paramedical aids, there are other forms of relief
provided for in the award thus far made, which will ameliorate the hardship of the
appellant’s disability and his loss of amenities. These include the use of a motor car for
social and what might be termed ‘non-medical’ purposes. The fact that he is now to
live at home with his family and an attendant, is likewise a compensatory
consideration. The trial Court, it should be noted, was mindful of this danger of
duplication when making an assessment of compensation for general damages under
this head.

COUNTER Damage claims of unconscious persons

POINT Should a person who is in an unconscious or vegetative state be able to
claim damages for pain and suffering? According to Gerke NO v Parity
Insurance Co Ltd,111 the test:

(a) is objective in that something falls to be awarded for what has been
called loss of happiness even in a case where the victim has been
reduced to a state in which he has never realised and will never realise
that he has suffered this loss; (b) is, however, subjective, in the sense
that the Court, in fixing quantum, will have regard to any relevant data
about the individual characteristics and circumstances of the plaintiff
which tend to show the extent and degree of the deprivation; (c) is
subjective, also, in the sense that any realisation which the plaintiff has,
or did have or will have, of what he has lost, is most material and
important. This is the true compensable suffering (as distinct from pain)
which will carry far heavier damages than the somewhat artificial and
notional award referred to in (a) above. This suffering will continue only
for the expected duration of his life.

In Reyneke v Mutual and Federal Insurance Co Ltd 112 the Court noted
that the plaintiff must have subjectively experienced the pain, and in
Collins v Administrator, Cape,113 the Court added a further requirement:
that the award must serve some purpose in that the plaintiff must be able
to know of and appreciate the award that is made. So, according to
Gerke, an unconscious plaintiff could at least claim under heading (a).
According to Reyneke, a court will only award damages if the
unconscious plaintiff at some stage subjectively experienced some pain,
while the Collins approach, which requires not only a subjective
experience of pain, but also that a plaintiff must be able to enjoy the
award, would result in an unconscious plaintiff receiving no award at all.
A fourth approach, similar to that in Reyneke, holds that unconscious
plaintiffs, or those in a vegetative state, should be treated no differently
 from other plaintiffs, in that if pain is experienced, damages should be
awarded. See Roberts NO v Northern Assurance Co Ltd.114
Whether an unconscious plaintiff is entitled to compensation for loss
of amenities of life is similarly not settled. In the Reyneke case, the Court
considered disfigurement, disability and loss of life expectancy to be
objective concepts that lend themselves to a more functional
assessment,115 and awarded damages for loss of amenities of life.
However, in the Collins case, the unconscious plaintiff did not succeed in
claiming damages under this head. Instead, the Court followed the
personal-functional approach, which attempts to provide the plaintiff with
an amount of money that is reasonably able to remove the unhappiness,
and since the unconscious plaintiff could not appreciate this, no damages
were awarded.

34.2.9 Actio iniuriarum: Determining the amount that would provide

appropriate satisfaction
Plaintiffs can claim under the actio iniuriarum in instances of intentional infringement of the right
to a person’s bodily integrity, dignity and good name. This type of harm is non-patrimonial and
cannot be measured in money. The purpose of the award is to provide some relief for the
infringement in the form of monetary satisfaction and to attempt to assuage the feelings of injustice
that the plaintiff may feel. In all instances courts make the award ex aequo et bono, which means
that the award reflects what a court considers to be fair and just in the circumstances.

34.2.9.1 Assault
An assault infringes a person’s right to bodily integrity and a person can claim satisfaction for the
mere fact that that right has been infringed. In addition, if one has incurred medical expenses, one
can also claim damages for such patrimonial loss under the lex Aquilia.
When assessing the harm and quantifying the satisfaction, courts have to consider a number of
factors, for example, the nature and seriousness of the assault, any indignity associated with it, and
any psychological after-effects. Whether the assault was the result of provocation could also play a
role. In Ramakulukusha v Commander, Venda National Force 116 policemen severely assaulted the
plaintiff while he was detained. In deciding upon the amount of damages, the Court considered the
period over which the assault took place, and the fact that in this particular case, the assault was
serious and also extremely vicious. The Court considered previous awards, but found that there was
no previous case that it could compare with the present one.

34.2.9.2 Sexual abuse

The factors that courts have to consider in sexual abuse cases are similar to those in assault cases.
However, courts consider sexual abuse to be a particularly serious violation of bodily integrity,
coupled with a high degree of humiliation and prolonged psychological impact.117 There is
constitutional justification for reassessing principles relating to quantum when a delict strikes the
foundational constitutional values of our society.118 Courts are therefore duty-bound to deal
decisively with the appropriate value to be placed on upon the loss of dignity of a victim.119
Van Zijl v Hoogenhout120
 The plaintiff had been subjected to many years of sexual abuse by her uncle. The Court
awarded general damages in the amount of R200000, but this was a lump sum that included
damages for pain and suffering, and those under the actio iniuriarum. The Court considered
the serious and prolonged nature of the abuse when making its award.121

34.2.9.3 Deprivation of liberty

Courts are particularly concerned about protecting individual liberty and their awards tend to reflect
this concern. Other factors that are relevant when assessing the amount of damages are the
plaintiff’s social standing and reputation, the circumstances of the detention, its length, and its
psychological effects.122
Minister of Safety and Security v Seymour123
This case provides some guidance as to how courts should assess general damages for this
type of case in a constitutional era. Seymour, a 63-year-old man had been unlawfully arrested
and imprisoned for a period of five days, first for a short while at a police station and later, for
the bulk of the period, in hospital. The Supreme Court of Appeal took the view that courts
have always given prime regard to the rights to liberty and dignity, and that earlier cases did
not place less value on personal liberty than ought to be placed today:

The real import of the Constitution has not been to enhance the inherent value of
liberty, which has been constant, albeit that it was systematically undermined, but
rather to ensure that those incursions upon it will not recur.124

The Court also commented that previous awards:

are a useful guide as to what other courts have considered to be appropriate but they
have no higher value than that.125

After reviewing the awards in a number of cases, the Court concluded that Seymour had had
free access to his family and medical adviser throughout his detention at the police station.
He suffered no degradation in excess of that which is generally associated with being
arrested and detained. Seymour had spent only the first 24 hours of his arrest in detention,
the remainder of the detention was in a hospital bed at the Rand Clinic. Conceding that the
experience throughout must have been traumatic and stressful, the Court noted that the
continuing depression and anxiety could not be attributed solely to the arrest and detention,
and concluded that an amount of R90000 was appropriate. The Supreme Court of Appeal’s
award was startlingly disparate from that made by the Court a quo (R500000), and serves as
a reminder that in a rights-based society, one should not encourage over-zealous litigation at
the expense of other rights that are no less important.

34.2.9.4 Dignity, privacy, and identity

When infringements of rights fall under the broad dignitas umbrella (dignity, privacy and identity)
the circumstances of each case guide courts when assessing satisfaction, particularly:
• The social status of the plaintiff, and his or her personality and exposure to publicity
• The nature and seriousness of the infringement
• The attitude of the defendant at the time (for example, whether there was malice, or abuse of a
personal relationship) and after the event (for example, whether an apology was tendered)
• The nature and extent of the humiliation and distress that the plaintiff experienced.126

The Constitutional Court has pronounced on the continued existence of the delictual action for
injury and insult to self-esteem (contumelia) and loss of comfort and society (consortium) as a
result of adultery. The Court reminded us in DE v RH 127 that the abusive conduct of a non-
adulterous spouse has always been relevant to quantum of damages, although not necessarily to the
question of liability.128

34.2.9.5 Defamation
Where a person’s right to a good name is infringed, the purpose of the award is to provide some
satisfaction. Courts determine the amount by considering the nature of the words used, the nature
and extent of the publication, the plaintiff’s character, status and reputation, the probable
consequences of the defamation, the subsequent conduct of the defendant (whether any attempt was
made to rectify the harm done, for example, an apology or a refusal to apologise), as well as
comparable awards and the diminishing value of money.129
• In Pont v Geyser 130 the defendant persisted with defamatory remarks and repeatedly refused to
apologise. In addition, the Court regarded the extreme nature of the defamation and the
perceived malice on the part of the defendant as aggravating factors. Mitigating factors
included factors such as the truth of the statements, retraction and apology, absence of malice,
and provocation. Interestingly, even misconduct on the part of the plaintiff can serve as a
mitigating factor.
• In Naylor v Jansen; Jansen v Naylor 131 the plaintiff had been accused by a fellow employee of
misappropriating company funds. The Court found that there had indeed been defamation, but
took into account the fact that the plaintiff had breached his contract of employment and acted
in conflict with his employer’s interests.132 Although the plaintiff’s conduct did not amount to
misappropriation or theft, the plaintiff nevertheless had been in breach of his duty of good faith
towards his employer, which also included an element of dishonesty. While the plaintiff’s
conduct did not serve as a factor that justified the remarks made by the defendant, the Court
nevertheless held it to be a factor that would reduce the damages award.
• In Iyman v Natal Witness Printing & Publishing Co (Pty) Ltd 133 the plaintiff sued the
newspaper because of an allegation that he had assaulted a person, because this person had not
voted for his father in a local election. The Court found for the plaintiff, but held that a
mitigating factor was that the plaintiff had indeed assaulted the person, albeit for another
reason.134

34.3 Retraction and apology

A retraction and apology may act as mitigating factors when assessing damages under the actio
iniuriarum and reduce the amount of the award. During the Middle Ages, the amende honorable
and amende profitable were two remedies that were available to the victim of an iniuria. The
amende profitable in effect was the same as the actio iniuriarum. The amende honorable did not
entail claiming money, but rather retracting defamatory words and issuing an apology. The remedy
was thought to have been abrogated by disuse, and therefore no longer applicable in South African
law.135 For many years it was relegated to the chapters of delict books that dealt with the historical
development of the actio iniuriarum. In 2002, the remedy reappeared in the case law, but courts
have yet to use it to its full extent.
Mineworkers Investment Co (Pty) Ltd v Modibane136
The plaintiff, a black empowerment company, instituted an action for defamation against the
defendant based, inter alia, on:
• A letter written by the defendant to certain senior persons at Johnnic Holdings, the holding
company of the magazine Financial Mail
• A telephone conversation between the defendant and a strategy manager at BP
• Statements made by the defendant to a journalist from The Star newspaper.

In all these instances, the statements made were to the effect that the plaintiff company had
betrayed the black empowerment cause and was acting for its own gain, or that it was
incompetent.
The plaintiff claimed satisfaction, and in addition, claimed an apology and retraction of the
offending words.

The remedy entails that the plaintiff demand the following:

• That the defendant withdraw the defamatory statement
• An admission of guilt as well as an apology.

Willis J held that the amende honorable had in fact merely been forgotten (‘a little treasure
lost in a nook of our legal attic’) and concluded that it was still very much part of our law.137 He
was convinced that even if the remedy had been abrogated by disuse, there were other
reasons why an analogous remedy should be available in our law. There are two reasons
why a damages award often does not strike a balance between freedom of expression and
the right to reputation:
• It fails to protect the reputation of the plaintiff
• It can impose restrictions on freedom of expression, because a damages award can
potentially ruin a defendant.
In this regard, the amende honorable or an analogous remedy would be far more effective. It
is far less expensive and potentially far less disastrous to a defendant, and so it will not inhibit
freedom of expression. On the other hand, a public apology can do far more to restore the
victim’s reputation and give him or her a feeling of satisfaction.
The Court then referred to section 173 of the Constitution of the Republic of South Africa,
1996, read with section 39(2), in terms of which courts are to develop the common law and,
when doing so, promote the ‘spirit, purport and objects of the Bill of Rights’. Willis J went on
to say:138

Even if the amende honorable had never existed, the imperatives of our times would
have required its invention. In my view, it is entirely consonant with ‘the spirit, purport
and objects’ of the Bill of Rights in our Constitution that a person who has committed a
wrongful act by defaming another should, in suitable circumstances, be given an
opportunity to make an appropriate public apology in lieu of paying damages; and, no
less importantly, that the victim of a defamation should similarly have the opportunity of
having a damaged reputation restored by the remedy of a public apology. In the
circumstances of this particular case, I am satisfied that it would be just and equitable
that the defendant be given a choice between making a public apology or paying
damages.

This decision differs from the decision in Young v Shaikh.139 The plaintiff in that case did not
claim an apology and retraction, instead the defendant offered the apology as a way of
avoiding the damages claim, that is, as a defence. However, because of the serious nature of
the defamation, the Court did not accept the appropriateness of this remedy:140

Even if the ‘little treasure’ can be recovered from a ‘nook in our legal attic’, I do not
believe that a published apology in this matter would serve the interests of justice.

Freedom of expression does not include the right to falsely attack the integrity of a
fellow citizen for selfish reasons or for reasons which have nothing to do with ‘public
benefit’.
 If the award which I intend to make will have a ‘chilling’ effect on possible future and
similarly baseless and selfish attacks on the integrity of others, it would certainly, in my
view, be an additional reason not to make use of the lost ‘little treasure’.

The Court in Young v Shaikh 141

also left open the question of whether the amende honorable
was part of our law.

Dikoko v Mokhatla142
The plaintiff was the chief executive officer of the Southern District Municipality. The
defendant was the executive mayor. The defendant had far exceeded his monthly cellphone
allowance of R300 and had accumulated an excess of R3200, payment of which was long
overdue. The Provincial Auditor-General had, in letters to the plaintiff, questioned the overdue
payment and was not satisfied with an agreement between the defendant and the council to
write off the debt. The Auditor-General called on the defendant to appear before the North
West Provincial Public Accounts Standing Committee to explain this debt. In the course of his
explanation the defendant made a defamatory remark about the plaintiff, for which the plaintiff
sued him for damages.
The High Court awarded damages in the amount of R110000. The defendant appealed to
the Supreme Court of Appeal, which dismissed the appeal without providing reasons. He
then resorted to the Constitutional Court, claiming that the words had been uttered on a
privileged occasion on the basis of section 28 of the Local Government: Municipal Structures
Act 117 of 1998, and section 3 of the North West Municipal Structures Act 3 of 2000. The
Constitutional Court dismissed the appeal.
The separate judgment of Sachs J is particularly significant in that it juxtaposes the African
value of ubuntu with the amende honorable. Sachs J was of the opinion that the actio
iniuriarum is not always a satisfactory remedy:143

The notion that the value of a person’s reputation has to be expressed in Rands in fact
carries the risk of undermining the very thing the law is seeking to vindicate, namely
the intangible, sociallyconstructed and intensely meaningful good name of the injured
person. The specific nature of the injury at issue requires a sensitive judicial response
that goes beyond the ordinary alertness that courts should be expected to display to
encourage settlement between litigants. As the law is currently applied, defamation
proceedings tend to unfold in a way that exacerbates the ruptured relationship
between the parties, driving them further apart rather than bringing them closer
together. For the one to win, the other must lose, the scorecard being measured in a
surplus of Rands for the victor.

Both Mokgoro J and Sachs J believed that in this regard, the amende honorable could
provide a more meaningful solution in that it could more effectively restore the plaintiff’s
reputation, and at the same time also effect some type of reconciliation between the parties.
The Justices believed that the amende honorable accorded well with the African notion of
ubuntu, which is one of the values that underlie our Constitution. In this regard, Mokgoro J
said:144

A remedy based on the idea of ubuntu or botho could go much further in restoring
human dignity than an imposed monetary award in which the size of the victory is
measured by the quantum ordered and the parties are further estranged rather than
brought together by the legal process … .

The focus on monetary compensation diverts attention from two considerations that
should be basic to defamation law. The first is that the reparation sought is essentially
for injury to one’s honour, dignity and reputation, and not to one’s pocket. The second
is that courts should attempt, wherever feasible, to re-establish a dignified and
respectful relationship between the parties. Because an apology serves to recognise
 the human dignity of the plaintiff, thus acknowledging, in the true sense of ubuntu, his
or her inner humanity, the resultant harmony would serve the good of both the plaintiff
and the defendant. Whether the amende honorable is part of our law or not, our law in
this area should be developed in the light of the values of ubuntu emphasising
restorative rather than retributive justice. The goal should be to knit together shattered
relationships in the community and encourage across-the-board respect for the basic
norms of human and social interdependence. It is an area where courts should be pro-
active in encouraging apology and mutual understanding wherever possible.

Although Sachs J and Mokgoro J supported the remedy, their remarks in this regard were
obiter, as the case in fact dealt with issues of quantum.

In 2011, the remedy appeared again, this time in the Constitutional Court.
Le Roux v Dey145
Two schoolboys published a digitally-modified image in which the faces of the deputy
principal (Dr Dey) and the principal of their school were superimposed on an image of two
naked male bodybuilders sitting in what the Court referred to as ‘a sexually suggestive
posture’. The genitals of the two men in the image were covered by the school’s crest. Dr Dey
instituted action for damages claiming that the images were defamatory and in the alternative,
that his constitutional right to have his dignity respected and protected had been violated. The
Constitutional Court was heavily divided, but the majority held that the publication of the
image constituted defamation. The Court ordered the boys to apologise to Dr Dey and also
ordered them to pay a sum of money in damages as compensation.
Froneman J and Cameron J in a minority judgment outline the purpose of an apology and
its role in assessing damages. They moreover highlight the importance of developing the law
insofar as this remedy is concerned because it reflects restorative justice as one of its
principal values. The judges note that common-law and customary-law principles have to be
brought in line with constitutional values and norms.146

34.4 Interdict
An interdict is a court order by which means a plaintiff can prevent conduct that causes harm, or
prevent a continuation of harmful conduct that has already commenced. Interdicts are either
mandatory or prohibitory. A mandatory interdict requires a positive action from the wrongdoer, for
example, to destroy a consignment of fake goods sold under a well-known brand name. A
prohibitory interdict requires the wrongdoer to desist from wrongful conduct or from continuing
wrongful conduct, for example, a person can obtain a prohibitory interdict against a newspaper to
stop it from printing a potentially defamatory article, or to stop it continuing with a particular story.
One can also classify interdicts as either final or temporary. In a final interdict, a court places a
permanent ban on the threatened conduct. In a temporary interdict, a court prohibits the threatened
conduct pending the outcome of another hearing. For example, if a newspaper wishes to print a
story exposing a well-known politician, the politician may go to court to obtain a temporary
interdict. If the application is successful, the publication will be interdicted until the date of the
hearing, on which day a court will either set aside the interdict, and the newspaper can print the
story, or make a final order, in which case the interdict stands and the newspaper will not be able to
print or continue with the story.
The requirements for an interdict are straightforward: (a) a clear right, (b) an actual or
threatened infringement of a right, and (c) the absence of another suitable remedy.
Hotz v University of Cape Town147
 Students allegedly caused extensive damage to University of Cape Town (UCT) property
during the course of a protest. Alleged threats of further damage led UCT to apply for an
interdict preventing five students from protesting within the university premises. The High
Court a final interdict against the five students, which the Supreme Court of Appeal
confirmed. UCT had demonstrated:
• A clear right
• An injury actually committed or reasonably apprehended
• The absence of similar protection by any other ordinary remedy.148

The SCA however highlighted that although the right to protest against injustice is protected
under the Constitution, the manner in which the right is exercised is also subject to
constitutional norms. Destroying, damaging or defacing property during the course of a
protest is unlawful conduct. This also applies to actions that have the effect of interfering with
the rights of others.149

If one adapts these requirements for an interdict to a delictual context, the requirements read: 150
1. Conduct on the part of the respondent: This conduct may be either a positive act or an
omission. The conduct must have commenced or be threatening.
2. The conduct will cause or threaten to cause irreparable harm: The applicant has to prove that
the respondent’s conduct threatens to infringe or is actually infringing a so-called ‘clear right’.
This ‘clear right’ involves a recognised subjective right. However, in the absence of such a
right, an interdict can also lie in the case of non-compliance with a legal duty, for example, in
the case of pure economic loss.151
3. The causing of harm must be wrongful: This requirement links with the previous requirement
of infringement (actual or threatened) of a clear right. One applies the usual test for
wrongfulness in this instance.
4. There is no other remedy available to the applicant.

It is not necessary to prove fault on the part of the respondent, because in this instance, harm has
not yet occurred. The purpose of the interdict is usually to prevent harm, not to redistribute loss that
has already occurred. So, the cornerstone for redistribution, which is fault, is not an essential
requirement.
RM v RB152
This case concerned an application for an interdict restraining the respondent from posting
defamatory postings about the applicant on her Facebook page. The applicant requested the
Court to order the respondent to:
• Remove the messages from her Facebook page
• Refrain from posting further defamatory statements about him on her Facebook
• Refrain from publishing defamatory statements about him in any other way.

The Court noted that some future defamatory statements about the applicant would not be
actionable if the contents were true and in the public interest. Restraining the respondent
from posting on Facebook would in such instances unduly limit the respondent’s right to
freedom of expression. However, if the respondent were to make derogatory or defamatory
statements about him in future, then the applicant would be entitled to approach a court for
relief in the form of an interdict or sue for damages.153
Accordingly, an interdict was not granted – although the Court did not expressly say so, for
want of proof that the future conduct would be wrongful.

After a court has calculated the amount of damages relevant in a particular case, it needs to
consider one further aspect, which is whether a particular defendant should be solely responsible, or
whether the responsibility for carrying the burden should be shared with someone else – the
plaintiff or another person. We discuss the principles that apply to reduction and apportionment of
damages in the next chapter.

34.5 Conclusion
• When a person suffers harm as a result of a delict committed by another, that person is entitled
to a remedy depending on the nature of the harm suffered.
• The remedies include an action for damages, an interdict, and an apology.
• The three most important actions in South African law are the actio legis Aquiliae (for
patrimonial loss), the actio iniuriarum, and the Germanic action (for non-patrimonial loss).
• Each of these actions has its own purpose. The purpose of the actio legis Aquiliae is complete
compensation, that of the actio iniuriarum is satisfaction, and in the case of the Germanic
action, the purpose is reparation. In the latter two instances, complete compensation is not
possible, because it is difficult to quantify non-patrimonial loss.
• The once-and-for-all rule means that a person has to claim all loss suffered as a result of a
delict (insofar as it is based on a single cause of action) in one lawsuit. This means that the
plaintiff has to claim damages for past and future losses in a single lawsuit.
• In the case of future losses (future loss of support, future medical expenses and loss of earning
capacity), courts have to make an educated guess to ascertain the exact harm that the plaintiff
will suffer in the future and then they have to quantify that harm into an award of damages by
considering various factors such as inflation, interest rates and contingencies.
• As a result of a delict a person may, in addition to suffering harm, also receive certain benefits,
such as insurance payouts. The question that arises in this regard is which of these benefits do
courts need to deduct from the damages claim to prevent the plaintiff from being over-
compensated, but on the other hand, to not let the defendant escape his or her responsibility to
compensate the plaintiff for the harm caused.
• The action used by the plaintiff and the rules that apply for quantifying the damages award
depend on the nature of the loss. In the case of patrimonial loss, the lex Aquilia applies, while
in the case of non-patrimonial loss, the plaintiff will use either the actio iniuriarum or the
Germanic action, depending on whether the harm was the result of an iniuria or a negligent
infringement of the physical-mental integrity.
• Each of these actions is subject to the general rules, such as the once-and-for-all rule and the
rule that relates to accounting for benefits. In addition, there are specific guidelines that our
courts have laid down regarding the quantification of damages for specific types of loss, such
as damage to property; harm that results from personal injury (in the case of the lex Aquilia);
losses that result from infringement of the body, reputation and dignity (in the case of the actio
iniuriarum); and pain and suffering and loss of amenities of life (in the case of the Germanic
action).
• In the case of defamation, the aggrieved party may also approach a court to demand an
apology from the wrongdoer.

1 [2014] 3 All SA 306 (SCA).

2 Para 42.
3 1991 (4) SA 23 (ZS).
4 Mutual & Federal Insurance Co Ltd v Swanepoel 1988 (2) SA 1 (A) at 10–11.
5 Neethling and Potgieter Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 224.
6 Van der Walt and Midgley Principles of Delict 4 ed (2016) para 192.
7 P.T.S., Inc., No. D-202 CV-93-02419, 1995 WL 360309 (Bernalillo County, N.M. Dist. Ct. Aug. 18, 1994).
 8 [2001] UKHL 29; [2002] 2 AC 122; [2001] 3 All ER 193.
9 1997 (3) SA 786 (CC) para 62.
10 Visser and Potgieter Visser and Potgieter’s Law of Damages 2 ed (2003) at 190.
11 Van der Walt and Midgley (2016) para 192.
12 Para 192.
13 1997 (3) SA 786 (CC).
14 Paras 70–72.
15 Refer to para 65 of the judgment and also to Van der Walt and Midgley (2016) paras 4 and 192.
16 Para 62.
17 The condictio furtiva is available where, for example, a person withdraws the thing from another’s possession or takes
and uses it while intending to restore possession after use. The remedy entitles the owner or possessor to claim the
highest value of the thing between the time it was stolen and litis contestatio.
18 Chetty v Italtile Ceramics Ltd 2013 (3) SA 374 (SCA) para 10.
19 2013 (3) SA 374 (SCA).
20 1980 (2) SA 814 (A). The meaning of ‘cause of action’ was confirmed in Duet and Magnum Financial Services CC
(In liquidation) v Koster 2010 (4) SA 499 (SCA) para 23.
21 At 838–839.
22 1980 (2) SA 814 (A).
23 See Southern Insurance Association Ltd v Bailey NO 1984 (1) SA 98 (A).
24 1967 (4) SA 188 (D).
25 1991 (3) SA 250 (D).
26 At 253H et seq.
27 1966 (3) SA 317 (A).
28 1973 (2) SA 146 (A).
29 1997 (1) SA 33 (A).
30 At 41C–H.
31 At 42A–B.
32 See Neethling and Potgieter (2015) at 239–240 for a comprehensive list.
33 See Neethling and Potgieter (2015) at 241–242 for a comprehensive list.
34 The Road Accident Fund v N F Timis (29/09) [2010] ZASCA 30 (26 March 2010). See also Mukheibir ‘Road
Accident Fund v Timis – Child Support Grants Not Res Inter Alios Acta’ (2011) 128(2) SALJ at 246.
35 See Mukheibir ‘Comparing the Casuistry of Compensating Advantages and Collateral Sources’ (2002) 23 Obiter at
328.
36 Visser and Potgieter (2003) at 83.
37 1987 (3) SA 577 (A).
38 At 612G–613D (our translation).
39 1991 (4) SA 23 (ZS).
40 [1955] 3 All ER 796 (HL).
41 1990 (4) SA 833 (A).
42 1989 (2) SA 173 (C).
43 1987 (3) SA 577 (A); 1990 (4) SA 833 (A) at 839F–G.
44 2013 (5) SA 356 (GSJ). See also Sil v Road Accident Fund 2013 (3) SA 402 (GSJ).
45 1994 (4) SA 747 (A).
46 At 776–777.
47 Visser and Potgieter (2003) at 23.
48 McGregor McGregor on Damages 17 ed (2003) para 35.107.
49 McGregor (2003) para 35.107.
50 1991 (4) SA 23 (ZS) at 34F–G.
51 1991 (4) SA 23 (ZS).
52 1979 (3) SA 953 (A).
53 Constantia Versekeringsmaatskappy Bpk v Victor NO 1986 (1) SA 601 (A).
54 2017 (4) SA 461 (GP).
55 Para 13.
56 2015 (4) SA 1 (CC).
57 Makhuvela v Road Accident Fund 2010 (1) SA 29 (GSJ) and Road Accident Fund v Timis (29/09) [2010] ZASCA 30
(26 March 2010).
58 Para 57.
59 Hentiq 1320 (Pty) Ltd v Mediterranean Shipping Co 2012 (6) SA 88 (SCA).
60 The ‘positive interesse’ formula is used when damages for breach of contract are assessed.
61 Visser and Potgieter (2003) at 72.
62 For detailed criticism of this rule see Visser and Potgieter (2003) at 69–73.
 63 1973 (2) SA 146 (A) at 150.
64 2005 (1) SA 299 (SCA) para 15.
65 1976 (2) SA 111 (C).
66 At 118G–H.
67 1969 (2) SA 1 (A).
68 At 18E–G.
69 This is contemplated in section 17(1A) of the Act. The Road Accident Fund Amendment Act 19 of 2005 introduced
this amendment which took effect on 1 August 2008. The prescribed method is found in regulation 3 of the Road
Accident Fund Regulations of 2008, which requires the third party to submit to an assessment by a medical
practitioner.
70 [2013] 1 All SA 543 (SCA).
71 Para 19.
72 Para 19.
73 See Paterson NO v Road Accident Fund 2013 (2) SA 455 (ECP) paras 11–12.
74 Visser and Potgieter (2003) at 402.
75 See section 34.2.7.6 below for the meaning of these terms.
76 See section 34.2.6.4 above.
77 1984 (1) SA 98 (A).
78 At 113–114.
79 1992 (1) SA 1 (A).
80 1992 (1) SA 1 (A) at 5E–G.
81 Terblanche v Minister of Safety and Security 2016 (2) SA 109 (SCA).
82 [2012] 4 All SA 26 (SCA).
83 Para 14.
84 Paras 13–14.
85 Visser and Potgieter (2003) at 409–410.
86 1984 (1) SA 98 (A).
87 1974 (4) SA 906 (A).
88 Para 199.
89 At 407.
90 Boberg The Law of Delict, Vol 1: Aquilian Liability (1984) at 538.
91 1974 (4) SA 906 (A).
92 2014 (1) SA 357 (WCC).
93 Para 42.
94 Para 42.
95 Para 44.
96 1985 (4) SA 843 (A).
97 Ferguson v Santam Insurance Ltd 1985 (1) SA 207 (C) at 208G–209A.
98 Santam Insurance Ltd v Ferguson 1985 (4) SA 843 (A) at 851D–G.
99 1949 (3) SA 1228 (T).
100 At 1237–1238.
101 1994 (4) SA 1 (A).
102 [2004] 2 All SA 565 (SCA) paras 60–61 (our translation).
103 1971 (1) SA 530 (A) at 535–536.
104 Sandler v Wholesale Coal Suppliers Ltd 1941 AD 194 at 199.
105 Gibson v Berkowitz 1996 (4) SA 1029 (W) at 1057; Van Zijl v Hoogenhout 2005 (2) SA 93 (SCA).
106 1994 (1) SA 862 (C).
107 1988 (3) SA 275 (A).
108 Administrator-General, South West Africa v Kriel 1988 (3) SA 275 (A) at 286–289.
109 1991 (1) SA 756 (A).
110 At 786A–E.
111 1966 (3) SA 484 (W) at 494F–H.
112 1991 (3) SA 412 (W).
113 1995 (4) SA 73 (C).
114 1964 (4) SA 531 (D).
115 Reyneke v Mutual and Federal Insurance Co Ltd 1991 (3) SA 412 (W) at 425–426.
116 1989 (2) SA 813 (V).
117 M v N 1981 (1) SA 136 (TkS).
118 Bridgman NO v Witzenberg Municipality (JL and Another as third parties) [2017] 1 All SA 466 (WCC) para 218.
119 Bridgman NO v Witzenberg Municipality (JL and Another as third parties) [2017] 1 All SA 466 (WCC) para 223.
120 2005 (2) SA 93 (SCA).
121 Paras 13–16.
122 Ramakulukusha v Commander, Venda National Force 1989 (2) SA 813 (V); Thandani v Minister of Law and Order
1991 (1) SA 702 (E); Mthimkhulu v Minister of Law and Order 1993 (3) SA 432 (E).
123 2006 (6) SA 320 (SCA).
124 Para 14.
125 Para 17.
126 Brenner v Botha 1956 (3) SA 257 (T); Kidson v SA Associated Newspapers Ltd 1957 (3) SA 461 (W); Bennett v
Minister of Police 1980 (3) SA 24 (C); Ramakulukusha v Commander, Venda National Force 1989 (2) SA 813 (V);
Jansen van Vuuren v Kruger 1993 (4) SA 842 (A).
127 DE v RH 2015 (5) SA 83 (CC).
128 Para 55.
129 Muller v SA Associated Newspapers Ltd 1972 (2) SA 589 (C) at 595A; Smith v Die Republikein (Edms) Bpk 1989 (3)
SA 872 (SWA); Afrika v Metzler 1997 (4) SA 531 (NM).
130 1968 (2) SA 545 (A).
131 [2005] 4 All SA 26 (C).
132 Paras 14 and 15.
133 1991 (4) SA 677 (N).
134 At 687A.
135 Hare v White (1865) 1 Roscoe 246 at 247.
136 2002 (6) SA 512 (W).
137 Para 24.
138 Para 28.
139 2004 (3) SA 46 (C).
140 At 57E–F.
141 2004 (3) SA 46 (C).
142 2006 (6) SA 235 (CC).
143 Para 111.
144 Paras 68 and 69.
145 2011 (3) SA 274 (CC).
146 Paras 197–202.
147 [2016] 4 All SA 723 (SCA).
148 Paras 29 and 35.
149 Paras 62–64.
150 Van der Walt and Midgley (2016) para 190.
151 Neethling, and Potgieter (2015) at 269.
152 2015 (1) SA 270 (KZP).
153 Paras 27–28.
 Chapter 35

Reduction and apportionment of

damages

35.1 Introduction

35.2 Apportionment between plaintiff and defendant: Contributory fault

35.2.1 The statutory provision
35.2.2 How are damages apportioned?
35.2.2.1 Approach 1
35.2.2.2 Approach 2
35.2.2.3 Approach 3
35.2.2.4 Approach 4
35.2.3 Contributory negligence or contributory fault?
35.2.4 Children’s contributory fault
35.2.5 Contributory fault and breach of contract

35.3 Apportionment between defendants: Joint wrongdoers

35.3.1 The statutory provision
35.3.2 Spouses
35.3.3 Claims by dependants
35.3.4 Claims by breadwinners

35.4 Apportionment between more than one plaintiff and more than one
defendant

35.5 Conclusion

35.1 Introduction
Where a defendant has committed a delict, but another person did something that contributed to the
plaintiff’s harm, then the defendant can claim that the amount required to compensate the plaintiff
should be shared between those that contributed to the harm. So, once a court has determined and
quantified the damages, the question that arises is whether one person (the defendant) should bear
the full loss or whether others (the plaintiff or other defendants) should share the burden. The co-
contributors could be any of the following:
• The plaintiff: In such an instance a court reduces the amount of damages it awards to the
plaintiff. This situation involves contributory fault.
• Another defendant: In such an instance the defendants share liability for the damages. This is a
case of joint wrongdoers.
• The plaintiff(s) and two or more defendants: In this instance a court reduces the amount it
awards to the plaintiff(s). The defendants share liability for the damages. This is a combination
of contributory fault and joint wrongdoer situations.

In each instance, a court apportions the damages so that the person who committed the delict pays
less than the full amount of compensation. Courts use the extent of each party’s fault as the
criterion to determine how to apportion the damages. After determining that there is a connection
between the other party’s conduct and the plaintiff’s harm, a court will look at the extent of the
parties’ fault and determine, in fairness to all concerned, to what extent the parties should share the
compensation amount.
Contributory fault used to be a complete defence against liability and where the plaintiff was
partially responsible for the harm, the defendant could escape liability entirely. This is no longer
the case and contributory fault does not negate the delict. It is now simply a device that courts use
for determining the extent to which compensation should be shared, after they have determined that
a delict exists.
Both the reduction of the award of damages due to the plaintiff’s contributory fault and the
sharing of responsibility for the compensation of harm between joint wrongdoers, are governed by
statutory rules, and not the common law.

35.2 Apportionment between plaintiff and defendant:

Contributory fault
The apportionment of damages between plaintiff and defendant is governed by the Apportionment
of Damages Act 34 of 1956.

35.2.1 The statutory provision

Where a defendant has committed a delict and a court can link the plaintiff’s conduct to the harm
that resulted, a court may reduce the amount of compensation that the defendant will ultimately pay
after considering the plaintiff’s fault. Section 1(1)(a) of the Apportionment of Damages Act
governs the reduction in the amount of damages:
Where any person suffers damage which is caused partly by his own fault and partly by the
fault of any other person, a claim in respect of that damage shall not be defeated by reason
of the fault of the claimant but the damages recoverable in respect thereof shall be reduced
by the court to such extent as the court may deem just and equitable having regard to the
degree in which the claimant was at fault in relation to the damage.

This section raises a number of issues:

• Apportionment is in respect of a plaintiff’s own fault, not that of another.
• The fact that a claim ‘shall not be defeated’ means that contributory fault is not a defence. The
defendant can use the plaintiff’s contributory fault only to claim a reduction in the amount of
damages, and not to escape liability completely.
• The reduction of damages is at a court’s discretion. A court reduces damages in accordance
with what it decides is ‘just and equitable’, that is, what it regards as fair and reasonable in the
circumstances.
• ‘Fault in relation to damage’ implies that there must be causal link between the plaintiff’s
conduct and the harm, and that the plaintiff must bear some of the blame for the result. Note
that the link is not to the event that caused the harm, but to the harm itself. So, the defendant
might have caused the accident, and the plaintiff’s negligent conduct could be something that
bore no relation to the accident, yet the plaintiff’s conduct nonetheless contributed to the
plaintiff’s harm. Failing to wear a seat belt that results in head injuries, is a classic example, as
are the facts in King NO v Pearl Insurance Co Ltd,1 where the passenger failed to wear a crash
helmet and suffered severe head injuries when the scooter she was travelling on collided with a
motor vehicle. The defendant succeeded in claiming a reduction in damages on the basis that
the plaintiff had contributed to her harm by not wearing a helmet.
Road Accident Fund v Myhill NO2
A mother and her two young children had been injured in a motor vehicle collision. Her
attorney lodged a claim with the Road Accident Fund (RAF). The mother was not awarded
any compensation for her own injuries, but the RAF offered to settle in respect of the
children’s injuries. The Fund then deducted 30% from that amount for the mother’s alleged
contributory negligence. Years later a curator appointed to represent the minors applied to set
aside the settlement on the grounds that the amount of damages awarded was inadequate
and moreover, set-off of the claims for a debt owed by the parent was not allowed in law.
The Court set aside the settlement agreement and furthermore held that for set-off to
operate between two parties the general principle is that there should be reciprocal
indebtedness. If both debts were equal, this would lead to their mutual discharge; if they
were not equal, the larger would be reduced by the amount of the smaller. Accordingly, it
was impermissible to reduce the RAF’s liability to the minors by way of setting off the alleged
personal liability of their mother against the children’s claims.3
This matter was decided by applying principles of contract. Could the same result have
been achieved by applying section 1(1)(a) of the Apportionment of Damages Act?

35.2.2 How are damages apportioned?

A trial court enjoys a wide discretion to award what it considers to be fair and adequate
compensation to an injured party and this applies equally to the way in which damages are
apportioned. A court of appeal will interfere only where there is a striking disparity between what
the trial court has awarded and what this court considers ought to have been awarded.4
Over the years, courts have applied different approaches to reduce the harm in cases of
contributory fault. We discuss these approaches in the next sections.

35.2.2.1 Approach 1
According to the method adopted in South British Insurance Co Ltd v Smith 5 and AA Mutual
Insurance Association Ltd v Nomeka,6 the degree of fault attributed to the defendant depends on the
plaintiff’s degree of negligence. Thus, if a court finds that the plaintiff was 30% negligent, it will
regard the defendant as 70% negligent. Note that the respective degrees of fault of the plaintiff and
defendant will add up to 100. So, if the total amount of damages (after reduction for contingencies)
is R100 000, the plaintiff will only receive R70 000. The plaintiff would have to bear the loss in
respect of the balance.

35.2.2.2 Approach 2
According to the method adopted in Jones, NO v Santam Bpk,7 the degree of fault that a court
attributes to the defendant does not depend on the degree of fault attributed to the plaintiff. In
Jones, NO the Court held that it had to establish the ‘respective degrees of negligence’ of the
parties. To do this, a court measures each party’s conduct independently against that of the
hypothetical reasonable person (which is 100%) and assesses the extent to which each party has
deviated from that norm. So, for example, if a court finds that the plaintiff’s conduct deviated 30%
from the reasonable-person standard, it does not automatically follow that the degree of the
defendant’s deviation will be 70%. Instead, a court will then also compare the defendant’s conduct
to the hypothetical standard and measure the deviation independently from that of the plaintiff. The
ratio between the respective deviations will then establish the formula for calculating the reduction
in compensation. Assume that a court finds that the conduct of a plaintiff deviated by 30% from the
norm and that of a defendant by 50%. The ratio of the two deviations is 30:50, or simplified, 3:5.
Therefore, the plaintiff’s degree of fault is ⅜ and the defendant’s is ⅝ of the total amount of
damages that was calculated. The percentage fault attributed to the defendant is ⅝ × 100 = 62,5%,
and if the total amount of damages is R100 000, the plaintiff will receive R62 500.

35.2.2.3 Approach 3
In General Accident Versekeringsmaatskappy SA Bpk v Uijs NO 8 the Court held that section 1(1)
(a) does not regard the plaintiff’s degree of fault as the deciding factor and that courts should also
consider other factors. In this case, Van Huyssteen had been severely injured in a collision. The
driver of the vehicle, Stander, had caused the accident, but the plaintiff, a passenger, had not
fastened his seatbelt and as a result his injuries were far more serious than what they would have
been had he done so. The defendant wanted the plaintiff’s claim to be reduced by 50%. The Court
held: 9
Section 1(1)(a) however does not provide that a plaintiff’s damages has to be reduced in
relation to his fault, but to the degree which, taking the extent of the plaintiff’s fault into
consideration, it will be fair and equitable. And in a case such as the present fairness and
equity require that one considers the fact that the plaintiff in no way contributed to the
collision and that his fault was of a different nature than that of (the driver).

The Court compared the degree of the plaintiff’s fault with that of the defendant, and held them to
be equally negligent. However, it did not reduce the damages by 50%. The Court considered other
factors and reduced the damages by 33% instead.

35.2.2.4 Approach 4
If one scans the law reports there are many instances in which courts apportion damages without
having regard to any of the three approaches mentioned above. In most of the cases courts simply
stipulate a percentage by which they will reduce the damages – a gut feeling, as it were, without
any comparative analysis or any reasons. Courts tend to assess contributory fault according to what
they consider to be fair in particular instances, and the more technical approaches are reserved for
problematic cases.
 COUNTER Apportioning damages
POINT According to Van der Walt and Midgley,10 none of the approaches
referred to previously is correct, because in each one, courts compare
the degrees of negligence of the plaintiff and the defendant. They argue
that section 1(1)(a) does not stipulate such a comparison, and that courts
should make any reduction according to what is fair and reasonable, with
regard to the plaintiff’s fault alone. The fourth approach is similarly
incorrect because the determinations make no reference to the plaintiff’s
fault.
• Are courts correct in following a comparative approach?
• Should the extent of the defendant’s fault be a relevant factor in deciding what
is just and equitable?
• Which of the above approaches conforms closest to the Act’s requirements,
and why?

Section 17(4)(c) of the Road Accident Fund Act 56 of 1996 places a limitation (‘a cap’) on the
amount of compensation payable in claims for loss of income. The cap applies to the actual loss
that the plaintiff has suffered.11 The purpose of this cap is not to interfere with the traditional way of
calculating damages, but instead to limit the sum to be paid by the defendant. At issue in Nel v
Road Accident Fund 12 was whether the apportionment of liability regarding the merits had to be
deducted before or after the application of the limit. The Court confirmed that, regardless of what
the plaintiff’s actual loss might be, the cap guarantees that the defendant will not pay more than the
limit. This case confirms that the statutory limit on compensation was introduced primarily for
policy reasons of a budgetary nature.13

35.2.3 Contributory negligence or contributory fault?

The Apportionment of Damages Act uses the term ‘contributory negligence’, as do courts. The
question is whether the term ‘fault’ in the Act refers to negligence only, or whether it also includes
intent. In Greater Johannesburg Transitional Metropolitan Council v ABSA Bank Ltd t/a Volkskas
Bank 14 the Court held that ‘fault’ includes intent and that section 1(1)(a) of the Act applies where
both the plaintiff and the defendant have acted intentionally in causing the harm:15
In my view the word ‘fault’ and its Afrikaans counterpart ‘skuld’ clearly include dolus … .
It should be noted that I have to do with a situation of dolus on both sides since both the
plaintiff’s servant, Mr Thoms, and the defendant’s Mrs Van der Walt intentionally caused
the harm which befell the plaintiff. Thus I do not have to consider the case where the
plaintiff’s fault may be negligence and that of the defendant dolus, or where the plaintiff has
dolus and the defendant is merely negligent … . Where there is dolus on both sides there
appears to me to be no reason not to give effect to the ordinary meaning of the words ‘fault’
and ‘skuld’.

Nevertheless, the majority of court decisions only refer to negligence. In its report on
apportionment of damages, the South African Law Commission believes that fault should include
both intent and negligence.16

PAUSE FOR Contributory fault

REFLECTION Courts have little difficulty in apportioning cases where both parties are
negligent, and presumably if both parties acted intentionally the situation
would be no different. However, how should a court assess contributory
 fault where the fault of the parties is not the same? Could any of the
approaches we have described so far satisfactorily be applied?

35.2.4 Children’s contributory fault

In Jones, NO v Santam Bpk,17 a case dealing with a child’s contributory fault, the Court determined
that it had to establish each party’s degree of negligence independently. In this case, the plaintiff
was a nine-year-old girl who had been struck by a motor vehicle. The defendant claimed
contributory fault on the part of the child. The Court held that it had to determine her degree of
fault objectively:
[T]he degree of … fault had to be determined on the same basis as if the child were an adult.
That was so because any proper requirement of subjectivity on account of his youth in
relation to his negligence was satisfied in the enquiry as to whether he was culpae capax. The
ascertainment of the degree in which either party in an action such as this was at fault in
relation to the damage is purely an objective enquiry related to the extent to which each has
been shown to have deviated, as set out above, from the standard of the bonus paterfamilias.

Therefore, the Court did not consider the fact that the plaintiff was a child in reducing her damages.
In Weber v Santam Versekeringsmaatskappy Bpk 18 the Court found that the decision in Jones
was unduly harsh towards children. The Court said: 19
If it is found that a defendant has been negligent, but that the child plaintiff was accountable
and also acted negligently, the question arises as to what the degree of negligence of the child
was. Where the defendant ought to have realised that he was dealing with the careless
conduct of a child and that his negligence consisted therein that he inter alia did not take
this into consideration, it follows that his degree of negligence ought to be higher than that
of the child, otherwise the child will paradoxically be penalised for conduct which the
defendant ought to have foreseen.

Even if one therefore applies the mathematical comparison of degrees of negligence as was
prescribed in the Jones case, it would follow that the defendant is guilty of a higher degree of
negligence than the child. The generally unsatisfactory results following from a
mathematical comparison of degrees of negligence of a child and an adult has to be ascribed
to the fact that this comparison is too rigid. A comparison of blameworthiness would arrive
at a more satisfactory solution.

Although Weber followed the Jones approach in assessing the negligence of the child in terms of
the reasonable-person test, it adopted a less rigid approach in apportioning the damages.
In Eskom Holdings Ltd v Hendricks 20 the Court referred to both the Jones and Weber cases.
Although there was no reference to the method of apportionment, the Court said the following
regarding the decision in Weber: 21
Experience revealed, Jansen JA said … that although children may be able to distinguish
between right and wrong, they will often not be able to act in accordance with that
appreciation; they became so engrossed in their play that they become oblivious of other
considerations and act impulsively. The learned Judge accordingly warned against ‘placing
an old head on young shoulders’ … . From what has been said above, it is clear that the
enquiry must in each case be related to the particular conduct which gave rise to the loss
forming the subject-matter of the claim.

35.2.5 Contributory fault and breach of contract

Can one apply the Apportionment of Damages Act to breach of contract? What happens if a person
suffers harm as a result of breach of contract on the part of the other contracting party, but that
person has also contributed to this harm?
Thoroughbred Breeders’ Association of South Africa v Price Waterhouse22
The plaintiff sued the defendant, a firm of auditors, for damages resulting from breach of the
auditing agreement. The plaintiff claimed that the auditor had failed to detect certain
irregularities, such as the encashment of a promissory note belonging to the plaintiff and theft
of its proceeds, and the theft of several amounts of money that had not been deposited into
the plaintiff’s bank account. The thefts were committed by one M, who had been the plaintiff’s
financial manager. The defendant denied liability, and submitted that M had a previous
conviction, that the plaintiff knowingly employed him and that the plaintiff, therefore, was in
part responsible for its own loss. The Court a quo held that the plaintiff was 80% responsible
for its own loss, and accordingly reduced the amount of damages awarded to the plaintiff by
that percentage.
The Supreme Court of Appeal found the defendant liable and in addition held, in four
separate judgments, that the Apportionment of Damages Act did not apply to breach of
contract. Therefore, the defendant was liable for the entire amount of damages awarded to
the plaintiff. In this regard, Nienaber JA said:23

There is, I believe, for the reasons stated by him, a pressing need for legislative
intervention in a situation such as the present where the defendant’s breach of
contract is defined in terms of his negligent conduct but the plaintiff, by his own
carelessness, contributed to the ultimate harm. But having said that, I am afraid that I
have reluctantly come to the conclusion that this particular piece of legislation does not
fulfil that function. I state my reasons for saying so with a minimum of elaboration.

Nienaber JA proceeded to quote the relevant sections of the Act:24

The extraneous defence of culpa compensatio was known to the common law in the
law of delict but not in the law of contract. In the law of contract the claim of the
claimant would not have been ‘defeated’ by his own culpa. (Of course, it would have
been a defence available to a defendant, even in a contractual setting, if the claimant’s
carelessness was the sole cause of the loss – but that would ex hypothesi not have
been a case where the damage was caused ‘partly by his own fault and partly by the
fault of any other person’.) That remained the position at the time the Act was
promulgated in 1956. The intention of the Legislature as to the scope and range of the
Act must be determined in the light of the situation prevailing at the time it was
enacted. At that time the concepts of both contributory negligence and ‘last
opportunity’ were unknown to a claim based on breach of contract. That being so, it
seems to me to follow that the Act was designed to address and correct a particular
mischief that was identified as such within the law of delict; that it was confined to that
particular mischief; and that the corresponding problem that might arise within the law
of contract was never within the Legislature’s compass. The express wording used in
the Act does not fit a contractual claim. In my view, the comfort of the Act was
accordingly not available to PW in this case to counter or curtail TBA’s claim for
damages.

So, the Court held that the common-law defence of contributory fault did not apply to
contractual liability and, after applying the rules of interpretation of statutes, that the Act did
not apply to breach of contract.

35.3 Apportionment between defendants: Joint wrongdoers

The Apportionment of Damages Act also apportions damages between joint wrongdoers.

35.3.1 The statutory provision

Where more than one person is responsible for causing the plaintiff’s harm, all these persons are
responsible for compensating the harm, provided that a court has held them to have committed a
delict.
Historically, a distinction was drawn between joint wrongdoers and concurrent wrongdoers.
Joint wrongdoers were those who consciously cooperated to cause the same harm. Concurrent, or
independent wrongdoers, were persons who, by independent and wrongful conduct, contributed to
the same harmful consequence. At common law courts could hold concurrent and joint wrongdoers
jointly and severally liable. Concurrent wrongdoers had a right of recourse against each other. In
other words, if one of the concurrent wrongdoers paid all the damages, he or she could recover the
pro rata share of the damages from the other wrongdoers. Joint wrongdoers had no right of
recourse. This distinction is no longer made and the Apportionment of Damages Act refers to both
categories as joint wrongdoers. Section 2(1) of the Act provides:
Where it is alleged that two or more persons are jointly or severally liable in delict to a third
person (hereinafter referred to as the plaintiff) for the same damage, such persons
(hereinafter referred to as joint wrongdoers) may be sued in the same action.

In Nedcor Bank Ltd t/a Nedbank v Lloyd-Gray Lithographers (Pty) Ltd 25

the Supreme Court of
Appeal noted:
The distinction between joint and concurrent wrongdoers is, of course, now largely
academic in view of the provisions of the Act which recognise and regulate a right of
contribution between ‘joint wrongdoers’ who are so defined as to include both joint and
concurrent wrongdoers at common law.

The plaintiff may sue one or more of the wrongdoers in the same lawsuit. As the wrongdoers are
jointly and severally liable, the plaintiff can choose to sue one of them for all the damages, or to sue
them together in the same action. If one of the wrongdoers pays the entire amount of damages, or if
judgment has been given against one wrongdoer, the other wrongdoer(s) is absolved from liability.
The wrongdoer who has paid the full amount has a right of recourse against the other wrongdoers.
Where joint wrongdoers are not sued in the same action a defendant may, before pleadings have
closed, give notice in terms of section 2(2) of the Act to a joint wrongdoer, in which event the
defendant would be entitled to recover a contribution from such joint wrongdoer after having
settled the claim in full. If no notice is given, the right is forfeited.26

35.3.2 Spouses
Where a plaintiff’s spouse, as well as the defendant, commits a delict against the plaintiff, special
rules apply if the spouses are married in community of property.27 Before 1984, if the spouses were
married in community of property, the defendant would have been liable for all the damages. The
reason for this was that parties who are married in community of property share a common estate
and cannot be delictually liable to one another.
Section 19 of the Matrimonial Property Act 88 of 1984 changed this position. It reads:
When a spouse is liable for the payment of damages, including damages for non-patrimonial
loss, by reason of a delict committed by him or when a contribution is recoverable from a
spouse under the Apportionment of Damages Act, 1956 (Act No 34 of 1956), such damages
or contribution and any costs awarded against him are recoverable from the separate
 property, if any, of that spouse, and only in so far as he has no separate property, from the
joint estate: Provided that in so far as such damages, contribution or costs have been
recovered from the joint estate, an adjustment shall, upon the division of the joint estate, be
effected in favour of the other spouse or his estate, as the case may be.

Courts can now hold a spouse married in community of property liable for delicts committed
against the other spouse and therefore one spouse may also be a joint wrongdoer in respect of the
other spouse. The Matrimonial Property Act provides that in appropriate cases both spouses can be
held responsible for the harm and courts will apportion the damages between them in terms of the
Apportionment of Damages Act. For this purpose, courts consider the spouses to have separate
estates and the damages will not come from or be paid into any communal estate. If the joint
wrongdoer spouse has a separate estate already, he or she will pay the damages from that separate
estate. In the absence of such a separate estate, the wrongdoer spouse will pay the damages from
the joint estate. However, if the marriage is dissolved, the court will make an adjustment in favour
of the plaintiff or the plaintiff’s estate. In both instances, the wrongdoer will make the payment into
a separate estate that belongs solely to the plaintiff.
Where the parties are married out of community of property, each has a separate estate and
courts can therefore hold each one liable in delict towards the other. Therefore, such spouses can be
joint wrongdoers in respect of each other.

35.3.3 Claims by dependants

Where a dependant’s breadwinner as well as the defendant commits a delict against the dependant,
they are treated as joint wrongdoers. For example, the defendant and the breadwinner are involved
in an accident in which the dependant is injured, and both the breadwinner and the dependant are at
fault. It does not matter whether the breadwinner is alive or dead: when the dependant plaintiff sues
the defendant, a court may draw the breadwinner or the estate into the action as a joint wrongdoer,
and it will apportion damages accordingly.28

PAUSE FOR A dependant’s claim and right of support

REFLECTION If successful, how would the dependant’s claim affect the right of support
that other dependants have? For example, if the dependant has suffered
future harm and is awarded R100 000 in damages, but the parent was a
joint wrongdoer and responsible for R40 000 of that amount, does the
dependant have a prior claim over other siblings?

A second example is where the defendant and the breadwinner are involved in an accident in which
the breadwinner dies, and both were at fault, and the dependant plaintiff suffers harm in the form of
loss of support. Courts would consider the breadwinner a joint wrongdoer in respect of that claim
and so the dependant can sue either the deceased estate or the third party, or both. If either party is
successfully sued or pays the entire amount, that party has a right of recourse against the other
party.
If the plaintiff receives a benefit from the deceased breadwinner’s estate (an inheritance) a
court will deduct the accelerated benefit of the inheritance from the plaintiff’s claim for loss of
support. If the third party has paid the damages in full, the right of recourse is retained, provided
that the benefit received by the plaintiffis not affected. Section 2(6)(a) provides as follows:
If judgment is in any action given against any joint wrongdoer for the full amount of the
harm suffered by the plaintiff, the said joint wrongdoer may, if the judgment debt has been
 paid in full … recover from any other joint wrongdoer a contribution in respect of his
responsibility for such damage of such an amount as the court may deem just and equitable
having regard to the degree in which that other joint wrongdoer was at fault in relation to
the damage suffered by the plaintiff, and to the damages awarded: Provided further that if
the court, in determining the full amount of the damage suffered by the plaintiff referred to
in subsection (1B), deducts from the estimated value of the support of which the plaintiff has
been deprived by reason of the death of any person, the value of any benefit which the
plaintiff has acquired from the estate of such deceased person no contribution which the
said joint wrongdoer may so recover from the estate of the said deceased person shall
deprive the plaintiff of the said benefit or any portion thereof.

PAUSE FOR Section 2(6)(a)

REFLECTION If successful, how would the dependant’s claim affect the right of support
that other dependants have? For example, if the dependant has suffered
future harm and is awarded R100 000 in damages, but the parent was a
joint wrongdoer and responsible for R40 000 of that amount, does the
dependant have a prior claim over other siblings?
• What is the practical effect of section 2(6)(a) where a dependent child suffers
R120 000 in loss of support and the deceased breadwinner and the defendant
are joint wrongdoers, with damages to be apportioned in the ratio 1:1, and the
accelerated benefit of the inheritance received by the child is assessed to be
R20 000? Note:
◆ The accelerated benefit of inheritance must be deducted from the claim for
loss of support.
◆ If the dependant child sues the defendant, who in turn claims a contribution
from the estate of the breadwinner, the latter claim cannot exhaust the
estate: at least R20 000 (the value of the benefit) should be left to pass to
the child. This means that the contribution claim may not be fully satisfied if
less than R20 000 is available in the estate.

COUNTER Inability to earn future income

POINT A situation may arise in which the breadwinner does not die in the
accident, but is severely injured, resulting in an inability to earn any
income in the future. One way of looking at this situation is that the
dependant’s right of support has not been infringed in such an instance.
The right still exists against the breadwinner, in which case the
dependant has not suffered any harm. In this scenario the breadwinner
will claim against the defendant and a court may apportion damages
according to the principles of contributory fault. Another way is to see if
section 2(1B) of the Act applies. It provides:

… if it is alleged that the plaintiff has suffered harm as a result of

any injury to or the death of any person and that such injury or
death was caused partly by the fault of such injured or deceased
person and partly by the fault of any other person, such injured
person or the estate of such deceased person, as the case may
be, and such other person shall for the purposes of this section
be regarded as joint wrongdoers.

Consider:
• Which is the correct approach? Could both be correct, depending upon the
circumstances?
 • What if the dependant’s support is reduced because the father’s income post-
accident is far less than before because a court reduced his damages claim
substantially as a result of his contributory negligence? Should the dependant
be able to claim for that loss of support from the wrongdoer?

35.3.4 Claims by breadwinners

Where a breadwinner sues an outside party for expenses resulting from injuries to the
breadwinner’s child, and the child was also at fault, the child will be a joint wrongdoer in respect of
the breadwinner’s claim. The outside party has a right to claim a contribution from the child
proportionate to its contributory negligence.

35.4 Apportionment between more than one plaintiff and

more than one defendant
It may be that a court has to apportion damages between more than one plaintiff and between more
than one defendant. In such a case, a court is dealing with both contributory fault and joint
wrongdoers. It appears that there is only one reported case in our case law that has dealt pertinently
with such a situation.
Harrington NO v Transnet Ltd29
The plaintiffs, two security guards patrolling the railway line in the night, were struck by a train
on its way from Cape Town to Woodstock. They sustained serious bodily injuries as a result
of the accident and proceeded to sue Transnet, the owner of the property on which the
accident took place, as well as the train driver. The Court found that the two plaintiffs had
been negligent in that they had not kept a proper lookout for trains while they were walking
along the railway line. The Court, following the approach in some Commonwealth countries,
compared the plaintiffs’ conduct to that of the defendants and, after having reduced the
damages in proportion to the plaintiff’s fault, apportioned the remainder between the
defendants.30 In this instance, the blameworthiness of each of the plaintiffs was ‘substantially
outweighed by the aggregate of the blameworthiness of Transnet and the train driver’, and
there were ‘no specific considerations of equity or justice, other than the comparison in
respect of blameworthiness’ that impacted on the issue.31 The Court accordingly reduced
each of the plaintiffs’ damages by one-third and held the two wrongdoers jointly and severally
liable for the balance.
The matter went on appeal, first to a full bench of the Cape High Court, and thereafter to
the Supreme Court of Appeal.32 The upshot of these appeals was that neither the train driver
nor the plaintiffs were found to have been negligent. These findings were made on the
evidence that was presented and it appears that the trial court’s reasoning regarding the
principles to be applied in such cases remains valid.

35.5 Conclusion
In certain instances, a defendant is not responsible for the entire amount of damages, because
someone else contributed to the harm. If the plaintiff contributed to the harm, a court will reduce
the amount of damages in accordance with the plaintiff’s degree of fault. If the co-contributor is
one of the defendants, the defendants share the responsibility of paying the damages. When both a
plaintiff and a defendant contributed to the harm, a court will reduce the amount of damages and
the defendants will have to share responsibility for this reduced amount between them. In all these
cases, courts carry out the apportionment in terms of the Apportionment of Damages Act.
• In certain instances, the defendant is not solely responsible for damages, because someone else
contributed to that harm.
• The co-contributor could be any of the following:
◆ The plaintiff
◆ A defendant
◆ The plaintiff(s) and two or more defendants.
• Where the plaintiff has contributed to the harm, one is dealing with contributory fault. The
defendant will not escape harm completely. Instead, a court will reduce the amount of damages
awarded to the plaintiff.
• Where a defendant (or more than one defendant) has contributed to the harm, one is dealing
with joint wrongdoers. In this case, the defendants share the responsibility for damages and
they will be jointly and severally liable for compensating the plaintiff.
• Where a plaintiff (or more than one plaintiff) and two or more defendants contribute to the
harm, one is dealing with both contributory fault and joint wrongdoers. In this case, a court
will reduce the amount awarded to the plaintiff(s), which will be shared between the
defendants.

1 1970 (1) SA 462 (W).

2 2013 (5) SA 426 (SCA).
3 Para 23.
4 Protea Assurance Co Ltd v Lamb 1971 (1) SA 530 (A) at 535A–B. See also Long v Jacobs (145/11) [2012] ZASCA
58 (2 April 2012) para 23; Tyco International (Pty) Ltd v Golden Mile Trading 547 CC (949/2013) [2016] ZASCA 44
(31 March 2016) paras 12 and 20.
5 1962 (3) SA 826 (A).
6 1976 (3) SA 45 (A).
7 1965 (2) SA 542 (A).
8 1993 (4) SA 228 (A).
9 At 235D–E (our translation).
10 Para 218.
11 The term ‘actual loss’ is not defined in the Act, however the court considers this as being damages calculated in the
normal traditional way after application of the relevant apportionment of damages for contributory negligence.
12 2016 (2) SA 517 (GP).
13 Para 7. See also Law Society of South Africa v Minister for Transport 2011 (1) SA 400 (CC) para 86.
14 1997 (2) SA 591 (W).
15 At 606F–I.
16 South African Law Reform Commission Project 96 the Apportionment of Damages Act 34 of 1956 Report (2003).
17 1965 (2) SA 542 (A) at 556A–C.
18 1983 (1) SA 381 (A).
19 At 401A–B (our translation).
20 2005 (5) SA 503 (SCA).
21 Paras 17–19.
22 2001 (4) SA 551 (SCA).
23 Para 72.
24 Para 74.
25 2000 (4) SA 915 (SCA) at 922H–I.
26 Lategan v Jansen Van Rensburg (1499/2012) [2016] ZAECGHC 69 (8 September 2016) paras 35–37.
27 See section 2(1A) of the Apportionment of Damages Act.
28 See section 2(1B) of the Apportionment of Damages Act.
29 2007 (2) SA 228 (C).
30 Para 91.
31 Para 92.
32 Harrington NO v Transnet Ltd t/a Metrorail 2010 (2) SA 479 (SCA). The full bench decision is not reported.
 Chapter 36

Legal and public policy considerations

that have justified the statutory
development of the law of delict

36.1 Introduction

36.2 Legal and public policy considerations that have justified the
statutory development of the law of delict
36.2.1 The need to combat the risk of receiving no compensation
36.2.2 The role of the Constitution and the promotion of the constitutional right to
social security
36.2.3 The evidentiary difficulties with proving fault
36.2.4 General time and cost-related problems with the civil procedural system
36.2.5 The ability of the legislature to regulate liability more comprehensively than
the judiciary
36.2.6 The need to prevent arbitrary outcomes
36.3 Conclusion

36.1 Introduction
There are many instances where the law of delict has been developed by legislation enacted to
compensate a variety of victims of harm. In this chapter, attention will be given to the legal and
public policy considerations that have justified the legislature’s decision to supplement the
compensatory response provided by the common-law principles of the law of delict.
The statutory compensation of harm should be distinguished from compensation payable as a
result of the breach of a statutory duty. For a discussion of the legal position relating to the latter,
see Chapter 20 above.
36.2 Legal and public policy considerations that have
justified the statutory development of the law of delict
In this chapter, the spotlight falls on the policy considerations that have justified the statutory
development of the law of delict through the enactment of the following statutes:
• The Compensation for Occupational Injuries and Diseases Act 130 of 1993 (‘COIDA’)
• The Road Accident Fund Act 56 of 1996 (‘RAF Act’), as amended by the Road Accident Fund
Amendment Act 19 of 2005 (‘RAFA Act’) and the proposed Road Accident Benefit Scheme
(‘RABS’)
• The Consumer Protection Act 68 of 2008 (‘CPA’).

Although there are several other pieces of legislation that have a significant impact on the law of
delict,1 the focus is placed on these statutes because they predominantly deal with the compensation
of specific groups of victims: victims of occupational injuries and diseases, motor vehicle accident
victims and those who suffer harm arising from defective consumer products. In some way or
another, all of these statutes have therefore singled out a collection of individuals for preferential
treatment while aligning themselves with the primary function of the law of delict, that is, the
compensation of harm.2

PAUSE FOR Are there any other areas of the law of delict that may benefit from
REFLECTION legislative intervention similar to the areas outlined above? Could the
considerations outlined below provide a theoretical framework on which
the potential future statutory development of the law of delict may
justifiably be based?

36.2.1 The need to combat the risk of receiving no compensation

The existence and extent of a risk of harm has played an important role in the legislature’s decision
to develop the South African law of delict. This has especially been the case in the context of
occupational injuries and diseases, motor vehicle accidents and defective consumer products, where
the growth in labour activities, expanded motor vehicle traffic and the rise of manufactured
consumer goods produced an increase of two types of risk. First, it brought about the significantly
higher risk of harm to road users, employees and consumers. Under the common law, when such
risk materialises as the result of the culpable conduct of another, the victim of the accident may
institute a common-law delictual claim against the wrongdoer in search of compensation.
Wrongdoers, however, are often unable to pay any or all of the damages required to repair the
victim’s harm.3 Therefore, the risk of harm is accompanied by a further risk of receiving no
compensation in the event of the harm materialising.
To protect road users from the realisation of the risk, the legislature decided to intervene by
enacting motor vehicle accident legislation. The RAF Act is the latest statute in a long line of
national legislation, beginning with the Motor Vehicle Insurance Act 29 of 1942. The primary
purpose underlying these statutes may be described as follows: 4
The obvious evil that [motor vehicle accident statutes like the RAF Act are] designed to
remedy is that members of the public who are injured, and the dependants of those who are
killed, through the negligent driving of motor vehicles may find themselves without redress
against the wrongdoer. If the driver of the motor vehicle or his master is without means and
 is uninsured, the person who has been injured or his dependants, if he has been killed, are in
fact remediless and are compelled to bear the loss themselves.

During the course of the twentieth century, more and more South Africans became employed in a
variety of sectors. Similar to the dramatic increase of motor vehicle traffic during that period, the
rise in labour activity was accompanied by a risk of harm to employees, who were often required to
endanger their well-being by confronting hazardous environments and taking on precarious
challenges for the gain of their employers. Again, this risk was accompanied by the risk of
receiving no compensation in the event of the materialisation of an employment related risk. This
consideration motivated the legislature to take action by enacting the COIDA as well as its
antecedent legislation.
The design, manufacture, distribution and sale of products are, generally, central to the wealth
and welfare of any society, but could also bring about disease, injury and even death for a wide
range of individuals.5 The rise of industrialisation in the nineteenth century and consumerism in the
twentieth century led to a substantial increase in the manufacturing and distribution of consumer
products.6 This meant that, more than ever before, consumers were being exposed to an unremitting
series of manufactured goods. Because technology grew more sophisticated and often coupled with
high levels of expertise, consumers knew very little about the products that reached them. It is
unsurprising that many of these products posed a significant risk to the well-being of consumers
who chose to make use of them and a further risk of receiving no compensation in the event of
harm resulting from the particular risk.7 In response to the rise in consumer products, the growing
risk of exposure to harm and the difficulty of holding manufacturers liable for the harm suffered by
consumers as a result of defective products, the legislature introduced a strict liability regime for
harm suffered as a result of defective products when it enacted the CPA,8 which is described in
further detail in Chapter 39.

36.2.2 The role of the Constitution and the promotion of the constitutional
right to social security
Chapter 2 dealt specifically with the general relationship between the law of delict and the
Constitution of the Republic of South Africa, 1996. From a general perspective, it is sufficient to
emphasise here the fact that the Constitution is the supreme law of the country, central to the
country’s legal system and determines the validity of all law, including the law of delict.9
More specifically, section 27(1)(c) of the Constitution entrenches the right to social security
while section 27(2) imposes upon the State a mandatory duty to take reasonable legislative and
other measures, within its available resources, to achieve the progressive realisation of each of
these rights.
The COIDA and the RAF Act are both regarded as legislation that promotes the right to social
security. In Law Society of South Africa v Minister of Transport 10 the Constitutional Court
described the RAF Act as follows:11
It seems plain that the scheme arose out of the social responsibility of the State. In effect, it
was, and indeed still remains, part of the social security net for all road users and their
dependants.

In the same judgment, the Court referred also to the proposed RABS and noted that the ultimate
vision appears to be the establishment of a system of compensation for road accident victims that:12
must be integrated into a comprehensive social security system that offers life, disability and
health insurance cover for all accidents and diseases.
The COIDA introduced significant changes in respect of the protection of employees’ rights and,
although it did not intend to provide a kind of general health cover for every accident or disease
which an employee may suffer from, it may nevertheless be regarded as social security legislation,
aimed at the provision of a more equitable compensation dispensation in regard to injuries suffered
and diseases contracted by employees. Where earlier legislation was based on the principle of
individual employer liability as covered by private insurance, the COIDA introduced the principle
of no-fault based liability and limited benefits covered by a public scheme.13 The introduction of
such a scheme, which does not require an injured or diseased employee to prove fault on the part of
the employer, weakens the likelihood of lengthy and costly legal disputes and provides a more
streamlined administrative process for the effective compensation of injured employees. As such,
this piece of ‘social legislation’, 14 like most statutory occupational injury and disease schemes
enacted in foreign jurisdictions, promotes the social and economic welfare of employees.

36.2.3 The evidentiary difficulties with proving fault

The evidentiary difficulties that are involved with proving fault has motivated the legislature to
develop the law of delict by abolishing the fault requirement in specific instances.
In the context of motor vehicle accidents, the Road Accident Fund Commission (the ‘RAFC’),
tasked with conducting a broad inquiry into the statutory compensation system under the RAF Act,
noted that it: 15
is increasingly felt that fault cannot really be determined accurately and there is also a
growing social concern for accident victims regardless of the role they played in causing the
accident.

Following the publication of a report by the RAFC in 2002, the legislature has proposed the RABS,
which will abolish fault as requirement for liability in the context of motor vehicle accidents. It is
envisaged that the proposed no-fault model will ease the administrative load regarding the process
of statutory claims, increase the speed with which those claims are processed and prevent lengthy,
costly legal disputes concerning the existence of negligence.16
This evidentiary difficulty was also an important consideration underlying the enactment of the
COIDA. When the proposed COIDA was tabled for discussion in Parliament, the then Minister of
Manpower remarked: 17
Under common law an injured employee or the dependents of a deceased employee may get
compensation from his employer if it can be proved that the injury or death was due to the
negligence of the employer, but in a modern industrial set-up in which, for example, a
number of employees jointly use sophisticated machinery, it may be virtually impossible for
an injured employee to prove negligence.

With the introduction of the COIDA, which obviates the need to prove fault within this context, the
employee is therefore able to obtain compensation much easier and quicker from a solvent entity.18
This evidentiary difficulty was also a major policy consideration in favour of statutory
intervention in the field of product liability, where the consumer is usually unable effectively to
scrutinise products for safety.19 In Wagener v Pharmacare Ltd; Cuttings v Pharmacare Ltd 20 the
Supreme Court of Appeal (‘SCA’) was requested to develop the rules of the common law of delict
so that it was no longer required for victims of defective products to prove that the manufacturer
had been culpable in manufacturing the product in question. Although the Court ultimately opted to
leave the development of this branch of the law to the legislature, it took cognisance of, and seemed
in agreement with, this criticism: 21
 A plaintiff has no knowledge of, or access to the manufacturing process, either to determine
its workings generally or, more particularly, to establish negligence in relation to the
making of the item or substance which has apparently caused the injury complained of.
And, contrary to what some writers suggest, it was urged that it is insufficient to overcome
the problem that the fact of the injury, consequent upon use of the product as prescribed or
directed, brings the maxim res ipsa loquitur into play and casts on the defendant a duty to
lead evidence or risk having judgment given against it. The submission is that resort to the
maxim is but a hypocritical ruse to justify (unwarranted) adherence to the fault
requirement.

36.2.4 General time and cost-related problems with the civil procedural
system
As mentioned in Chapter 1, the law of delict may be said to be ineffective in its principal aim of
compensating harm arising from especially personal injury and death.22 Obtaining compensation by
instituting civil proceedings has been described as slow, cumbersome and expensive. 23
The statistics quoted from royal reports and academic writing in Chapter 1 provide convincing
evidence that, from a comparative perspective, the English law of torts plays an underwhelming
role in the compensation of harm victims.
Although there are no up-to-date statistics to put alongside those provided above, it has been
argued that ‘there is little reason to think that the basic picture is significantly different now’. 24
Further, as suggested in Chapter 1, although figures for South Africa are not known, they are likely
to follow similar trends.
Instituting civil litigation in South Africa is expensive and only a limited number of people can
afford the accompanying legal transaction costs, thereby restricting the general access to justice.
The enactment of the COIDA, the RAF Act and the CPA is a legislative attempt to enhance access
to justice. The COIDA and the RAF Act established statutory compensation funds and adopt a
quicker and more cost-efficient administrative process towards compensation. The CPA also
contributes to the reduction of transaction costs linked to civil litigation by the removal of the
requirement to prove fault, which generally may involve greater amounts of time, money and
technical resources.

36.2.5 The ability of the legislature to regulate liability more

comprehensively than the judiciary
In Wagener v Pharmacare Ltd; Cuttings v Pharmacare Ltd,25 the SCA took account of the
arguments in favour of the introduction of a strict liability regime for harm caused by defective
consumer products. The Court noted that product liability reform in foreign jurisdictions had
largely been achieved through legislation and ultimately concluded that South Africa should adopt
the same route: 26
[i]f strict liability is to be imposed, it is the Legislature that must do it.

In its judgment it held that the legislature was better equipped to investigate the variety of questions
and concerns that would have to be addressed prior to introducing a strict liability regime in the
context of defective products.
The SCA held that single instances of litigation could not provide the opportunity for
conducting the thorough investigation, analysis and determination that was necessary to produce a
cohesive and effective structure by which to impose strict liability.27 The Court’s recommendation
was heeded and the legislature, with the benefit of more empirical data, more time and more
product liability expertise, enacted the CPA.28
36.2.6 The need to prevent arbitrary outcomes
Some scholars argue that the delict/tort system is essentially a ‘lottery’ 29 and that it ‘produces
arbitrary outcomes.’ 30 They contend that the imposition of delictual/tortious liability and the
subsequent payment of damages are based on ‘considerations unrelated to what the parties
deserve.’ 31 Instead, the outcome of litigation is seemingly determined by contingent factors,
including the availability of evidence, the quality of counsel, the limits of insurance coverage, the
financing of litigation, the whims of judges and juries, the identities of the respective litigants and
many other factors that are not conducive to the consistent and principled application of law.32 The
argument that the delict/tort system is both unfair and unpredictable have been advanced to justify
reform proposals that would supplement the delict/tort system in some way or another in order to
provide fairer and more reliable compensation to accident victims.

36.3 Conclusion
The South African law of delict has seen significant development relating to the compensation of
motor vehicle accident victims, victims of occupational injuries and diseases and those who suffer
harm arising from defective consumer products, through the enactment of the RAF Act, the COIDA
and the CPA. The enactment of these statutes have been justified by similar legal and public policy
considerations.
In future, if the legislature were to consider the further legislative development of the law of
delict insofar as the compensation of a single category of victims is concerned, attention may be
given to the considerations that have been identified in this chapter. It may provide the lawmakers
with a general theoretical basis against which to determine if future statutory endeavors could be
justified on similar grounds. Obviously, this is important because all statutes require justification,
especially if they are to entail a supplementation of the existing common-law compensatory regime.
In addition, it could provide a justifiable policy basis so that the purpose and scope of the relevant
statute is clear and all affected parties are provided with an underlying rationale on the basis of
which the legislative provisions may be interpreted.

1 For example, the Apportionment of Damages Act 34 of 1956, which was described as being the ‘most important piece
of law reform that has been carried out in the field of private law since Union’ – see McKerron (1956) The
Apportionment of Damages Act at 1.
2 For a discussion of the functions of the law of delict, see Macintosh Negligence in Delict 1 ed (1926) at 1; Van den
Heever, Aquilian Damages in South African law (1944) at 3; Van den Heever (1944) at vii; McKerron The Law of
Delict: A Treatise on the Principles of Liability for Civil Wrongs in the Law of South Africa 7 ed (1971); Van der
Merwe and Olivier Die onregmatige daad in die Suid-Afrikaanse reg 6 ed (1989) at 1–3; Neethling and Potgieter
Neethling-Potgieter-Visser Law of Delict 7 ed (2015) at 3; Van der Walt and Midgley Principles of Delict 3 ed
(2016); Chapter 1 above.
3 For example, with regard to wrongdoers who have culpably caused motor vehicle accidents, Moseneke DCJ remarked
in Law Society of South Africa v Minister of Transport 2011 (1) SA 400 (CC) para 50 that:
in his view, the number of drivers and owners who would be able to pay would be very small.
4 Aetna Insurance Co v Minister of Justice 1960 (3) SA 273 (A) at 285. This dictum has been confirmed and applied in
a series of motor vehicle accident cases: see RAF v Makwetlane 2005 (4) SA 51 (SCA) para 7; Engelbrecht v RAF
2007 (6) SA 96 (CC) para 23; Nel v RAF 2016 (2) SA 517 (GP) para 10.
5 Van Eeden Consumer Protection Law in South Africa (2013) at 367.
6 Stapleton Product Liability (1994) at 9–16.
7 Van Eeden (2013) at 370.
8 The Act came into effect in 2010.
9 Section 2 of the Constitution.
10 2011 (1) SA 400 (CC).
11 Law Society of South Africa v Minister for Transport 2011 (1) SA 400 (CC) para 17.
12 Paras 45–46.
13 Olivier ‘Social Security: Framework’ (2012) LAWSA 13(2) para 9.
14 In Molefe v Compensation Commissioner (25579/05) [2007] ZAGPHC 365 (25 September 2007) para 5, Seriti J
found that the:
Compensation for Occupational Injuries and Diseases Act … is a social legislation and according to
section 39(2) of the Constitution, it must be interpreted in such a manner that the said interpretation
promotes the spirit, purport and objects of the social security right as enshrined in section 27 (l)(c) of
the Constitution.
15 Satchwell Road Accident Fund (South Africa) Commission (2002) at 119.
16 Minister of Transport Policy Paper (2011) at 5.
17 South Africa Proceedings of the Extended Public Committee Debates of Parliament Hansard 1993-1994 (1994) at
12305.
18 For a German perspective on this point, see Markesinis and Unberath The German Law of Torts (2002) at 727.
19 Loubser and Reid Product Liability (2012) at 4.
20 2003 (4) SA 285 (SCA).
21 Para 10.
22 Cane Atiyah’s Accidents, Compensation and the Law 8 ed (2013) at 459–495; Deakin, Johnston and Markesinis Tort
Law 7 ed (2013) at 51–59; Hedley ‘Tort and Personal Injuries, 1850 to present’ in Arvind and Steele (Eds) Tort Law
and the Legislature (2013) at 235–249.
23 Ison The Forensic Lottery (1967); Elliot and Street Road Accidents (1968); Atiyah Accidents, Compensation and the
Law (1970); Cane (2013) at 459–495; Deakin, Johnston and Markesinis (2013) at 51–59.
24 Cane (2013) at 19–21.
25 2003 (4) SA 285 (SCA).
26 Para 38.
27 Para 37.
28 From a comparative perspective, see also Cambridge Water Co Ltd v Eastern Counties Leather Plc [1994] 2 AC 264
at 305:
I incline to the opinion that, as a general rule, it is more appropriate for strict liability in respect of
operations of high risk to be imposed by Parliament, than by the courts.
29 Sugarman ‘Serious Tort Law Reform’ (1987) 24(4) San Diego Law Review 795 at 795–796.
30 Franklin ‘Replacing the Negligence Lottery: Compensation and Selective Reimbursement’ (1967) 53(4) Va. L. Rev.
774 at 778:
[T]he fault system is little more than an immoral lottery for both plaintiffs and defendants. See also
Atiyah The Damages Lottery (1997) at 143.
31 Lytton, Rabin and Schuck ‘Tort as a litigation lottery: a misconceived metaphor’ (2011) Vol 52 Boston College Law
Review267 at 268.
32 Lytton, Rabin and Schuck (2010) at 268–269.
 Chapter 37

The Compensation for Occupational

Injuries and Diseases Act 130 of 1993

37.1 Introduction

37.2 Position under the COIDA

37.3 Section 35

37.1 Introduction
Under the common law, it is well established that an employer owes its employees a duty to:1
take reasonable care for their safety … and an obligation to take reasonable steps to protect
them from physical harm. |
Therefore, employees who suffered harm arising from an accident that occurred in their workplace
could claim damages in delict from the employer if they could prove that the employer culpably
and wrongfully caused their harm. If the employee was contributorily negligent, the damages
payable would be reduced in accordance with the Apportionment of Damages Act 34 of 1956.
Furthermore, employees could only obtain compensation if they had the time and means to resort to
expensive and time-consuming litigation and the employer had the financial means to pay
compensation.
Since 1886 a series of statutes have therefore been enacted to improve the position of
employees who had suffered harm as a result of their employment. The Compensation for
Occupational Injuries and Diseases Act 130 of 1993 (‘COIDA’) repealed the Workmen’s
Compensation Act of 1941 and came into operation in 1994. The COIDA replaces the common-law
position by abolishing the employee’s common-law delictual claim against his or her employer and
introducing a compensation system that allows employees to obtain limited compensation from a
statutory compensation fund to which employers are obliged to contribute. The system introduced
by the COIDA is no-fault based and compensation is therefore payable regardless of whether the
injury or disease was caused by negligence on the part of the employer or anyone else.
The purpose of the COIDA is to provide compensation to employees for harm arising from
occupational injuries or diseases sustained or contracted by employees in the course of their
employment. The enactment of this statute, like its predecessors, has amounted to the statutory
development of the law of delict and it has been justified on the basis of the legal and public policy
considerations outlined in Chapter 35. The Constitutional Court has described this development as
follows: 2
The [COIDA] supplants the essentially individualistic common-law position, typically
represented by civil claims of a plaintiff employee against a negligent defendant employer,
by a system which is intended to and does enable employees to obtain limited compensation
from a fund to which employers are obliged to contribute.

37.2 Position under the COIDA

The operation of the COIDA can be summarised as follows. An employee who is harmed in the
course of his or her employment has the right to claim patrimonial loss only3 through an
administrative process,4 which requires the Director-General of the Department of Labour
(‘Director-General’) to determine the amount that the employee is entitled to.5 Compensation is
paid from a statutory compensation fund 6 to which employers are obliged to contribute, with a
criminal sanction for non-compliance.7
As stated earlier, payment of compensation does not depend on proof of negligence, on the
part of the employer or anyone else. Therefore, the employee’s contributory negligence will not
reduce the compensation.8 However, if an accident is caused by an employee’s ‘serious and willful
misconduct’, no compensation is paid, unless the accident results in serious disablement, or if the
employee dies and leaves a dependant who is wholly financially dependent upon him.9 The amount
of compensation may be increased if the employer or a co-employee was negligent, but not beyond
the extent of the claimant’s actual patrimonial loss.10
Although the COIDA has abolished the employee’s common-law delictual claim against his or
her employer, it retains a common-law action against a wrongdoer other than the employer
(referred to as a ‘third party’).11 However, damages recoverable from such a third party are reduced
by the amount of compensation recoverable under the Act, so that the employee cannot recover
double compensation.12 The Director-General also has a claim against such a third party ‘for the
recovery of compensation that he is obliged to pay in terms of this Act’. 13
If an employee is dissatisfied with the Director-General’s interpretation of the Act or an award,
there is provision for lodging an objection, within 180 days, in the prescribed manner.14 If the
employee is still dissatisfied, the employee can appeal to the High Court, which has jurisdiction on
the following matters:

• The interpretation of the Act or any other law

• Whether an employee’s serious and willful misconduct resulted in an accident or occupational
disease that caused the disablement or death of the employee
• Whether the amount of any compensation awarded is so excessive or so inadequate that the
award could not reasonably have been made
• The right to increased compensation in terms of section 56.15
In terms of section 35 of the Act, employees’ common-law delictual claims against their employers
have been abolished. This section was subject to constitutional scrutiny and is discussed in more
detail below.

37.3 Section 35
Section 35 provides as follows:
(1) No action shall lie by an employee or any dependant of an employee for the recovery of
damages in respect of any occupational injury or disease resulting in the disablement or
death of such employee against such employee’s employer, and no liability for
compensation on the part of such employer shall arise save under the provisions of this
Act in respect of such disablement or death.
(2) For the purposes of subsection (1) a person referred to in section 56(1) (b), (c), (d) and
(e) shall be deemed to be an employer.

In other words, employees who qualify for compensation under the Act cannot institute a common-
law delictual claim against their employer. The validity of this exclusion of employers’ liability has
been challenged under the Constitution of the Republic of South Africa Act 200 of 1993. In Jooste
v Score Supermarket Trading (Pty) Ltd (Minister of Labour Intervening)16 an employee who slipped
and fell at her workplace as a result of her employer’s alleged negligence, argued that section 35
infringed on her constitutional right to equality because, by being deprived of the common-law
right to claim damages against her employer, she was placed at a disadvantage when compared to
other people who were not employees and who thus retained the right.17
The case ultimately turned on whether the impugned section was rationally connected to a
legitimate government purpose.18 In this regard, the Court was confronted with the disparity
between the general purpose of the Act, that is, to promote the interests of employees, and the fact
that section 35 operated to employees’ disadvantage. The Court confirmed that the rights of
employees to claim damages at common law were curtailed by the section, but emphasised also that
the Act allowed employees to obtain limited compensation without having to prove negligence. In
the end, it was held that it was not for the Court to determine whether the statutory compensation
scheme under the Act was to the disadvantage of employees. Rather, whether an employee should
be able to keep the common-law right to claim damages, either over and above, or as an alternative
to the advantages conferred by the COIDA, was a highly debatable, controversial and complex
matter of policy. It involved a policy choice that the legislature, and not a court, were best
positioned to make.
The Court found that the legislature had considered that it was appropriate to grant employees
certain benefits not available at common law, while excluding certain common-law rights.
Therefore, section 35(1) of the Act was logically and rationally connected to a legitimate
government purpose, namely the comprehensive regulation of compensation for harm arising from
occupational injuries or diseases sustained or contracted by employees in the course of their
employment.
For the exclusion of employer’s liability in terms of section 35 to apply, the parties and the
claim in question must meet the requirements as set out in relevant definitions.
• The claimant must be an ‘employee’, which is defined as:
a person who has entered into, or works under a contract of service or of apprenticeship or
learnership, with an employer, whether the contract is express or implied, oral or in writing,
and whether the remuneration is calculated by time or by work done, or is in cash or in
kind.
• In Crown Chickens (Pty) Ltd t/a Rocklands Poultry v Rieck 19 the Supreme Court of Appeal
held that a person contracted out to a company by a labour broker was an employee of the
labour broker and not of the company where she worked. Therefore, she could sue the
company for injuries arising out of an incident in which she was taken as a hostage by robbers,
and where security guards employed by the company negligently caused her injuries when
firing at the robbers.
• An ‘employer’, on the other hand, includes any person, including the State, who employs an
employee; any person controlling the business of an employer; a temporary employer to whom
the services of an employee is lent or let; and a labour broker who against payment provides a
person to a client to supply a service or perform work. Section 35 uses the term in an extended
sense. In terms of section 35(2), read with section 56(1), ‘employer’ includes:
◆ An employee charged by the employer with managing or controlling the business or any
branch or department thereof
◆ An employee who has the right to engage or discharge employees on behalf of the
employer
◆ An engineer appointed to be in general charge of machinery, or of a person appointed to
assist such engineer
◆ A person appointed to be in charge of machinery in terms of any regulation made under
the Occupational Health and Safety Act 85 of 1993.
• Employees have no claim against their employer for harm arising from an ‘occupational injury
or disease’. 20 The concept of ‘occupational injury’ means a personal injury that results from an
‘accident’, which is in turn defined as ‘an accident arising out of and in the course of an
employee’s employment’. 21
◆ In Rauff v Standard Bank Properties (A Division of Standard Bank of South Africa Ltd) 22
the Court held that an employee, who was injured while leaving work when the lift in the
building that she worked in malfunctioned, did not suffer an ‘occupational injury’,
because she was no longer working at the time and the injury did not arise ‘out of and in
the course of employment’. Therefore, section 35 did not apply and she could sue her
employer.
◆ In DN v MEC for Health Free State 23 the plaintiff, a female medical doctor employed by
the defendant, brought a delictual action against the defendant, her employer, in order to
recover the harm she had suffered after being raped by an intruder who had gained access
to the employer’s premises. The doctor was raped while discharging her duties, while
walking from one hospital building to another. At the time of the incident, building
construction was being carried out at the hospital and a portion of the parameter fencing
was under temporary repair. In addition, the elevator between the ground and first floor in
the building was defective and the lights on the first floor were also not working. The
doctor’s assailant was not a patient of the relevant hospital, had no authorisation to be on
its premises and was later convicted of rape and sentenced to 15 years’ imprisonment. The
Court had to determine whether the rape of the doctor could be regarded as an ‘accident’
arising ‘out of and in the course and scope of employment’. The Court confirmed that the
approach developed in the majority judgment of Chief Justice Rumpff in Minister of
Justice v Khoza 24 was still good law for establishing whether an incident is an ‘accident’
which arose out of or in the course of employment. Essentially the Khoza approach
required in the broad sense a causal connection 25 between the employee’s employment
and the relevant accident.26 Applying the approach in Khoza, the Court held that the attack
on the doctor bore no relationship to her employment. In doing so, the Court emphasised
that the rape, although unexpected, was intentional and fell outside the scope of meaning
ordinarily ascribed to ‘accident’ by South African courts.27 Furthermore, the fact that the
 doctor was intentionally injured by a person unauthorised to be on the hospital’s premises,
together with the perpetrator’s motive, played a significant role in concluding that there
was no causal connection between the doctor’s employment and the crime.28
• The concept of ‘occupational disease’ means any disease contemplated in section 65(1) (a) or
(b). Section 65(1)(a) allows an employee to claim compensation if he or she can show that the
disease contracted is mentioned in Schedule 3 to the Act and that such disease has arisen out of
and in the course and scope of employment. Notwithstanding, section 65(1)(b) allows an
employee to claim compensation for a disease not mentioned in Schedule 3, if he or she can
prove to the Director-General that such disease arose out of and in the course and scope of
employment.

Section 35 also applies to claims under the Road Accident Fund Act 56 of 1996. In Mphosi v
Central Board for Co-operative Insurance Ltd 29 the Court held than an employee has no claim
against the predecessor of the current Road Accident Fund if the claim is based on his or her
employer’s negligence, because the liability of the Fund depends on whether a claim under
common law exists, which in this case would have been excluded by the predecessor of the current
section 35.
In Mankayi v Anglogold Ashanti Ltd 30 the Constitutional Court had to decide whether section
35(1) of the COIDA also extinguished the common-law delictual claim of mine workers to recover
damages in delict for an occupational injury or disease from their negligent employers (mine
owners), notwithstanding the fact that they were not entitled to claim compensation under the
COIDA, but under other legislation, namely the Occupational Diseases in Mines and Works Act 78
of 1973 (‘ODIMWA’). Earlier, the Supreme Court of Appeal had held that section 35(1) of the
COIDA extinguished mine workers’ common-law claim and thus extended the protection offered
against common-law liability under section 35(1) also to mine owners.31 The Constitutional Court
rejected this decision and held that the ‘employee’ referred to in section 35(1) of the COIDA is
limited to an employee who has a claim for compensation under the COIDA and it therefore
excluded mine workers.32 It confirmed that section 35(1) of the Act does not cover an ‘employee’
who qualifies for compensation in respect of ‘compensatable diseases’ under the ODIMWA.33

1 Media 24 Ltd v Grobler 2005 (6) SA 328 (SCA) para 65.

2 Jooste v Score Supermarket Trading (Pty) Ltd (Minister of Labour Intervening) 1999 (2) SA 1 (CC) para 15. See also
MEC for Education, Western Cape Province v Strauss 2008 (2) SA 366 (SCA) paras 11–12; Healy v Compensation
Commissioner 2010 (2) SA 470 (E) para 11; Sanan v Eskom Holdings Ltd 2010 (6) SA 638 (GSJ) para 8; MEC for
Health, Free State v DN 2015 (1) SA 182 (SCA) paras 6–7; Thomas v Minister of Defence and Military Veterans
2015 (1) SA 253 (SCA) para 6.
3 Sections 47–64: no compensation for non-patrimonial harm such as pain and suffering.
4 Sections 38–46.
5 Section 4.
6 Section 15.
7 Section 87.
8 Section 22(1).
9 Section 22(3)(a).
10 Section 56(4).
11 Section 36(1)(a).
12 Section 36(2).
13 Section 36(1)(b).
14 Section 91(1)–(4).
15 Section 91(5).
16 1999 (2) SA 1 (CC).
17 Para 10.
18 Para 12.
19 2007 (2) SA 118 (SCA).
20 Section 1.
21 Section 1 contains the following definitions: ‘occupational injury’ means a personal injury sustained as a result of an
accident; and ‘accident’ means an accident arising out of, and in the course of an employee’s employment and
resulting in a personal injury, illness or the death of the employee.
22 2002 (6) SA 693 (W).
23 2014 (3) SA 49 (FB); MEC for Health, Free State v DN 2015 (1) SA 182 (SCA).
24 1966 (1) SA 410 (A).
25 See Olivier ‘Social Security: Core Elements’ LAWSA 2 ed Vol 13 Part 3 (2013) para 126, where the authors refer to
Basson v Ongevallekommissaris [2000] 1 All SA 67 (C), in which the employee had a pre-existing back condition and
was subsequently involved in an accident. With regard to the causal connection between the accident and the
employment, the Court held that it was not required that the injury suffered by the employee should be exclusively the
result of the accident in question. In other words, it was sufficient that the accident was merely a ‘contributing factor’
to the injury. Put differently, the Court thus held that there could be more than one factual cause of the injury: both the
pre-existing back condition and the accident. This differs from the common law of delict, which requires of a plaintiff
to prove a necessary cause (conditio sine qua non) and thus imposes a greater evidentiary burden on the plaintiff.
26 See Minister of Justice v Khoza 1966 (1) SA 410 (A) at 417:
‘In die loop daarvan’ beteken dat die ongeval moet plaasvind terwyl die werksman besig is met sy
werksaamhede en dit ontstaan ‘uit sy diens’ as die ongeval in verband staan met sy werksaamhede.
27 MEC for Health, Free State v DN 2015 (1) SA 182 (SCA) at 190J–191D.
28 At 191D.
29 1974 (4) SA 633 (A).
30 2011 (3) SA 237 (CC).
31 Para 113.
32 Para 113.
33 Paras 113–114.
 Chapter 38

Road Accident Fund Act 56 of 1996

38.1 Introduction

38.2 Operation of the RAF Act

38.3 Scope of liability under the RAF Act

38.3.1 The liability of the Fund
38.3.2 Limitations of the Fund’s liability under the RAF Act
38.3.3 Exclusions of the Fund’s liability under the RAF Act

38.4 Damages

38.5 Causation of harm, locality, ‘driving’ and ‘motor vehicle’

38.6 Liability fault-based |

38.7 Delimitation of RAF claims

38.8 Prescription, procedure and right of recourse

38.9 The proposed RABS

38.1 Introduction
Motivated by the policy considerations outlined in Chapter 35, the legislature has introduced a
series of statutes aimed at compensating road users who suffer harm arising from motor vehicle
accidents. The current statute, the Road Accident Fund Act 56 of 1996 (‘RAF Act’), has the same
object as that of its predecessors, namely the ‘payment of compensation … for loss or damage
wrongfully caused by the driving of motor vehicles’. 1 The RAF Act achieves this through the
establishment of a statutory compensation fund, financed from fuel levies. The RAF Act transfers
liability from the negligent driver, or owner or employee of the owner of a motor vehicle to the
Road Accident Fund (the ‘Fund’) resulting from bodily injury or death. Like its forerunners, and in
contrast to the Compensation for Occupational Injuries and Diseases Act 130 of 1993 (‘COIDA’),
the Act is based on the common-law elements for delictual liability and retained fault as the basis
for liability.
The RAF Act was amended by the Road Accident Fund Amendment Act 19 of 2005 (the
‘RAFA Act’). The amendments sought to achieve the simplification of the claims procedure,
promoting access to justice and ensuring the financial stability and sustainability of the RAF. The
most important amendments introduced by the RAFA Act, which relate to the scope and extent of
the RAF’s liability and the abolition of the victim’s common-law claim against the wrongdoer, are
discussed in section 38.3 below.
The fault-based compensation system established under the RAF Act has been criticised as
being ‘unreasonable, inequitable, unaffordable and unsustainable’ 2 on the basis that it contributes
to the financial decline of the Fund. The amendment of the RAF Act by the RAFA Act may be
regarded as a first step towards greater reform, which is ultimately focused on the restructuring of
the Fund so that compensation is paid to victims of motor vehicle accidents on a no-fault basis.3 To
achieve the desired reform, the legislature published the draft Road Accident Benefit Scheme Bill
(the ‘RABS Bill’) in 2014. Should it be enacted, the current fault-based system of liability
administered by the Fund will be replaced by a new social security scheme for road accident
victims.

38.2 Operation of the RAF Act

If a person (referred to in the Act as a ‘third party’) suffers injuries or dies as a result of the
negligent driving of a motor vehicle, the Fund is liable to pay compensation, instead of the driver,
owner or employee of the owner that would have otherwise existed in common law, except where
the Fund is unable to pay compensation, conceivably for reasons of insolvency.4 The third party
may institute a statutory claim against the Fund to claim limited patrimonial and non-patrimonial
damages from the Fund (see further section 38.4 below).
As indicated below, the statutory compensation system in terms of the current RAF Act is
fault-based, which means that the plaintiff has to prove the negligence of the driver or of the owner
of the motor vehicle or of the owner’s employee in the performance of his or her duties as
employee. Normally, the negligent conduct consists of the driving of a motor vehicle, but it can
also be negligent maintenance or use of the vehicle by the owner or an employee of the owner in
the performance of duties.
The institution of the third party’s claim involves an administrative procedure set out in section
24 of the Act, which requires a victim, among other things, to lodge a prescribed statutory claim
form.5 Once the claim is lodged, the Fund will commence with its investigation to determine the
claim’s validity and the amount of damages payable.
If the Fund does not object to the validity of the claim within 60 days from the date on which a
claim was sent by registered post or delivered by hand to the RAF, the claim is deemed to be valid
in law in all respects.6 In terms of section 24(6) of the Act, a summons may not be served on the
Fund before the expiry of a period of 120 days from the date on which the claim was sent or
delivered by hand, unless the Fund repudiates the claim in writing before the expiry of this period.
If the claim is valid, the Fund is liable to the third party and the negligent driver, owner, or
owner’s employee concerned normally incurs no liability.7 However, in certain circumstances, the
Fund has a right of recourse against the driver or the owner of the vehicle, as set out in section 25,
including cases of driving without a licence and driving under the influence of alcohol or other
drugs.
 Claims under the Act are subject to prescription provisions contained in section 23.

38.3 Scope of liability under the RAF Act

38.3.1 The Liability of the Fund

The key provision of the current compensation scheme is section 17(1) of the RAF Act, which
provides:
The Fund or an agent shall … be obliged to compensate any person (the third party) for any
loss or damage which the third party has suffered as a result of any bodily injury to himself
or herself or the death of or any bodily injury to any other person, caused by or arising from
the driving of a motor vehicle by any person at any place within the Republic, if the injury
or death is due to the negligence or other wrongful act of the driver or of the owner of the
motor vehicle or of his or her employee in the performance of the employee’s duties as
employee. Provided that the obligation of the Fund to compensate a third party for non-
pecuniary loss shall be limited to compensation for a serious injury as contemplated in
subsection (1A) and shall be paid by way of a lump sum.

The effect of section 17 is that the liability of the Fund replaces the liability of the driver, or owner
or owner’s employee that would have otherwise existed at common law, except where the Fund is
unable to pay compensation.
Because the Fund’s liability is fault-based, the amount of damages recoverable by a plaintiff
may be reduced to the extent that a plaintiff’s fault contributed to his injury or death and in
accordance with sections 1 and 2 of the Apportionment of Damages Act 34 of 1956.
The RAF Act refers to the claimant as ‘the third party’, a concept that the Act does not define,
but gives content to by implication: The liability transferred to the Fund in terms of section 17 is
common-law liability for negligently caused injury or death,8 and a ‘third party’ is any person who
would have had a common-law claim for damages arising from bodily injury or death, including
the injured person, a dependant of a person who died in a motor vehicle accident, a supplier of
medical services, and any other person who, in terms of the ordinary principles of delict, has a
claim based on the injury or death of another.9
The Fund is liable only for harm (‘loss or damage’) that arises from a bodily injury or death
and therefore does not include property damage. The Fund may incur limited liability for both
patrimonial and non-patrimonial harm (see further section 38.4 below).

38.3.2 Limitations of the Fund’s liability under the RAF Act

Even if the requirements of section 17 are met, there are certain limitations to the liability of the
Fund, as set out in sections 17 and 18 of the RAF Act. They include the following:
• Section 17(1)(1A) of the Act limits the Fund’s liability for non-patrimonial harm to
‘compensation for a serious injury … and shall be paid by way of a lump sum.’ In accordance
with section 17(1)(1A)(a) the assessment of a serious injury shall be based on a ‘prescribed
method adopted after consultation with medical service providers and shall be reasonable in
ensuring that injuries are assessed in relation to the circumstances of the third party.’ The
assessment must be carried out by a registered medical practitioner registered under the Health
Professions Act 56 of 1974.
• Section 17(4)(c) limits the Fund’s liability insofar as it provides that, where compensation is
claimed for loss of income or loss of support, the claim will be capped to a specific amount
that may be adjusted by the legislature, irrespective of the actual loss. The constitutional
 validity of this subsection was challenged in Law Society of South Africa v Minister for
Transport.10 On behalf of the applicants it was argued that the section failed to comply with the
constitutional principle of rationality and unjustifiably limited various constitutional rights.11
The Constitutional Court upheld the relevant provisions, which were considered to have the
effect of making the Fund financially viable and rendering the compensation regime more
transparent, predictable and equitable. Therefore, the Court also concluded that the provisions
did not constitute an arbitrary deprivation of property.
• Section 18(2) provides that where the loss or damage is suffered as a result of bodily injury or
death of a person who was being conveyed in or on the motor vehicle concerned, and who was
an employee of the driver or owner of that motor vehicle, and the person who can claim
damages from the Fund (the third party) is also entitled to compensation under the COIDA, the
liability of the Fund is limited. The Fund is only liable for the amount representing the
difference between the amount which that third party could, but for this paragraph, have
claimed from the Fund and any lesser amount to which that third party is entitled by way of
compensation under COIDA. The Fund shall not be liable for the amount of the compensation
to which the third party is entitled under the COIDA. The effect of section 18(2) is that, where
compensation is recoverable under the COIDA, the amount of such compensation must be
deducted from damages that are recoverable from the Fund. Furthermore, in cases where the
claim arises from the injury or death of a passenger in a vehicle who was an employee of the
driver or owner of that motor vehicle, there is also no right of recovery for the compensation
fund under COIDA against the Fund. COIDA makes provision for such a right of recovery in
certain other cases.
• Section 18(4) also states that the liability of the Fund for funeral expenses that arise as the
result of the death of a person shall be limited to the necessary actual costs to cremate the
deceased or to inter the person in a grave.

38.3.3 Exclusions of the Fund’s liability under the RAF Act

Section 19 further excludes liability for the Fund in the following cases:
• If the claim was not instituted and prosecuted by the third party, or on behalf of the third party
by a person entitled to practise as an attorney or any person who is in the service, or who is a
representative of the State or government or a provincial, territorial or local authority.
• If the third party had agreed to pay someone other than the category of persons mentioned
above a part of the compensation or some other remuneration for handling the claim.
• If the injured person unreasonably refuses or fails to subject himself or herself, at the request
and cost of the Fund, to any medical examination by medical practitioners designated by the
Fund; or refuses or fails to furnish the Fund, at its request and cost, with copies of all medical
reports in his or her possession that relate to the relevant claim for compensation; or refuses or
fails to allow the Fund, at its request, to inspect all records that relate to himself that any
hospital, or his or her medical practitioner possess.
• If the third party refuses or fails to submit to the Fund, together with the claim form or within a
reasonable period thereafter, an affidavit that fully sets out the particulars of the accident that
gave rise to the claim; or to furnish the Fund with copies of all statements and documents that
relate to the accident that gave rise to the claim, within a reasonable period of coming into
possession of them.
• If the loss or harm is suffered as a result of an emotional shock sustained when someone
witnessed, observed, or was informed of the bodily injury or the death of another person as a
result of the driving of a motor vehicle.12 In other words, a victim of so-called ‘secondary
emotional shock’ 13 may not institute a claim against the Fund and must institute a common-
 law delictual claim against the culpable wrongdoer for the harm that results from the emotional
shock.

The current section 21 of the RAF Act is the result of the amendments introduced by the RAFA Act
and severely limits the common-law right to compensation of motor vehicle accident victims.
Under the older version of the RAF Act, victims were entitled to claim compensation from the
wrongdoer in respect of the harm that was not compensable under the Act. Significantly, section 21
(1) now states that no claim for compensation in respect of:
loss or damage resulting from bodily injury to or the death of any person caused by or
arising from the driving of a motor vehicle shall lie caused by or arising from the driving of
a motor vehicle shall lie
(a) against the owner or driver of a motor vehicle; or
(b) against the employer of the driver

This subsection abolishes the victim’s residual common-law delictual claim for harm which is not
covered by the RAF Act. A motor vehicle accident victim may look only to the Fund for
compensation and, to cover harm not compensable under the Act, must take out appropriate
insurance.

PAUSE FOR In Law Society of South Africa v Minister for Transport14 the Constitutional
REFLECTION Court considered the constitutional validity of the abolition of the motor
vehicle accident victim’s residual common-law delictual claim against a
wrongdoer.15 The applicants argued that the amendment failed to comply
with the constitutional principle of rationality and unjustifiably limited at
least one of the following fundamental rights contained in the Bill of
Rights: sections 12(1),16 25(1),17 27(1)18 and 38.19 Despite the far-reaching
consequences proposed by the amended section 21 of the RAF Act, the
Court ultimately held that it passed the necessary rationality test.20 It
furthermore confirmed that the abolition of the victim’s common-law claim
against the wrongdoer was a justifiable infringement of section 12(1) of
the Bill of Rights, and that it did not infringe upon section 38 of the
Constitution.21 In addition, it was confirmed that the proposed change did
not constitute an arbitrary deprivation of property.
The judgment raises questions pertaining to the function of the law of
delict that require consideration: Should the law of delict be implemented
as an instrument with which to distribute loss and to promote social
security, or should it be restricted to correcting wrongs by forcing a
culpable wrongdoer who has caused the victim’s harm to pay
compensation? What is the implication of the judgment for the role that
the law of delict must play in promoting the central moral notion of
personal responsibility, which is regarded as the justificatory principle for
a fault-based system of liability? What effect will the abolition of the
victim’s common-law right to claim damages from the wrongdoer have on
deterring future drivers from causing harm through the negligent driving
of motor vehicles?

38.4 Damages
The Fund is liable only for harm (‘loss or damage’) that arises from a bodily injury or death and
therefore not for property damage. The Fund may incur limited liability for both patrimonial and
non-patrimonial harm that arises from a bodily injury or death. The patrimonial harm compensable
under the Act includes past and future loss resulting from bodily injury or death which a road
accident victim may have suffered. This includes: 22
• Hospital and medical expenses
• Loss of income
• Loss of earnings
• Loss of support
• Funeral and cremation costs
• Travelling expenses
• The cost of a nurse or an aide.
The claimant is also entitled to claim damages for non-patrimonial harm. However, as indicated
above, section 17(1)(1A) of the Act limits the Fund’s liability for non-patrimonial harm to
‘compensation for a serious injury and shall be paid by way of a lump sum.’ In this context, the
third party would be able to claim for pain and suffering, psychological trauma, emotional shock,
disfigurement, loss of the amenities and enjoyment of life, loss of general health and shortened life
expectancy.
As indicated in section 38.3, the Fund’s liability is limited and/or excluded in some cases and
certain damages will therefore be excluded or limited.

Figure 38.1 Structure of the Road Accident Fund compensation scheme

38.5 Causation of harm, locality, ‘driving’ and ‘motor

vehicle’
Causation of harm can be direct or indirect (‘caused by or arising from th riving of a motor
vehicle’). Inrectly caused injury or dea, which falls under section 17, includes, f example, injury
from a fire following a negligently caused accident,23 and injury from an attack on a bus that was
allwed to be driven in a dangerous area.24
The liability of the Fund can arise from accidents ‘at any place within the Republic’, and not
only on public or other roads. The accident must involve the ‘driving’ of a ‘motor vehicle’. Secion
20 partially defines ‘driving’, with deeming provisions, namely that a vehicle is ‘deemed to be
driven by the person in control of the vehicle’, that ‘a person who has placed or left a motor vehicle
at any place shall be deemed to be driving that motor vehicl while it moves from that place as a
result of gravty, or while it is stationary at that place or at a place to whih it moved from the first-
mentioned place as a result of gravity’, and that ‘(W)heneve any motor vehicl has been placed or
left at any place, it shall, for the purposes f this Act, be prsumed, until the cotrary is proved, that
such vehicle was placed or left at such place by the owner of such vehicle.’
Section 1 of the Ac defines a ‘motor vehicle’ as ‘any vehicle designed or adapted for
propulsion or haulage on a road by means of fuel, gas or electricity, including a trailer, a caravan,
an agricultural or any other implement designed or adapted to be drawn by such motor vehicle’.

38.6 Liability fault-based

The liability of the Fund is fault-based, which means that the plaintiff has to prove ‘negligence or
other wrongful act of the driver or of the owner of the motor vehicle or of his or her employee in
the performance of the employee’s duties as employee’. The ‘negligence or other wrongful act’ can
be (and mostly is) that of the driver, but it can also be related to the maintenance or use of the
vehicle by the owner or an employee of the owner in the performance of duties.

38.7 Delimitation of RAF claims

Road accidents that do not involve ‘driving’ a ‘motor vehicle’ or fault on the part of the driver,
owner or employee of the owner as required by section 17, can only give rise to general delict
claims in terms of the ordinary principles of the law of delict.

Figure 38.2 Delimitation of RAF claims

38.8 Prescription, procedure and right of recourse

Claims under the Act are subject to prscrption proviions contained in section 23, and must be
instituted according to the procedural requirements as set out in section 24.
Where the Fund is liable to the third party in terms of the Act, the negligent driver, owner, or
employee of the owner of the motor vehicle concerned normally incurs no liability.25 However in
certain circumstances, the Fund has a right of recourse against the driver or the owner of the
vehicle, as set out in section 25, including cases of driving without a licence and driving under the
influence of alcohol or other drugs.
38.9 The proposed RABS
The fault-based compensation system under the RAF Act has been described as ‘unreasonable,
unaffordable and unsustainable’ 26 and a few points of criticism are summarised below:
• Because the liability of the Fund is fault-based, victims are required to prove negligence.
• Proving fault in the form of negligence often leads to disputes that may only be resolved
through litigation, which is an expensive and time-consuming exercise. In addition, while
litigation is pending, victims have to pay for medical and other expenses themselves, which is
often impossible for some victims.
• The RAF Act is open to abuse due to fraud, opportunistic claims, nuisance and overinflated
claims, mismanagement, professional malpractice and human failing.
• The structure of the current compensation system encourages perverse incentives: Victims are
presented as permanently disabled, maimed and without the capacity to continue earning an
income, irrespective of their actual condition. There is thus no financial incentive for injured
accident victims to return to work or to undergo rehabilitation.

The legislature heeded the criticism and has proposed a new scheme for the compensation of motor
vehicle accident victims: The RABS was published for comments in 2014. It aims to introduce a
‘social security scheme for the victims of road accidents’. This will be done by providing ‘a set of
defined benefits on a no-fault basis’ to persons who have suffered harm as a result of ‘bodily injury
or death caused by or arising from road accidents’. Apart from its abolition of fault, the RABS will
exclude the delictual liability of ‘certain persons responsible for bodily injuries or death caused by
or arising from accidents’. The proposed RABS furthermore seeks to simplify the claims procedure
and reduce potential legal disputes.

1 Section 3 of the RAF Act.

2 Department of Transport Policy Paper for the Road Accident Fund Benefit Scheme (RABS) Notice 815 GG 34765 21
November 2011) at 6. See also Chapter 35 for a discussion of the way in which the fault requirement has provided
motivation for legislative development of the law of delict.
3 Law Society of South Africa v Minister of Transport 2011 (1) SA 400 (CC) paras 44–45.
4 Section 17(1) of the RAF Act.
5 The form should provide the basic information of the victim, the vehicles and parties involved, the date and place of
the accident, the amount claimed by the victim, a medical report compiled by the treating doctor, and an affidavit that
contains further reports and statements pertaining to the accident and the amount claimed. The completed claim and
accompanying documentation must be sent by registered post or delivered by hand to the Fund at its principal, branch
or regional office.
6 Section 24(5) of the RAF Act.
7 Section 21 of the RAF Act.
8 The reference in section 17 of the RAF Act refers to injury or death ‘due to the negligence or other wrongful act’ and
is a reference to common-law fault-based liability. Furthermore, section 19(a) of the Act in effect excludes liability of
the Fund where no common-law liability exists.
9 Section 17(5) of the RAF Act.
10 2011 (1) SA 400 (CC).
11 The right to security of the person, the right not to be arbitrarily deprived of property, the right of access to health-care
services and the right to an adequate remedy.
12 See section 19(g) of the RAF Act for the exclusion of the Fund’s liability for instances of so-called secondary
emotional shock.
13 See section 21(2)(b) of the RAF Act. For an example of secondary emotional shock, see the facts in Barnard v Santam
Bpk 1999 (1) SA 202 (SCA).
14 2011 (1) SA 400 (CC).
15 Para 15.
16 The right to security of the person.
17 The right not to be arbitrarily deprived of property.
18 The right of access to health-care services.
19 The right to an adequate remedy.
20 Law Society of South Africa v Minister for Transport 2011 (1) SA 400 (CC) para 35.
21 Paras 75–80 and 103.
22 See sections 17(4)–17(6) and 18(4) of the RAF Act.
23 Groenewald v Protea Assurance Co Ltd 1979 (1) SA 354 (C).
24 General Accident Insurance Co of South Africa Ltd v Xhego 1992 (1) SA 580 (A).
25 Section 21 of the RAF Act.
26 Department of Transport Policy Paper (2011) at 6.
Chapter 39

Strict liability for harm caused by goods:

Section 61 of the Consumer Protection
Act

39.1 Introduction

39.2 What is the nature of the liability created by the Act?

39.3 Why should there be strict liability for defective products?

39.4 What are ‘goods’ for the purpose of strict liability?

39.4.1 Immovable property
39.4.2 Information products
39.4.3 Contaminated blood

39.5 Defectiveness of goods

39.5.1 Unsafe |
39.5.2 Failure
39.5.3 Defect
39.5.4 Hazard
39.5.5 A puzzling interplay of definitions

39.6 Inadequate instructions or warnings provided to the consumer

pertaining to any hazard

39.7 Who is a consumer?

39.8 Who is liable?

39.9 For what type of harm will liability be imposed?

39.10 Defences
 39.10.1 Section 61(4)(a): Compliance with public regulation
39.10.2 Section 61(4)(b)(i): Absence of defect at time of supply
39.10.3 Section 61(4)(b)(ii): Compliance with instructions
39.10.4 Section 61(4)(c): Defects not reasonably discoverable by the distributor or
retailer
39.10.5 Section 61(4)(d): Time limit for claims

39.11 Conclusion

39.1 Introduction
In response to the risk of harm that can be caused by consumer products, and the difficulty of
proving negligence on the part of manufacturers as required for Aquilian liability, discussed in
Chapter 19, a strict liability regime for harm caused by defective products was introduced by
section 61 of the Consumer Protection Act 68 of 2008, which came into effect on 1 April 2011.
South Africa has followed the example of other countries: strict liability for defective products has
become part of the law in many other jurisdictions.

39.2 What is the nature of the liability created by the Act?

Section 61 of the Act provides:
(1) Except to the extent contemplated in subsection (4), the producer or importer,
distributor or retailer of any goods is liable for any harm, as described in subsection (5),
caused wholly or partly as a consequence of:
(a) supplying any unsafe goods;
(b) a product failure, defect or hazard in any goods; or
(c) inadequate instructions or warnings provided to the consumer pertaining to any hazard
arising from or associated with the use of any goods irrespective whether the harm
resulted from any negligence on the part of the producer, importer, distributor or
retailer, as the case may be.

The consumer has an action against the producer, importer, distributor or retailer, without having to
prove negligence. The consumer can be the purchaser or merely the user of the product,1 and the
action is not limited by the ambit of whatever warranty was given for the product.

39.3 Why should there be strict liability for defective

products?
The arguments favouring strict liability for defective products are basically about fairness, the
difficulty of proving negligence, economic efficiency, and deterrence. Normally, consumers lack
the resources to analyse or scrutinise products effectively for safety. The efficiency argument is that
the burden of harm that results from using defective products should fall on the producer, who is
best able to control the danger or to spread the losses when they do occur, either through insurance
or through increased prices. Judge Traynor, in the well-known American case of Escola v Coca-
Cola Bottling Co of Fresno,2 summed it up well:
 The cost of an injury and the loss of time or health may be an overwhelming misfortune to
the person injured, and a needless one, for the risk of injury can be insured by the
manufacturer and distributed among the public as a cost of doing business.

Apart from the ‘down-stream’ compensatory function of strict liability, there is also an ‘up-stream’
preventative or deterrence function. In the USA in particular, courts have stressed that product
liability is a powerful way of inducing product safety.

39.4 What are ‘goods’ for the purpose of strict liability?

The Act defines ‘goods’ as follows:
(a) anything marketed for human consumption;
(b) any tangible object not otherwise contemplated in paragraph (a) including any medium
on which anything is or may be written or encoded;
(c) any literature, music photograph, motion picture, game, information, data, software,
code or other intangible product written or encoded on any medium, or a licence to use
any such intangible product;
(d) a legal interest in land or any other immovable property, other than an interest that
falls within the definition of ‘service’ in this section; and
(e) gas, water and electricity.

This wide and general definition includes all kinds of substances produced and tradable as
commodities. In combination, paragraphs (a), (b), and (c) include any tangible object whether
marketed for human consumption or otherwise, as well as a wide range of intangible products,
including information. In addition to goods produced by industrial process – from machinery to
pharmaceuticals – it includes raw and processed foodstuffs. A ‘tangible object’ also includes
natural substances such as body parts, blood products, minerals and water. Water is in any event
mentioned specifically, together with gas and electricity.

39.4.1 Immovable property

The Act also covers land transactions. Its definition of goods includes:
a legal interest in land or any other immovable property, other than an interest that falls
within the definition of ‘service’ in this section.

This involves liability for structural or design defects in buildings and hazards that occur on land.
Damage to the product itself – the land or buildings – is apparently also recoverable.

PAUSE FOR Liability for defects in immovable property

REFLECTION Liability for defects in immovable property is an area governed by well-
settled principles of contract and property law. Consider, for example, the
situation where defective property sold subject to a ‘voetstoots’ clause
causes injury to the purchaser.

39.4.2 Information products

The definition of ‘goods’ includes ‘any literature, music, photograph, motion picture, game,
information, data, software, code’. In this respect, the Act goes much further than its counterparts in
Europe. The Act creates strict liability not only for mass-produced informational products, such as
software packages, but also for the producer of defective software for products such as aeroplanes,
cars or heart-lung machines. There appears to be no distinction between defective information
accompanying the product, which renders it dangerous, and defective information obtained from
the product (for example, a recipe contained in a cookery book that encourages the use of
potentially poisonous mushrooms).

PAUSE FOR Strict liability for information products

REFLECTION Should there be strict liability for defective information?
Strict liability for an open-ended category of information products, with
potential indeterminate liability for economic loss, may have far-reaching
effects. Defective information could have harmful consequences with a
considerably wider reach than defective tangible objects. Consider
whether strict liability could inhibit the socially and economically desirable
free sharing of ideas and theories.

39.4.3 Contaminated blood

Contaminated blood and blood products also qualify as ‘goods’ (tangible object) and this means
that the South African National Blood Service (SANBS) potentially will be liable for contaminated
blood and blood products.

PAUSE FOR HIV/AIDS and liability

REFLECTION The prevalence of HIV and AIDS in South Africa creates a high risk of
potential liability for the SANBS in the event of the supply of
contaminated blood. Consider whether the imposition of strict liability on
a non-profit organisation such as the SANBS was intended, and, if it was,
whether strict liability could have undesirable consequences.

39.5 Defectiveness of goods

The remedies provided by section 61(1) are available for the causing of harm in consequence of (a)
the supply of ‘unsafe’ goods; (b) a product ‘failure’, ‘defect’ or ‘hazard’ in goods; or (c)
‘inadequate instructions or warnings’ provided with goods. The definitions of these key terms,
‘unsafe’, ‘failure’, ‘defect’ or ‘hazard’ are contained in section 53. Therefore, strict liability can
arise as a result of several alternative modes of conduct by the producer, importer, distributor or
retailer of any goods. One can only properly understand these by referring to the definitions
contained in sections 1 and 53.

39.5.1 Unsafe
According to section 53(1)(d), ‘unsafe means that, due to a characteristic, failure, defect, or hazard,
particular goods present an extreme risk of personal injury or property damage to the consumer or
to other persons’. In turn, the subsections give ‘failure’, ‘defect’ and ‘hazard’ their own statutory
definitions, although it does not define ‘extreme risk’. Liability arises if goods are ‘unsafe’ and
present an ‘extreme risk’ ‘due to a characteristic, failure, defect or hazard’. An accident occurs
typically because of a failure in the manufacturing process of goods that has caused a safety
problem, or a design defect in the goods that has rendered it unsafe to use.
39.5.2 Failure
‘Failure’ is relatively unproblematic. It means (according to section 53(1)(b)) that the product ‘did
not perform in the intended manner or to the intended effect’. This indicates a typical
manufacturing defect, for example, where a machine tool malfunctions and causes injury, or the
brakes of a car fail and cause an accident.

39.5.3 Defect
In the definition of ‘defect’, the Act broadly adopts a ‘consumer expectations’ standard. In terms of
section 53(1)(a) a ‘defect’ means:
(i) any material imperfection in the manufacture of the goods or components, or in
performance of the services, that renders the goods or results of the service less
acceptable than persons generally would be reasonably entitled to expect in the
circumstances; or
(ii) any characteristic of the goods or components that renders the goods or components
less useful, practicable or safe than persons generally would be reasonably entitled to
expect in the circumstances.

This definition of ‘defect’ is complex and is based on consumer expectations of performance,

coupled with vague and overlapping notions of ‘acceptable’, ‘useful’, ‘practicable’ and ‘safe’.
The European experience of the ‘consumer expectations’ approach has not been entirely
satisfactory. There are obvious difficulties, for example, ‘Are consumers entitled to expect more
than the exercise of reasonable knowledge, skill and care by producers?’ and ‘What does the
consumer generally expect?’ The test is too vague to provide much guidance and it can be used to
explain almost any result that a court chooses to reach. Jane Stapleton, a prominent critic, describes
the consumer expectations test as ‘impenetrable to analysis’. 3 It cannot mean that courts must
somehow determine the actual expectations of consumers generally: the consumer expectations test
means that courts should determine what consumers are entitled to expect. This is not simple,
because the definition of ‘defect’ employs a number of other undefined terms that are open-ended
and vague, such as ‘acceptable’, ‘useful’ and ‘practicable’. The courts are likely to fall back on
reasonableness to determine whether goods are defective.

39.5.4 Hazard
A ‘hazard’, in terms of section 53(1)(c)(ii), means a characteristic that ‘presents a significant risk of
personal injury to any person, or damage to property, when the goods are utilised’.

39.5.5 A puzzling interplay of definitions

The liability referred to in section 61(1)(a) – the supply of ‘unsafe’ goods – therefore, inter alia,
relates to supplying goods that present an ‘extreme risk’ of personal injury or property damage, due
to a ‘hazard’, which is a characteristic that presents a ‘significant risk’ of personal injury to any
person, or damage to property. This is a puzzling interplay of definitions. Even if it did make sense
to link extreme risk and significant risk in this convoluted way, the different levels of risk are not
defined and are difficult to distinguish. It is also difficult to see why a claimant would rely on an
‘extreme risk’ situation when it is also possible to rely on a ‘significant risk’ situation.
The Act refers to these characteristics of goods failure, defect and hazard alternately, so in
effect, section 61(1)(b) refers to various alternative subcategories of liability. However, given that
the same terms are all within the definition of ‘unsafe’ in section 61(1)(a), these categories are in
effect tautologous.
 If the criteria in sections 61(1)(a) and 61(1)(b) are alternatives, there appears to be no reason
why a claimant should not simply rely on the lowest and most general of these standards in section
61(1)(b), which is that of a ‘defect’ that relates to whether the goods are ‘acceptable’, whatever
judicial meaning can be ascribed to this.

PAUSE FOR Could Eskom Holdings Limited v Halstead-Cleak4 have been argued
REFLECTION differently?
In Eskom Holdings Limited v Halstead-Cleak the Court held that only a
‘consumer’ as defined in the CPA can claim under section 61. In this
instance, a cyclist who came into contact with a low-hanging live power
line over a footpath and suffered severe burns did not qualify as a
consumer vis-à-vis Eskom, as producer and distributor of the electricity.
This decision made it unnecessary to elaborate on the meaning of ‘a
product failure, defect or hazard’ or ‘unsafe goods’ in section 61, read
with section 53. Nevertheless, the Court expressed the following views in
this regard:

In any event it cannot be found that the harm the

respondent suffered was as a result of the electricity itself
failing, or that the electricity had a defect. Failing in this
context would be if the electricity were unable to perform in
the intended manner. This was not the case. The electricity,
in the context of the case did not suffer from a material
imperfection in the manufacture of the goods. Likewise, the
electricity did not have a characteristic that rendered it less
useful or safe than a person would generally expect in the
circumstances. The same applies to the electricity not
possessing a characteristic that presented a significant risk
of injury to any person when the goods are utilised. It is
clear that the respondent was not utilising the electricity
when he was harmed.5

Electricity is known and accepted to be inherently dangerous and as the

Court noted, this characteristic in itself does not constitute ‘a material
imperfection’ in its manufacture, in terms of the definition of ‘defect’ in
section 53(1)(a)(i). Nor does it render electricity ‘less useful, practicable
or safe than persons generally would be reasonably entitled to expect’, in
terms of the definition of ‘defect’ in section 53(1)(a)(ii). There is also no
evidence that the electricity in this case suffered from ‘inability of the
goods to perform in the intended manner’, in terms of the definition of
‘failure’ in section 53(1)(b).
However, is it not also arguable that there was ‘a product failure,
defect or hazard’ or ‘unsafe goods’ in respect of electricity in the Eskom
case? 6

PAUSE FOR What is ‘acceptable’?

REFLECTION The difficulty for the consumer is that when courts are required to apply
such vague standards as what is ‘acceptable’, there may be an
understandable tendency for courts to resort to look at the general
reasonableness of the producer’s behaviour. In this case, the standards
to be applied are barely distinguishable from those used under fault-
 based liability, and ‘strict’ liability could bring little real benefit to
consumers. The objective of this legislation is to impose liability on a
producer, importer, distributor or retailer without requiring the consumer
to prove fault. To make this possible, these provisions must provide clear
guidelines for courts. The definitions contained in the Act are not clear
and will be challenging to interpret.

39.6 Inadequate instructions or warnings provided to the

consumer pertaining to any hazard
Section 61(1)(c) refers to a situation where the consumer received inadequate instructions or
warnings ‘pertaining to any hazard arising from or associated with the use of any goods’. This
refers to a situation, for example, where the goods are inherently dangerous, but could have been
made safe by instructions for proper use or a warning making clear the risks of using the goods. An
obvious example is toxic cleaning substances. Although such products are clearly hazardous, they
are of undoubted household utility and present an acceptable level of risk if appropriately
packaged, bearing a sufficient warning of toxicity, and directions as to use. Where adequate
instructions and or warnings are provided the goods are not to be regarded as unsafe or defective
for the purposes of section 61(1), and the ultimate responsibility for accident prevention in effect
shifts to the consumer.
There is a similar category of ‘information’ defect in European and American law. However,
in South Africa section 61(1) employs the vague concept of ‘adequate’ in this regard. Moreover,
the focus is only on instructions or warnings ‘pertaining to any hazard’, bringing into this category
also the uncertainties of the concepts of ‘hazard’ and ‘significant risk’, as discussed above. In short,
section 61(1) in its present form, read with the definitions in sections 1 and 53(1), presents a
convoluted array of different standards.

39.7 Who is a consumer?

Section 1 of the Act defines a consumer in respect of any particular goods or services as follows:
(a) a person to whom those particular goods or services are marketed in the ordinary
course of the supplier’s business;
(b) a person who has entered into a transaction with a supplier in the ordinary course of
the supplier’s business, unless the transaction is exempt from the application of this Act
by section 5(2) or in terms of section 5(3);
(c) if the context so requires or permits, a user of those particular goods or a recipient or
beneficiary of those particular services, irrespective of whether that user, recipient or
beneficiary was a party to a transaction concerning the supply of those particular goods
or services;
(d) a franchisee in terms of a franchise agreement, to the extent applicable in terms of
section 5(6)(b) to (e).

It is clear from this definition that a consumer includes the person to whom the goods or services
are sold or supplied. The consumer is also the user of the goods or services, who was not a party to
the transaction for the supply of the goods or services.
Eskom Holdings Limited v Halstead-Cleak7
The plaintiff, a cyclist who came into contact with a low-hanging live power line over a
footpath and suffered severe burns, claimed damages under section 61 from Eskom as
 producer and distributor of the electricity. The plaintiff succeeded in the High Court, but lost
on appeal, on the basis that he did not qualify as a ‘consumer’ vis-à-vis Eskom, because he
(a) had not entered into any transaction with Eskom as a supplier or producer of electricity in
the ordinary course of Eskom’s business; and (b) was not at the time of his injury either a
user, recipient or beneficiary of the electricity.
For the meaning of ‘user’ the Court referred to a dictionary definition which defines ‘utilise’
as ‘make practical and effective use of’.8 The injured plaintiff could hardly be regarded as a
‘beneficiary’ of the electricity. The Court did not comment on the meaning of ‘recipient’, which
arguably could have a wide meaning including a person who accidentally comes into contact
with electricity, thereby inadvertently and unwillingly ‘receiving’ the electricity.
The Court found that sections 61(5)(a) and 61(5)(b), which define the ‘harm’ for which
damages can be claimed, indicated that harm in the form of death or injury can only be
suffered by a natural person, as opposed to a juristic person; and that these subsections do
not mean that ‘any natural person’ can be a claimant under section 61. It follows from this
conclusion that a claimant can be any person, natural or juristic, who fits the description of
‘consumer’ – in terms of paragraphs (a ) and (b) of the definition, either as a person to whom
the defective goods were marketed or who received the defective goods pursuant to a
transaction with the supplier; or in terms of paragraph (c), as a user, recipient or beneficiary
of the goods. The Court held that a person who is a user of the goods qualifies as a
‘consumer’ in terms of paragraph (c), where there is a ‘transaction to which a consumer is a
party, or the goods are used by another person consequent on that transaction’.9
The Court said that section 3 which outlines the legislative purposes of the CPA, together
with the definitions of ‘consumer’ and ‘transaction’, clearly indicate that the ‘whole tenor of
the Act is to protect consumers’.10 Section 61, the Court noted, falls within Chapter 2 of the
CPA, dealing with ‘Fundamental Consumer Rights’, and in particular in Part H, which deals
with the ‘right to fair value, good quality and safety’; and this indicates that ‘the harm
envisaged in section 61 must be caused to a natural person mentioned in section 61(5)(a) in
his or her capacity as a consumer. This is the only business-like interpretation possible.’11

However, section 61, read with section 53, does not unambiguously exclude bystanders injured by
goods as potential claimants, for example, a person injured when touching an open and live
electricity cable, or a person who happens to be nearby when a defective glass container explodes.
Therefore, it is arguable that bystanders were also meant to be potential claimants.12

39.8 Who is liable?

The Act creates liability for ‘the producer or importer, distributor or supplier’. 13 ‘Supplier’ and
‘supply’ have a particularly wide meaning, including ‘sell, rent, exchange and hire in the ordinary
course of business for consideration’, and in respect of services, ‘to sell the services, or to perform
or cause them to be performed or provided, or to grant access to any premises, event, activity or
facility in the ordinary course of business for consideration’. This covers leasing and seemingly
also the common triangular arrangement in which a finance company buys goods from a retailer to
supply to a customer on lease or hire purchase. Therefore, the Act appears to create strict liability
for finance companies engaged in selling or leasing goods.

39.9 For what type of harm will liability be imposed?

Section 61(5) imposes liability for the following types of harm:
(a) the death of, or injury to, any natural person;
(b) an illness of any natural person;
 (c) any loss of, or physical damage to any property, irrespective whether it is movable or
immovable; and
(d) any economic loss that results from harm contemplated in paragraph (a), (b) or (c).

Harm under these heads is assessed in terms of common-law principles. Harm includes loss or
damage of ‘any property’. This apparently includes damage to the defective product itself. Loss
may derive from the product itself in various ways, for example, the costs of replacement, repair, or
remedying a safety hazard. Therefore, the Act allows significant additional compensation, over and
above whatever contractual remedies may be available. The Act also expressly allows
compensation for economic loss, thereby opening up a potentially vast area of liability, for
example, for loss of profit.

39.10 Defences

39.10.1 Section 61(4)(a): Compliance with public regulation

Section 61(4)(a) provides a defence where the characteristic that rendered the product unsafe was
‘wholly attributable to compliance with any public regulation’. Legislation such as the Foodstuffs,
Cosmetics and Disinfectants Act 54 of 1972 or the Medicines and Related
Substances Control Act 101 of 1965 aims to promote safety (non-compliance with the provisions of
these statutes is a criminal offence), and compliance should not therefore render the producer or
supplier liable under section 61(4)(a). Compliance with codes of practice or voluntary standards
would also not of itself provide a defence under section 61(4)(a). In the European experience this
defence is rarely invoked since ‘as the purpose of most regulations is to make products safe, it will
rarely be the case that compliance will force the producer to make an unsafe product.’ 14

39.10.2 Section 61(4)(b)(i): Absence of defect at time of supply

A further defence is available if the product defect ‘did not exist in the goods at the time it was
supplied by [the defendant] to another person alleged to be liable’ (section 61(4)(b)(i)). The
concept ‘supply’ includes selling renting, exchange and hiring for consideration. At what point is
‘supply’ regarded as having taken place for the purposes of this defence? It is suggested that the
time of supply is the point when the defendant relinquishes possession in favour of another party.
The purpose of this provision would appear to be to allow the defendant to escape liability if the
defect has arisen after the goods left his control. If goods have become defective due to
mishandling or inappropriate modification, the producer and those who have supplied goods in
their original, safe, condition should not be held liable.
Component parts (for instance nuts, bolts and industrial fasteners) also qualify as ‘goods’ as
defined in section 1 of the Act. Section 61(4)(b)(i) allows the producer or supplier of a component
part to escape liability if it can show that that component was sound at the point when it was
delivered to the producer of the complex product.

39.10.3 Section 61(4)(b)(ii): Compliance with instructions

Section 61(4)(b)(ii) deals with the situation where a supplier (A) has passed on goods to another in
the retail chain (B), and in so doing A has provided B with instructions, for example, regarding
their use or safekeeping. Where harm occurs as a result of compliance with these instructions, this
subsection provides a defence for B, and also provides that A cannot use the defence in section 61
(4)(b)(i) above to exonerate itself.
39.10.4 Section 61(4)(c): Defects not reasonably discoverable by the
distributor or retailer
In terms of section 61(4)(c) there is no liability if, it is ‘unreasonable to expect the distributor or
retailer to have discovered’ the product defect, ‘having regard to that person’s role in marketing the
goods to consumers’. This exemption from liability is broadly drafted and has the potential to
readmit fault-based liability through the back door. The application of a reasonableness test to the
conduct of distributors and retailers is hardly different from the negligence test that is the hallmark
of Aquilian liability. There are many defects that one could not reasonably expect even a highly
responsible distributor or retailer to discover, for instance the so-called ‘development risks’, which
only becomes apparent in a new product once it has been used. Also, in many cases retailers sell
packaged goods as delivered by a manufacturer or distributor, and cannot reasonably be expected to
open the goods for inspection. In most cases one cannot expect a distributor or retailer to detect a
latent manufacturing or design defect in a packaged product, or a hairline crack in a bottle that later
explodes and injures a consumer, or the inadequate instructions or warnings accompanying a
product. Yet, to permit all distributors and retailers to evade liability on the basis of reasonableness
would significantly narrow the ambit of strict liability.

39.10.5 Section 61(4)(d): Time limit for claims

The introductory part of section 61(4) and its subsection 61(4)(d), read together, state that liability
in terms of section 61 does not arise if the claim for damages is brought more than three years after
the occurrence of certain events or the acquisition of certain knowledge. The Prescription Act 68 of
1969 generally deals with the effect of the passage of time on a ‘debt’.15 Prescription begins to run
as soon as the ‘debt’ is ‘due’.16 Instead of using the established terminology of the Prescription Act,
section 61(4)(d) refers to a ‘liability’ which ‘does not arise’ if the ‘claim’ is not ‘brought’ within
three years. Section 61(4)(d) is probably intended to co-exist with the operation of prescription
under the Prescription Act, but it creates a number of interpretation problems which could have
been avoided if the section had simply stated that liability under section 61 constitutes a ‘debt’ for
the purposes of the Prescription Act.
Under section 61(4)(d)(i) liability does not arise if the claim for damages is brought more than
three years after the ‘death or injury’ occurs. Under the Prescription Act prescription begins to run
in respect of a claim for damages arising from death or injury as soon as the ‘debt’ to pay damages
becomes ‘due’.
Section 61(4)(d)(iv) provides that liability in terms of section 61 does not arise if the claim for
damages is brought more than three years after the ‘latest date on which a person suffered any
economic loss contemplated in subsection 5(d)’. The provision in section 61(4)(d)(iv) appears to be
aimed at creating an exception to the ‘once-and-for-all rule’ where economic loss is involved. The
three-year period provided for by section 61(4)(d)(iv) begins to run not when the first loss occurs,
as under the Prescription Act, but when the last loss occurs.
Does knowledge of the existence of the facts affect the running of the time period in terms of
subsections 61(4)(d)(ii) and 61(4)(d)(iii) in the same way as it affects the running of a prescription
period under the Prescription Act? Under these subsections the beginning of the time period is
respectively ‘the earliest time at which a person had knowledge of the material facts about an
illness’ and ‘the earliest time at which a person with an interest in any property had knowledge of
the material facts about the loss or damage to that property’. It seems that in respect of the
knowledge requirement there is no inconsistency in principle between subsections 61(4)(d)(ii) and
61(4)(d)(iii) and the Prescription Act in the limited areas where there is an overlap, and, therefore,
when interpreting subsections 61(4)(d)(ii) and 61(4)(d)(iii) the courts can apply the provisions on
the knowledge requirement contained in section 12 of the Prescription Act, and the relevant case
law.
Is the running of the time period under section 61(4)(d) delayed in the same way as a
prescription period under the Prescription Act? The Prescription Act provides that where certain
circumstances or ‘impediments’ exist, such as minority of the creditor or marriage between the
creditor and debtor, the completion of prescription is delayed. It is highly unlikely that the
legislature intended to abolish the principles concerning delay of the completion of a prescription
period in respect of claims under section 61 of the Consumer Protection Act. One must conclude,
therefore, that the legislature simply failed to deal with this aspect, and that there is no
inconsistency between section 61(4)(d) and the Prescription
Act regarding delay, so that the courts can apply section 13 of the Prescription Act and the relevant
case law to claims under section 61 of the Consumer Protection Act.
Under the Prescription Act the running of prescription is interrupted by service on the debtor
of ‘any process whereby the creditor claims payment of the debt’. 17 The running of prescription is
also interrupted by an express or tacit acknowledgement of liability,18 which acknowledgment must
be clearly intended by the debtor.19 Although section 61(4)(d) is silent on the matter of interruption,
it is suggested that the courts should apply sections 14 and 15 of the Prescription Act and the
relevant case law to claims under section 61 of the Consumer Protection Act, with the result that
expiry of the time period in terms of section 61(4)(d) is prevented by either service of judicial
process or acknowledgement of liability.

39.11 Conclusion
The problems of proving defectiveness and negligence for the purposes of common-law (Aquilian)
liability were the main catalysts for reforming the law on product liability in South Africa. Section
61 of the Consumer Protection Act introduced strict liability for harm caused by defective products,
Because this statutory remedy eliminates the need to prove negligence on the part of a
manufacturer, importer, distributor and retailer of defective products (‘goods’) which cause harm, it
is likely that claims for damages involving defective products will in future in most cases be
brought under this section. However, the common-law liability for defective products continues to
exist as an alternative basis for recovery of damages for harm caused by defective products. In
unusual circumstances not covered by the provisions of section 61 of the Consumer Protection Act
claims will still be brought by way of an Aquilian action, as discussed in Chapter 19.

1 See definition of ‘consumer’ in section 1.

2 150 P 2d 436 (1944) at 440.
3 Stapleton ‘Restatement (third) of torts: products of liability, an Anglo-Australian perspective’ (2000) 39(3) Washburn
LJ 363 at 376–378.
4 (599/2015) [2016] ZASCA 150; 2017 (1) SA 333 (SCA) (30 September 2016).
5 Para 24.
6 It is suggested that when considering whether electricity presents a ‘defect’ or ‘hazard’ or is ‘unsafe’, the meaning of
‘electricity’ in the definition of ‘goods’ in the CPA should not be restrictively understood to refer only to electrically
charged particles (current), but rather in a wider sense, to include the manner in which the current is conducted and
made accessible. It is suggested further that electricity as a species of ‘goods’ in the commercial sense comprises both
the current and the means whereby the current is conducted and made accessible. Accordingly, if this occurs
irregularly and inappropriately, as in the case of a live and unprotected power line overhanging a footpath and
exposing persons to harm, the electricity is arguably ‘less useful, practicable or safe than persons generally would be
reasonably entitled to expect’, therefore presenting a ‘defect’ in terms of section 53(1)(a)(ii).
7 (599/2015) [2016] ZASCA 150; 2017 (1) SA 333 (SCA) (30 September 2016).
8 Para 24.
 9 Para 15.
10 Para 16.
11 Para 21.
12 Subsections 61(1)(c) and 61(2) refer to ‘the consumer’, but subsections 61(5)(a)–(b), which deal with the kinds of
harm for which damages can be claimed, refer to death or illness of, or injury to ‘any natural person’. Further, the
definition of ‘unsafe’ in section 53(1)(d) refers to ‘an extreme risk of personal injury or property damage to the
consumer or to other persons’. This definition does not deal only with consumers, but specifically refers to an extreme
risk presented by the goods ‘to the consumer or to other persons’. Section 5(5) provides that, even where goods are
supplied in terms of a transaction that is exempt from the CPA, those goods and the importer, producer, distributor
and retailer of those goods are nevertheless subject to section 60 (safety monitoring and recall) and to section 61
(liability for damage caused by goods). Section 5(1)(d) read with section 5(5), arguably highlights the intention of the
legislature to provide general redress for persons harmed by defective goods, even if they did not receive the goods
pursuant to a ‘transaction’ or as a ‘consumer’ within the meaning of paragraph (b) of the definition of ‘consumer’.
13 All these terms are defined in section 1.
14 Grubb and Howells (Eds) The Law of Product Liability (2000) para 4.211.
15 See, generally, Loubser Extinctive Prescription (1996) at 1–8.
16 Section 12(1) of the Prescription Act 68 of 1969.
17 Section 15(1).
18 Section 14(1).
19 Road Accident Fund v Mothupi 2000 (4) SA 38 (SCA) paras 36–37.